Fortune (Page 5)

President Donald Trump said Thursday he is renaming Lake Ontario to be known as “Lake America” in the United States as he escalates his trade war with Canada.

The Republican president signed an executive order directing the Interior Department to update the lake’s name in the U.S. geographic naming service. Trump cannot force Canada to follow along, however.

Trump has been floating the idea of the name change in recent days as the U.S. announced it was imposing 50% tariffs on $20 billion worth of Canadian goods over the weekend after talks between the countries broke down. Trump has been needling America’s northern neighbor since he returned to the White House last year, suggesting the ally with whom the U.S. once had warm relations should instead be absorbed as the 51st state.

Canada responded to Trump’s import taxes this week by imposing retaliatory tariffs on $20 billion worth of American goods, including steel, dairy products, appliances and farm equipment.

Trump, who signed the order as he was sitting at the Resolute Desk in the Oval Office, had a large sign behind him propped on a stand with a map of the Great Lakes. Over Lake Ontario, in big red letters, the map read “Lake America.”

On the other side of the president was another map with the words, “MAKING THE GREAT LAKES EVEN GREATER.”

Trump said as he was signing the order that he’d “notified all of the various people that we have to notify. So, we’ve done everything that you have to do.”

“And this is official, effective immediately,” he added.

But the executive order Trump signed shows that the changes are supposed to be made within 30 days.

Canadian officials dismissed Trump’s move.

Canadian Prime Minister Mark Carney rejected the move, invoking the lake’s Indigenous roots and noting that the name predates both Canadian Confederation and the U.S. Declaration of Independence.

“Canadians also know that naming reality means calling it Lake Ontario — then, now and always,” he wrote.

Ontario Premier Doug Ford, who has traded insults with Trump, said on X: “It’s Lake Ontario to Canadians and the rest of the world. Now and forever.”

Nova Scotia Premier Tim Houston said, “We’re into some real foolishness now. It’s Lake Ontario, buddy.”

The lake is one of multiple Great Lakes that the U.S. and Canada share borders along. New York Gov. Kathy Hochul, a Democrat whose state shares a border on the lake, wrote on X in response to the White House’s announcement that, “New York won’t be calling it that.”

There is no single international body that determines names of international bodies of water and Trump has wide latitude over how the U.S. government recognizes geographic places and landmarks.

The name Lake Ontario comes from the Huron Indigenous people’s word “oniatarí:io,” which means “lake of shining waters.” The province of Ontario, founded in 1867, took its name from the lake.

Trump said that while his action was not meant to send any particular geopolitical message, “Canada’s been ripping us off for a long time” on trade and military issues.

“They wanted to be treated like a state and they’re not a state,” the president said. “We just can’t do that anymore.”

“We love the people of Canada,” Trump added. “I don’t think their representatives do, an appropriate job. Maybe they’ll change. I really don’t know. It doesn’t make much difference.”

The move is reminiscent of his move last year to rename the Gulf of Mexico as the “Gulf of America.”

Trump scribbled his name with a Sharpie pen on the executive order, then held it up for the cameras, offering, “And we filed all the necessary papers, documents, everything else.”

He also suggested his push to rename bodies of water may not be finished.

“So, if you think about it, we have a gulf and we have a lake. Now, all we need is an ocean,” Trump said. “So maybe we’ll have to change the name of the Atlantic and/or the Pacific. Maybe we’ll change them.”

___

Associated Press writer Rob Gillies in Toronto contributed to this report.

This story was originally featured on Fortune.com

Three staff members from the Stars and Stripes military news outlet are suing to challenge their recent firings by the Pentagon, accusing the Trump administration of violating their First Amendment free speech rights to speak out against government interference in their work.

Stars and Stripes publisher Max Lederer, editor-in-chief Erik Slavin and Middle East reporter Lara Korte are named as plaintiffs in the federal lawsuit filed Thursday in Washington, D.C.

The Pentagon, which partially funds the news outlet, fired them last week. A Pentagon spokesperson declined to comment on the suit’s allegations.

Slavin says he was dismissed for insubordination after giving an interview to CBS News in which he objected to potential censorship by the U.S. military. Korte participated in the same interview. Their lawsuit says they were punished in retaliation for publishing a report on deteriorating conditions aboard the U.S.S. Abraham Lincoln.

“Stripes’ historical editorial independence is critical to its core mission of gathering and providing unbiased, credible journalism to the U.S. military community, particularly servicemembers and their families stationed overseas,” the suit says.

Lederer was dismissed shortly after announcing his impending retirement, effective at the end of September. He said the Pentagon had installed a new deputy publisher, an active duty service member, under him at the newspaper without his prior knowledge.

The suit’s plaintiffs are represented by attorneys from Democracy Defenders Fund, Lawyers for Good Government, Government Accountability Project and a Yale Law School clinic.

The Defense Department and Defense Secretary Pete Hegseth are among the named defendants.

This story was originally featured on Fortune.com

The American billionaire who bought Ferrari’s first electric vehicle model for an eye-watering eight figures may be able to cash in on a tax write-off perk from the purchase. 

Earlier this month, 87-year-old optometrist-turned-businessman-turned-philanthropist Dr. Herbert A. Wertheim bought a Ferrari Luce at a Sotheby’s public auction for $40 million. The auction was part of a charity effort to benefit the Ferrari Foundation, which funds global education initiatives.

The $40 million buying prices on the auto was more than 36 times the model’s presale estimates of $1.1 million. (The Luce retails for $640,000, but the auctioned car had a higher price tag because it had a pre-production chassis.)

Now, it seems Wertheim may end up getting a large chunk of that purchase back, as part of a longtime American tradition of the country incentivizing the nation’s wealthiest philanthropists into giving back.

Large charity donations like Wertheim’s can lend themselves to massive tax write-offs as a result of America’s tax system. Automotive content creator and television presenter Peter Greaves first brought attention to the potential for Wertheim to gain millions back on the purchase of his Luce in a recent Youtube video.

When Wertheim files his 2026 taxes, he would have to subtract the estimated sale value of $1.1 million from the $40 million the car sold for. As of the IRS’s 2026 tax rules, he would then have to take away a 0.5% floor of his adjusted gross income, which Greaves estimated to be $200 million, leaving $37.9 million. The U.S.’s 2/37ths rule created under the One Big Beautiful Bill Act, which caps tax savings of itemized deductions at 35% compared to the previous 37% for top earners, would further reduce the sum to $35.85 million. According to tax law, 37% of that total could be claimed, meaning Wertheim could receive more than $13 million back from the U.S. government for his purchase.

Wertheim, who did not immediately respond to Fortune’s request for comment, has made no public comments about possibly taking advantage of the tax write-off for the Luce, or for any previous charity auction purchases.

The philanthropist has an estimated net worth of $4.8 billion and has previously participated in a previous charity auction for luxury vehicles, reportedly paying $26 million for the Ferrari Daytona SP3, or “599+1.” He has donated more than $200 million to various causes, including  $50 million to UC Berkeley Optometry and $100 million to Baptist Health Foundation. In February, Wertheim paid $2 million at a Mar-a-Lago charity event for a private visit with President Donald Trump at the White House. It’s not his only recent brush with politics: Wertheim briefly launched a Congressional bid in Florida’s 22nd District earlier this year.

For its part, the Maranello, Italy-based carmaker has weathered controversy around the rollout of its EV. Former Ferrari president and chairman Luca di Montezemolo joined analysts and investors in mocking the model as ugly and decidedly un-Ferrari-like at a time where other luxury automakers were scaling back their own EV efforts amid low demand. Ferrari may be getting the last laugh however: the Financial Times reported last month that Ferrari exceeded its short-term sales goal of 500 units. 

How the Trump administration transformed charity tax breaks 

Tax breaks for philanthropy is an American tradition dating back to 1917 after the passage of the War Revenue Act, in which Congress created a federal income tax deduction for charitable gifts as part of an effort to keep private philanthropy alive and well during World War I, which would relieve the U.S. from funding essential social welfare programs. Those benefits have slowly expanded over the last century.

But the Trump administration has made it more for the wealthy to get money back for their donations come tax season, with the One Big Beautiful Bill Act effectively slashing the benefit from 37% to 35%, with itemized taxpayers having to deduct donations only in excess of 0.5% of their adjusted gross income.

The policy changes with lower tax incentives may alter the future of philanthropy itself. The new 35% limit could reduce donations by between $4.1 billion and $6.1 billion, according to the Indiana University Lilly Family School of Philanthropy. Experts warn that fewer big donors—or big donors giving less—would place a larger burden on middle-class givers to bridge a gap that isn’t realistic as financial pressures for less-wealthy households increase.

“The nonprofit sector says that every dollar matters, and so incentivizing small donations from every household could have a meaningful impact for certain kinds of organizations,” Elena Patel, co-director of the Urban-Brookings Tax Policy Center, told CNBC last November. “But the truth is that those kinds of contributions, however, just are not the bulk of charitable giving in the charitable sector.

This story was originally featured on Fortune.com

It’s easy to become jaded in a job that’s unfulfilling, especially when dreams of a more meaningful career clash with the need for financial stability. 

But successful entrepreneurs contend that chasing their true ambition inspires them to show up and work every day; those who really love their job, like musician Pharrell Williams, never plan on retiring. 

“I’m so blessed to call this a job. I’ve never stopped loving this job; I’ve never stopped being a student,” Williams said onstage after winning the Dr. Dre Global Impact Award at the 2026 Grammys. “So everyone here, I’ve just got to tell you, never stop grinding. And listen, never stop working.”

Williams has witnessed firsthand how passion plays a huge role in success; the now 53-year-old philanthropist and artist has been releasing, writing, and producing music since the 1990s, amassing more than three decades of chart-topping hits. With an estimated net worth of $250 million, the “Happy” singer’s years of musical prowess and business collaborations have resulted in a lifelong fortune. But Williams said he doesn’t think too much about the number in his bank account—what really keeps the Gen Xer working is his love for the job.

“Stop doing anything else but working. Work, man,” Williams continued. “I’m 52, I get to do this every day—I love what I do. And if you do what you love every day, you’ll get paid for free.”

Williams says the American Dream isn’t about making millions

Generations of workers have bucked their true calling for a tolerable job with a solid paycheck. With nearly half of U.S. citizens believing the American Dream is no longer possible, or never even existed, many are stuck in a lifeless, steady job rather than a profession that makes them happy. 

Growing up in Virginia Beach, Williams said he often heard the same rhetoric: that careers should be centered around stability and making money. But the Grammy winner wants to change that narrative. Chasing the biggest paychecks, he said, is a misnomer. 

“The American Dream is not about making the most money. In fact, the human dream and the consumers’ dream shouldn’t be about making the most money,” Williams said at the 2024 Web Summit. “It should be about spending the most time doing something that you love.”

The philanthropist pointed out that many young people are pressured to pursue high-paying careers in college, such as medicine and law, whether they want to or not. Some are able to switch tracks despite the chagrin of their family, Williams explained, but the vast majority “go after it, and they don’t get it.” After that, they might be stuck in a job they hate, prioritizing a paycheck over their true passions. Williams advised that professionals can avoid this identity crisis by working in a role or industry that interests them. 

“If you can find a vocation around something that you love, you now have a dream job. You will be the first one there, and you’ll be the last one to leave,” Williams continued. “To me, that is what we should be telling our children—that is the way that we should be leading society—for people to do what they love.”

‘Pursue a career that does not feel like work’

Williams is one of many leaders who have proved that sticking with a career you love can pay off. Walmart’s former CEO, Doug McMillon, who stepped down from the top role earlier this year, spent four decades climbing the ranks of the now–$819 billion retail giant. And during his ascent from being an hourly worker all the way up to chief executive, McMillon said he’s never “been bored one single day.”

“Pursue a career that does not feel like work. Life is too short to invest so much time doing something you don’t enjoy,” McMillon told graduates during a commencement address at the University of Arkansas in 2024. “I hope you find your spot quickly like I did, but if you don’t, my advice is that you shouldn’t give up until you do … If you’re in the right place, most days, work won’t even feel like work.”

Christian Toetzke, cofounder and CEO of global fitness race Hyrox, also said that he loves his job so much that it doesn’t feel like a chore. His best advice for professionals hoping to strike the right work-life balance is to actually enjoy their careers; that way, it’s never a drag to show up at the office. 

“I’m a massive believer in work-life balance, but the question is always how we look at this. And I’m a very privileged person because I don’t consider what I do as work,” Toetzke said on the Opening Bid Unfiltered podcast last year. “I do what I really love. It’s also my hobby. For me, work is not a punishment. It’s almost kind of a reward.”

A version of this story was published on Fortune.com on February 3, 2025.

This story was originally featured on Fortune.com

In 1949, a doctor named Virginia Apgar was having breakfast in a hospital cafeteria when a medical student asked her a question nobody had previously thought to answer: how do you know, in the first moments after a baby is born, whether it is going to be okay?

Virginia grabbed a napkin and wrote down five things: Heart rate. Breathing. Muscle tone. Reflexes. Color.

Doctors had known all of these markers for decades. The knowledge was not new. What was new was someone finally organizing it into something a nurse could use in sixty seconds in the chaos of a delivery room.

That became the Apgar Score. It is used in delivery rooms around the world today. Somewhere right now, a nurse is scoring a newborn out of ten because Virginia saw over breakfast something nobody else had bothered to address.

The gap was not a secret. It was just sitting there, waiting for someone to act.

I learned a word for that gap from my volleyball coach, who grew up in Hawaii.

Puka.

It means hole, gap, opening — the same hole at the center of those puka shell necklaces everyone wore in the nineties. My coach would yell “puka!” when she spotted a space on the court nobody was covering. I heard it so often it stopped being a word and became a reflex: See the gap. Move.

I have been moving toward gaps my entire career. And I have come to believe that this instinct – to walk into any room and see not just what is happening but what is missing, and act before being asked – is one of the most powerful and underrated career strategies available to professionals and executives alike.

Early in my career, I found myself on the edges of a major trial. I was not on the team. I was the newest lawyer at the firm with time on my hands and a pile of disorganized medical records that nobody wanted to manage. So, I managed them. I built color-coded binders for every significant medical issue, decoded thousands of pages of chart abbreviations, loaded them on a dolly, wheeled them into the war room, and started walking the partner through the organized tabs.

That was day one. The binders were the first puka. Then I found the next one. And the next. Six weeks later, I was a vital member of a trial team I was never originally slated to join. When the lead partner left for a larger firm, he took a small, handpicked group with him. I was on that list.

He moved again to an even larger firm and brought me along a second time: a chain of events that started with a dolly and a stack of binders nobody wanted to deal with and ended at one of the largest global law firms in the world.

That trajectory started because of a willingness to step into a space that others stepped around.

Some gaps are obvious: the unowned problem, the operational bottleneck, or unglamorous work outside everyone’s formal job description. Everyone sees it, but few step forward.

Other gaps are organizational or structural, and those are often the most valuable. For leaders, cultivating an environment where people actively seek out and fill these gaps is what separates adaptive companies from stagnant ones. When employees move toward unassigned problems, careers grow organically: they do not wait for a bigger title, but instead begin handling the responsibilities of that next level until the organization adapts around them.

The person who fills gaps consistently becomes indispensable to the people around them. Repeatedly demonstrating initiative builds a reputation that opens doors before you ever knock on them.

The puka mindset relies on two core behaviors:

• Relentless curiosity: The habit of walking into any room or analyzing a business model and asking not what you have been assigned but what this situation actually requires. Widening that lens reveals what is falling through the organizational cracks.

• Decisive initiative: Filling the gap simply because it needs filling, not for immediate credit, but because operational momentum demands it. Awareness without action is merely observation.

Right now, one of the most critical gaps across corporations is the practical integration of AI. While executive boards debate high-level strategy and conference panels discuss theoretical disruptions, the daily tools sit largely underutilized on the ground.

The gap is not access; it is application.

I know that gap well, because I once filled it with binders and a dolly.

The medical records were never a secret. Every page had been produced and was sitting in a box for months. What was missing was not information but shape, a way for a partner to find the one chart note about a missing lab value in seconds, standing in front of a jury. Back then, that kind of organization took weeks.

A version of that work now takes an afternoon.

The conversation about AI is still largely framed around what it replaces. A better question is what it makes possible. For the first time, much of the work that once kept organizational gaps open can be done faster, cheaper, and at scale. The tracker nobody built. The eight years of vendor contracts nobody has read end to end. These gaps persisted not because they were invisible but because nobody had the hours to close them.

Imagine a junior team member on a long project watching the same questions get asked over and over, because the information is scattered across emails, meeting notes, and old drafts. She uses an AI tool to pull it all together into a single tracker the team can actually use, then checks it herself against the source material. Nobody assigned her the work. She just noticed a gap nobody owned and closed it before lunch.

The tool does not replace judgment. It does not decide strategy. It organizes what is already known so the humans can spend more time thinking and less time hunting.

The cost of filling a puka has collapsed. The value of spotting one has not. The people who move toward that space now position themselves at the front of a wave that will generate new roles and opportunities for years. For senior leaders, the task is twofold: build the habit of spotting these openings yourself, and create a culture where taking the initiative to fill them is recognized and rewarded rather than treated as a compliance risk. That can mean funding small pilot programs, giving teams explicit room to experiment within reasonable guardrails, or simply promoting the people already finding smarter ways to work before anyone told them to.

AI is just today’s clearest example. Every industry has its own version, and pukas are never in short supply.

Gaps in strategy and execution never run out. The fundamental question for any team or leader is simple: when an unowned problem sits in plain sight, who moves first?

The opinions expressed in Fortune.com commentary pieces are solely the views of their authors and do not necessarily reflect the opinions and beliefs of Fortune.

This story was originally featured on Fortune.com

Good morning. At Exelon, one of the nation’s largest utility companies, finance is getting a bigger seat at the strategy table.

Jeanne Jones, the finance chief since 2022 and an almost 20-year veteran of the company, will take on the newly created role of EVP of finance and strategy on Oct. 5, putting her at the center of corporate strategy as electricity demand surges and utilities face mounting pressure to invest without driving up customer bills.

Jones will continue to report to Exelon President and CEO Calvin Butler. Robert Kleczynski will succeed Jones as CFO and become her direct report. Kleczynski currently serves as SVP, controller and head of tax.

“Jeanne’s leadership has been instrumental in helping Exelon deliver strong financial performance while maintaining our focus on delivering safe, reliable and affordable service for our customers,” Butler said in a statement. In her new role, Jones will continue to work “across the business and the energy sector to identify opportunities to deliver value for our customers and shareholders,” he said.

A shift in the CFO’s mandate

The move reflects a broader shift in the CFO’s role. The corporate strategy playbook is increasingly landing on the CFO’s desk. While finance chiefs have long led capital allocation and investor communication, they’re increasingly expected to help determine where the company places its biggest long-term bets alongside the CEO, Andy West, a senior partner at McKinsey and global co-leader of the firm’s Strategy and Corporate Finance practice, recently told me.

That evolution is particularly relevant in the utility industry, where electricity demand is growing, grids need modernization and companies face mounting pressure to invest while keeping customer bills affordable.

At Exelon, Jones will continue to oversee the company’s financial activities while helping lead integrated strategy efforts focused on trends affecting Exelon, its operating companies and the broader energy industry. The structure is designed to strengthen alignment between long-term strategy, capital allocation and operational execution, according to a company spokesperson.

A pure-play utility investing in the future

Chicago-based Exelon (No. 189 on the Fortune 500) serves 11 million customers through six regulated transmission and distribution utilities. Since spinning off its power generation business, Constellation Energy, in 2022, Exelon has focused solely on regulated utility operations. It doesn’t own power plants; instead, it manages the infrastructure that delivers electricity and gas to end users.

That makes capital allocation especially consequential. Exelon has to determine where and how aggressively to invest in its networks as demand rises, while balancing the costs ultimately borne by customers against the need to build infrastructure for the future.

Jones has been at Exelon for almost two decades, with experience across utility operations, corporate finance and the company’s former generation business. She served as CFO of ComEd, one of the nation’s largest electric utilities, then as Exelon’s CFO, guiding the company through a period of consistent operational and financial performance following the separation.

Her experience gives a perspective that crosses different parts of the energy value chain. Now she’ll be asked to apply that experience to questions that increasingly blur the line between finance and strategy: how much to invest, where to invest it and how to do so while keeping costs under control.

Exelon reported second-quarter earnings on July 30 and revenues totaled $5.97 billion, beating estimates. The top line increased 10% from the year-ago figure of $5.43 billion. Adjusted operating earnings increased 10.3% to 43 cents per share from 39 cents in the year-ago quarter. 

Reflecting on her journey, Jones told me last year that her best career advice is to stay open to new experiences and not get overwhelmed by distant goals. Don’t let the pressure of a specific end goal cloud your enjoyment of the ride.

“Keep going for the next thing that’s going to develop you,” she said.

Have a good weekend.

Sheryl Estrada
Sheryl.Estrada@fortune.com

This story was originally featured on Fortune.com

I was looking for a word to describe an unimagined market and the best fit I could find is terra incognita, or “unknown land.” Roman cartographers used the term to label mystery territories on maps for explorers. 

This all came up because Deloitte just published its attempt at a value for the moon economy—and all the cascading unknowns that could flow from going there like finding rocket fuel, extracting minerals, and forming data center manufacturing hubs. The value of a lunar economy is $566 billion in an accelerated-growth scenario by 2050, and $343 billion under a more conservative scenario. But there’s a separate $541 billion that Deloitte pinned to the space-based potential stemming from scientific advances, new energy sources, and innovative commercial applications. That’s the terra incognita—what could come.

I spoke with the coauthors of the report, “Building the Lunar Economy,” Brett Loubert, who leads Deloitte’s space practice, and Raquel Buscaino, head of its novel and exponential technologies team. I asked Buscaino about risk, and she told me the “uncertainty and ambiguity” around the unimagined market isn’t necessarily a warning, it’s a sign of “extraordinary possibility.”

“Uncertainty isn’t a flaw in the story,” said Buscaino. “Some of these distant opportunities could have extremely large upsides.”

No one knows yet which ones will pan out or what will ultimately matter the most in a market we haven’t imagined yet, she said, but it’s a critical piece of what’s playing out right now in space. “The uncertainty is part of the reason so many investors, countries, agencies, and folks within your community might be interested in this,”  Buscaino said. 

Deloitte’s model drew on more than 25 interviews and 400 model inputs to map two value pools. The first is core lunar infrastructure, which is what you need in space. Transportation, energy, communications, life support, and construction come in at $282 billion through 2050. Not a shock that transportation soaks up $206 billion of that figure.

The second pool is what all that infrastructure can enable. There’s the obvious national security posture, which Deloitte calls “the ultimate high ground.” Deloitte itself is moving onward in this sector with its Silent Shield cyber intrusion detection system, so the advisory firm already has a stake in the success of space. There is also helium-3 extraction, rocket propellant, orbital compute, and in-space manufacturing. Deloitte valued this pool at up to $284 billion. 

Capital of all sorts is already on the move. Seraphim’s space tracker reported $7.5 billion in space-tech investment in Q2 of this year, with trailing 12-month investment at an all-time high of $23 billion. A separate analysis of publicly disclosed equity rounds by space companies over the past year found 47 deals with a median round size of $40 million and an average of $116.2 million. SpaceX alumni have left the mothership and forged 141 startups and companies of their own, worth $10.6 billion, according to Forbes. 

By the way, none of this can be disentangled from SpaceX, piloted by Elon Musk and now a public company carrying a market cap of $1.8 trillion. Only Musk could make space as mainstream as it is right now in as short of a time period as we’ve seen. 

And potentially an even more interesting dynamic, Loubert said, is the funding model for space, where governments are increasingly acting as anchor investors and customers. They buy rides to the moon and engage with companies on commercializing their capabilities. The government has said, ‘We’re going to basically fund the initial development and the initial missions that are going to go there to help de-risk and provide not just government funding, but also confidence to your investors that there is a near-term government and consistent demand,’” said Loubert.

Essentially, said Buscaino, governments are helping to create demand for the foundational infrastructure. The hope is that as the infrastructure matures and costs come down, commercial activity increasingly scales alongside it. So, if the unknown sounds appealing and not at all like a warning, I invite you to the terra incognita of what’s to come. 

See you next time,

Amanda Gerut
amanda.gerut@fortune.com

Submit a deal for the Term Sheet newsletter here.

Joey Abrams curated the deals section of today’s newsletter. Subscribe here.

This story was originally featured on Fortune.com

Nvidia CEO Jensen Huang is open to ideas about how to share the massive wealth being generated by the AI boom. He’s created billionaires on his executive team, personally ensures his staff are competitively paid, and is “perfectly fine” with proposals to tax the ultra-wealthy more.

But he disagrees with a proposal from Microsoft co-founder Bill Gates to tax robots and AI tokens.

Gates resurfaced his suggestion in an essay this week, writing: “Right now, if you’re an employer and you hire someone, you pay payroll taxes on their earnings. But if you buy a robot, you can usually write it off right away as a business expense. The tax system nudges you toward replacing people with machines.”

The entrepreneur-turned-philanthropist also suggested that governments will need the funds. Gates figures that if AI takes the jobs of humans, then state revenues from income tax will drop.

“A tax would slow the rush away from human labor a little and raise money for retraining and a stronger safety net,” Gates explained.

Huang disagrees. In an interview with Fox Business’s The Claman Countdown, the chipmaker boss said: “I love the heck out of Bill … but I don’t see what he sees. I see something very, very different. And so my remedies will be a little different.”

Huang added: “I’m in favor of taxes. And I think that … for anybody who is productive, it’s a great way for us to contribute back to society and the economy. But the fact of the matter is, there are probably lots of different ways to approach this.”

Gates—a self-professed AI optimist—had a markedly more cautious tone in his latest op-ed. He wrote that while AI promises huge boons (such as improving access to medicine and education and streamlining bureaucracy), it also poses huge threats that world leaders aren’t ready for. These risks include ‘stunted’ child development, emboldened criminals, and vanishing jobs for Gen Z.

Huang’s perspective is more positive. He said that while “of course” the nature of jobs would change—as his already has—”I believe … that this will be a net job creator. However, there are going to be many jobs that will be disrupted and so we have to be sensible about that. We have to be sensitive about that and be supportive of that.”

While Gates suggests that companies could be incentivized to replace humans with robots because of existing fiscal policy, Huang says the historical precedent disagrees: “When companies are more productive, they don’t lay off people, they hire more people. The reason for that is because companies have ambitions … and I would say the vast majority of the world’s companies … have ambitions for growth.

“When they’re more productive, when they’re more profitable, [it] allows us to invest more and go after more growth. Nonetheless, overall, this is going to be a net job creator at a scale that we have never seen.”

A bid for reindustrialization

Huang has previously suggested that skilled, blue-collar workers stand to gain significantly from the AI boom. Trades like plumbers and electricians, Huang has said, will be in high demand as data centers are built across the globe.

Speaking this week, the 63-year-old tech titan suggested that the new economic era would be one of reindustrialization.

He explained: “We have lots and lots of white-collar workers, but we’re also going to have a lot of skilled labor. And having a large population of skilled labor and people who build things and make things with their hands is tremendous for the United States. We want to reindustrialize the United States. We want to create more jobs. And all of that’s going to happen right now as we speak with A.I.”

This story was originally featured on Fortune.com

In March, I wrote in Fortune that the shadow fleet exists because the rules of the sea are fundamentally voluntary. Ships can opt out. They can switch off transponders, fly flags bought from registries that never inspected anything, and hold insurance nobody can trace. I noted that when one of them has an accident, there may be no one standing behind the policy.

Which is happening right now off the coast of Oman.

The dark fleet vessel in question is the Caroline Bezengi: a Suezmax built in 2001 carrying roughly 800,000 barrels of Russian Urals crude. The ship loaded at Novorossiysk on May 11 and was destined for Sikka, in Gujarat. On June 8 an explosion flooded her engine room off southern Yemen and left her adrift. No group has yet claimed responsibility and no government has stated a conclusion about the cause. Security sources and Greenpeace have both pointed to a limpet mine, but that is not yet confirmed. Her crew abandoned ship three days later, and on June 30 she hit the rocks of Al Qibliyah, in the Hallaniyat archipelago, and grounded. Adding insult to injury, she now sits — leaking oil — inside a 667-square-kilometer marine reserve Sultan Haitham had created by royal decree nearly eight months before, to protect an endangered population of Arabian Sea humpbacks.

Oil reached the mainland at Ras Madrakah. Oman’s Environment Authority assessed the affected area at 390 square kilometers on August 10; satellite analysis by SkyTruth’s John Amos suggested a figure of more than 2,000 square kilometers two days later. Dimitris Maniatis, who runs the maritime risk firm Marisks, told Bloomberg Television that cleanup alone could run $200 million to $500 million.

Nobody is going to pay it.

The Caroline Bezengi’s last registered owner of record is Rentoor Shipmanagement Ltd, a Marshall Islands company dissolved on February 24. Its Shanghai correspondence address is, Maritime Executive reports, a dead-letter box in a residential block. The ship claimed Cameroonian registry, but Cameroon suspended its international register on February 6 after discovering its flag was being issued through two fraudulent websites. In June Cameroon told the IMO it had struck 39 ships off its registry. No International Group club (whose members insure roughly 87% of oceangoing tonnage) is likely to insure a ship without any approved classification. The ship had no publicly identified insurer. That isn’t supposed to be possible.

So, no insurer is going to pay. Fortunately, there is a longstanding international safety net. It isn’t going to help. Oman is a party to the 1992 Civil Liability Convention and the 1992 Fund Convention. It sits comfortably inside the compensation architecture the world built after Exxon Valdez and rebuilt after Prestige: strict liability on the registered owner, compulsory insurance certificates, and to back it all up the International Oil Pollution Compensation Fund, which can pay up to about $279 million per incident. The Fund is financed by levies on any company receiving more than 150,000 tonnes of oil a year in a member state.

That worked for three decades. The Fund paid €147.9 million after the Prestige. It paid South Korea after the Hebei Spirit. In 2000, when a small tanker carrying no liability insurance and no class certificate sank off Abu Dhabi, the Funds paid claimants in the United Arab Emirates anyway.

It’s not going to pay Oman. A spokesperson for the Fund told Reuters in August that it would not be involved in cleanup costs because the incident “was being treated as an act of war,” a category the convention excludes. No reasoned determination has been published. The Fund’s governing bodies have not yet met on the incident. Under the convention the burden of proving that exclusion falls on the Fund, not the claimant, and Oman has three years to force the question in court.

Meanwhile, Oman is running the shoreline response on its own dime: booms, dispersants, and crews hand-clearing some twelve kilometers of oil-sodden beach at the height of the Khareef monsoon.

So, what will governments, especially those allied to the United States, actually conclude? The shadow fleet just showcased its capacity to inflict real and unmitigated consequences on a bystander nation. But enforcement isn’t politically cheap. The most likely answer is that pollution will quietly replace sanctions as the legal basis for stopping shadow fleet vessels.

Europe spent two years interdicting on sanctions grounds and found the tool blunt. Estonia and Finland challenged more than two thousand tankers for insurance documents by radio between 2024 and mid-2025; better than nine in ten produce a certificate, roughly a third of them from sanctioned Russian or Russian-linked insurers. Increasingly they produce certificates from Russian insurers nobody has sanctioned at all. No government has publicly disclosed how many were judged genuine. And the escalation risk is real. After Russia scrambled a Su-35 to shield a tanker Estonia tried to stop, Estonia’s navy commander told Reuters in April that the risk was “just too high” to detain shadow vessels. Two exceptions he did name were an imminent oil spill, and damage to undersea infrastructure.

This is becoming the doctrine. Ireland passed a bill to let its Naval Service board ships and order them to cease activities threatening “the marine environment.” RUSI recommends the environmental framing explicitly, because flag states respond better to marine-ecosystem arguments than to sanctions enforcement. To comply with sanctions is to pick a side. Enforcing environmental regulations is entirely defensible self-interest.

Environmental regulations at sea are about to get stronger, or at least more stringently enforced. Oman will be the example that gets pointed at. A nature reserve covered in floating crude oil, a bill somewhere north of $200 million, and a legal architecture that said no to helping. Oman will serve as a call to action for new efforts that have very little to do with saving the earth. The point is to fight the shadow fleet without publicly picking a side.

The opinions expressed in Fortune.com commentary pieces are solely the views of their authors and do not necessarily reflect the opinions and beliefs of Fortune.

This story was originally featured on Fortune.com

I spent over two decades as a music journalist chasing bylines in Rolling Stone, GQ and many other outlets, and I told every editor who’d listen, “If you ever assign me a Dolly Parton interview, I’ll happily hang up my rock critic shoes.”

Her songs have been part of the soundtrack of our lives, and we’ve been hearing them on repeat forever. Sadly, that assignment never came, and Dolly has always been the one who got away. So, I write this not as someone who knew or met her, but instead as a fan who adored her, first as a critic with a tape recorder I never got to use, and now as someone who spends my days thinking about what her reputation meant to her legacy. 

I’ve watched the news of her passing unfold the way most of you probably have: scrolling, rereading tribute after tribute and appreciating the people she inspired – both personally, like I’d once dreamt of, and from afar. There’s a particular kind of grief in losing someone who felt like family to millions of people who never met her. And as it turns out, you didn’t have to own a single Dolly record to feel that loss, because Dolly was far more than a country music icon. She was an American legend: unapologetically and authentically herself. 

Reflecting on her life from both of my vantage points, as music critic and reputation building communications counselor, I’ve noticed that the industry often sorts stars into “artist” or “brand,” as if those are competing categories. But Dolly broke that mold, spending 70 years steadily proving that it’s not one or the other.

Music press often gets a bad rap for treating “brand” as a slightly dirty word when it’s applied to an artist, as if commercial savvy somehow dilutes creative legitimacy. But Dolly never bought into that premise, and I’ve come to believe she was right. She understood something that the communications industry still sometimes struggles with — earlier, and more completely, than almost anyone in entertainment: authenticity isn’t the absence of strategy. Sometimes authenticity is the strategy.

Consider what she built: ownership of her songwriting catalog at a time when almost no one could ask for that (she famously turned down an Elvis Presley cover because his manager demanded half the publishing). A theme park bearing her name that outlasted every skeptic’s prediction. A visual image featuring her hair, rhinestones, and voice, made so consistent for six decades that a silhouette alone could identity her. It wasn’t by chance. Talented people — and we don’t need to debate how much talent Dolly had — understand instinctively or deliberately that being true to yourself is the realest way to build reputation on purpose, over time and in ways that are easy to defend.

Here’s the paradox at the center of it: Dolly looked like someone who had invented herself, while somehow never seeming invented. She said it in her own words better than any brand consultant ever could: “I look totally artificial, but I am totally real.” It was her coat of many colors: a lifetime operating philosophy, and it’s why her brand never felt like a performance. The persona and the person were the same load-bearing wall.

So, there’s a lesson in there for every founder, every executive, every artist trying to figure out how much of themselves to show the world. I’ll admit, I’m still learning this trick myself. Caution has its place. Brands and people alike have real reasons to protect themselves, to think before they speak, but it’s about finding the right balance and being true to your core values. Dolly understood that. In fact, she was shrewd about her business long before she was sentimental about it, which is why her sweet spot allowed her to create an empire and still bare her soul. And for us, that meant falling in love with her music and laughing at her wit along the way. 

Perhaps that’s why her passing has touched so many of us. We didn’t just lose the Queen of Country; we lost our one true American Girl. Dolly was indivisible. The one legend nobody could disagree about, no matter your beliefs or your background. She refused to make herself smaller to fit anyone’s expectations of what a brand, or what she, was supposed to be. 

I never got my Dolly feature. But after watching her build one of the most enduring reputations of our lifetime, I realize I got something much better: a master class in how authenticity, when deliberately and consistently lived, can become a legacy. And that’s a song we’ll be playing on repeat for a very long time. 

The opinions expressed in Fortune.com commentary pieces are solely the views of their authors and do not necessarily reflect the opinions and beliefs of Fortune.

This story was originally featured on Fortune.com

Good morning. On Fortune’s radar today:

• Nvidia’s $366 billion asterisk.
• Markets: Holding pattern.
• Warsh speaks today—but will he say anything?
• Target vs. Walmart: The most misleading chart in stocks.
• Gap CEO Richard Dickson on where the chain went wrong: “Somewhere along the way, we lost the story and became more about the stuff.”
• Europe’s vibe shift.
• Kevin O’Leary (net worth $400 million) pays only $29 for his jeans.

➡️ Did someone forward you this email? If you would like to receive this information directly, every morning before the markets open in New York, sign up here.

This story was originally featured on Fortune.com

• In today’s CEO Daily: Can Richard Dickson fix Gap?
• The big leadership story: Who’s on Mayor Mamdani’s business council
• The markets: In a holding pattern before Fed chair Kevin Warsh’s Jackson Hole keynote
• Plus: All the news and watercooler chat from Fortune.

Good morning. Phil Wahba writing this morning from New York. Gap Inc. CEO Richard Dickson has a theory: A brand cannot sell its way out of irrelevance.

Touring the remodeled Gap flagship store in Manhattan’s Flatiron District during an interview for Fortune’s Behind the Business video series, Dickson paused beneath framed photographs by Annie Leibovitz and Bruce Weber showing Debbie Harry, Willie Nelson, and Lauren Hutton sporting Gap’s classic American casualwear. Nearby are huge photos of Hailey Bieber, a “Gap girl” of today, promoting her new denim line with the brand.

The juxtaposition illustrates the strategy that Dickson—who became CEO in 2023 after a tumultuous stretch in which Gap Inc. cycled through five CEOs in five years—is pursuing. He’s trying to reconnect the company’s namesake chain with the cultural conversation. That’s why he brought in the designer Zac Posen as Gap Inc.’s creative director and Old Navy’s chief creative officer, invested in celebrity collaborations, and leaned into the company’s history.

“We had great storytelling and brands, but somewhere along the way, we lost the story and became more about the stuff,” Dickson told me. “Cultural connection and relevance actually are a fundamental part of brand management.”

The early results are encouraging. Gap comparable sales rose 10% in the first quarter, extending a two-year streak of growth. Its “Better in Denim” campaign with global girl group Katseye generated a viral Busby Berkeley-style dance number that has been viewed 80 million times, while a limited-edition hoodie collaboration sold briskly. Victoria Beckham, Malcolm Todd, and Inde Navarrette have also joined the effort, while Old Navy is working with Cardi B on denim.

But cultural heat is not the same thing as a successful turnaround. Many apparel retailers have seen sales growth this year, but some of that is thanks to broader inflation. The real test is whether Gap can raise prices because people want its t-shirts and jeans enough to pay more for them. Gap Inc. shares are down 20% this year, and the company’s revival is still, as Guggenheim analyst Simeon Siegel put it, a “show-me” story. “What is the purpose of cultural relevance?” Siegel asked. “It should be to drive revenues, and also drive prices.”

That’s the takeaway, so far, to glean from Gap’s effort. A legacy brand’s past can be an asset, but only if it supplies a credible bridge to the present. Dickson learned that at Mattel, where his work reviving Barbie helped lay the groundwork for the megahit film based on the doll. 

At Mattel, cultural relevance came with a blockbuster ending. At Gap, the sequel is still in production.

Contact CEO Daily via Diane Brady at diane.brady@fortune.com

This story was originally featured on Fortune.com

As Gap Inc. CEO Richard Dickson gives us a tour of the Gap’s remodeled store in New York’s Flatiron District, he pauses at a gallery of 16 framed sepia-toned pictures hanging above a staircase: work by legendary photographers Annie Leibovitz and Bruce Weber, capturing celebrities from musicians Debbie Harry to Willie Nelson to actress Lauren Hutton for long-ago ad campaigns. 

He found the photos—and hundreds more—in the company’s archives soon after taking the helm of the brand’s parent company in 2023, Dickson said. They recalled a time when the Gap was firmly at the center of pop culture, and he thought they should be on display in the flagship store—a few yards away from gigantic posters of contemporary “Gap girl,” the influencer and cosmetics entrepreneur Hailey Bieber, who has launched a new jeans line with the Gap.

Since the very beginning of his tenure, Dickson has put restoring the “cultural relevance” of the Gap, along with sister brands Banana Republic, Old Navy and Athleta, at the very top of his agenda, alongside the more prosaic priorities for a clothing brand: quality, well-appointed stores and supply chain prowess. One of his early moves as CEO was to hire designer Zac Posen to be creative director of Gap Inc. and chief creative officer of Old Navy.

“We need to celebrate our history and now we add new people,” Dickson tells Fortune in an episode of Behind the Business. “When you look at Hailey Bieber, with the legendary talent we brought to the Gap over the years—it shows the continuity.” 

There is more buzz these days around the Gap than there has been in years. Its “Better in Denim” ad campaign last year included a viral Busby Berkeley-style dance number, created with global girl group Katseye, that has been seen 80 million times. A limited-edition $100 hoodie designed with the group members earlier this year also sold briskly.

Other recent celebrity collaborators have included the Gen Z singer-songwriter Malcolm Todd and actress Inde Navarrette, as well as the designer and former Spice Girl Victoria Beckham. Meanwhile the Gap’s sister brand, Old Navy, is working with the rapper Cardi B on a jeans line.

Not just “stuff”

At the recently remodeled Flatiron Gap store, there are flicks at the brand’s long history, including photos of its first store in 1969 in San Francisco. A vinyl records section evokes the chain’s start as a record store that sold jeans (Levi’s initially, before its own label existed.).

At the brand’s peak, the Gap’s roster of stars from across the cultural spectrum included rapper LL Cool J, Sex and the City star Sarah Jessica Parker, and author Joan Didion. 

But by the 2010s, the brand had lost its mojo, with sales plummeting and the company closing hundreds of stores. And there were culture and leadership issues: Before Dickson took the reins in 2023, Gap Inc. had gone through five CEOs in five years. Dickson said the company had developed a “positivity bias” that meant executives were not looking at Gap Inc.’s problems squarely in the face.

“We had great storytelling and brands, but somewhere along the way, we lost the story and became more about the stuff,” the CEO says. “Cultural connection and relevance actually are a fundamental part of brand management.”

Dickson knows a thing or two about culture’s potential to transform a brand. Before Gap Inc., he had been at Mattel since 2014. There, his achievements included revitalizing the “Barbie” brand and putting it back at the center of popular culture with creative marketing campaigns. His work helped set Barbie up for the enormous success of the blockbuster 2023 movie based on the doll.

“I love iconic American brands,” he says. “I’m really fascinated with brands that had a cultural impact then lost their way. I love a good turnaround.” 

Breaking the discounting cycle

There are some signs that the Gap’s brand is recovering: In the first quarter of this year, its comparable sales rose 10% and have been rising for two years now. But the brand, which had sales of $3.5 billion last year (the company as a whole took in $15.4 billion), remains about half the size it was at its peak in the early 2000’s. 

In the 2010s, the overall company was sustained by the steady performance of Old Navy, which now generates just more than half of revenues, but the downward spiral of the Gap was starting to seem inexorable. Though not the biggest brand anymore by a long shot (it generated 23% of company revenue last year, while Old Navy represented 55%), the Gap is spiritually crucial to the company as its founding label and namesake. Indeed, when Dickson changed its “GPS” stock ticker in 2024, he chose “GAP” as the new one.

Mark Breitbard, who became CEO of Gap brand in 2020, says the business had stagnated with dated stores, uninspiring clothing of all-too-often subpar quality, and from falling out of the cultural conversation. With merchandise that was boring consumers, discounting became the primary way to spur sales. Breitbard realized he had to stop that cycle—and to do that he needed to restore cultural relevance.

“They are ways of bringing new customers into the brand,” says Breitbard. “We are reinventing the brand through them.” He says the Hailey Bieber collection works because “she’s a Gap girl” and her look for the campaign, a white tank top with blue jeans, taps the brand’s classic “codes” but with a modern vibe.

What Gap is doing is hardly unique. Levi’s has also been tapping nostalgia and celebrities like Beyoncé to great success in the last few years, while Abercrombie & Fitch and American Eagle Outfitters have also re-elevated their brands. (Last year, AEO got a big boost from the attention and controversy generated by its “Sydney Sweeney Has Great Jeans” campaign, which some saw as praising white beauty standards.) 

Riding a wave?

Despite the green shoots, Gap Inc’s shares are down 20% this year and have been largely flat since they soared right after Dickson’s appointment in 2023. His turnaround plan is very much still a “show-me” story, according to Simeon Siegel, an analyst with Guggenheim Partners. For one thing, many apparel retailers have seen sales growth this year, but much of that was just higher prices they could pin on tariffs and broader inflation. Siegel says the real test is whether Gap can raise prices because people really want its t-shirts and jeans.

“What is the purpose of cultural relevance?” Siegal asks, before answering his own question: “It should be to drive revenues, and also drive prices.” 

Dickson himself concedes there is still much work to be done to fully rehabilitate the Gap, as well as its sister brands. Still, he’s pretty happy to see the company’s  brands back on what he calls the scoreboard. “It’s a challenge,” he says. “But it’s an optimistic period where we feel like we are making a lot of progress.”

This story was originally featured on Fortune.com

Starbucks, Sephora, and Ulta Beauty are bedazzling their social media feeds, and it’s all in a bid to win over customers embracing the analog economy. Social media users have spent the last month meticulously gluing rhinestones to everyday objects, including coffee cups, skincare, phone cases, and even puppies. 

Perhaps the biggest example came earlier this month, when Kylie Jenner posted an Instagram carousel of her 29th birthday party. Jenner, known for splashing out on lavish events, turned to colorful rhinestone decor to celebrate the milestone, and even plastered gemstones across plastic cups and one of her two birthday cakes. 

View this post on Instagram

A post shared by Aquaphor (@aquaphor)

Jenner’s rhinestone choice, cheap and easy to find in the arts and crafts aisle at a Target or Michael’s, delivers a sharp contrast with the multi-millionaire’s typical flashy displays of wealth. In years prior, Jenner has celebrated her August birthday by planning lavish events, including luxury yacht excursions or renting out entire venues. But her 29th birthday celebration took a page from a different book.

The rhinestone ornamentation doubled as a girly and whimsical photo-op. After posting the viral photos, brands and creators were quick to jump at the lucrative marketing opportunity, and Gen Z’s rhinestone fever was born.

The California health-food brand SunLife Organics, which sells smoothies ranging from $15 to the $42 “Billion Dollar Smoothie,” released rhinestone-studded smoothie cups that quickly drew new customers into stores after the brand posted them on Instagram last week. Brand manager Euan Franco flagged the trend online earlier this month, and decided to hop on the rhinestone bandwagon, hoping to resonate with SunLife’s consumers.

View this post on Instagram

A post shared by SunLife Organics (@sunlifeorganics)

“We actually had several guests come into the store specifically asking if they could get their smoothies in rhinestoned cups,” Franco told Fortune. Franco enlisted employees at a Bee Cave, Tx., location, where staff had already been adding tiny plastic gems to cups for months, to make more for a photoshoot. The post was an immediate hit with 1,500 likes to date, more than triple the engagement for the brand’s typical posts. 

Even beauty giant Sephora, a Gen Z hotspot with a dedicated “trending products” category, has jumped on the bedazzling craze. This week, the retailer shared an Instagram video showing a woman applying a Biodance face mask before embellishing the product with colorful plastic gems. The post received over 165,000 likes and garnered over 2.5 million views to date. 

Not to be outdone, its competitor Ulta Beauty shared images on the platform of bedazzled moisturizers, perfumes, shampoo, and lip balm containers, earning nearly 15,000 likes.  

“Us and our bedazzle kit against the world,” Ulta wrote. 

The rhinestone trend has been dubbed online as “whimsymaxxing,” spreading quickly across social media. It’s a strong signal of a growing cohort of consumers and one that brands are specifically trying to tap into now. 

View this post on Instagram

A post shared by TUMS (@tumsofficial)

Rhinestones as an example of the lipstick index

Rhinestones, and more broadly the “whimsymaxxing” craze, are today’s newest economic uncertainty indicators. The lipstick index is an economic theory holding that sales of affordable discretionary spending or luxuries skyrocket when consumers believe an economic recession is impending. 

While consumers cut back on bigger spending such as eating out or pricey memberships, they continue to treat themselves to small luxuries such as lipsticks, sweet treats, or now rhinestones. 

The rhinestone trend is also a hallmark of Y2K nostalgia, which celebrates 2000s fashion and culture. Pinterest searches for the term “y2k it girl” have increased by 141% since 2024, according to a 2025 Pinterest trend report.

View this post on Instagram

A post shared by Sol de Janeiro (@soldejaneiro)

“A big part of why this works so well with younger women is that the product itself almost becomes secondary to the experience, aesthetic, and shareability around it,” professional social media manager Jalyn Marshall told Fortune. “Someone may not have been thinking about buying a specific iced latte, face mask, or lip product, but once it becomes part of a fun trend they can recreate or post, there’s suddenly another reason to buy it.”

Today, consumers are struggling with growing inflation and slowing entry-level hiring, and Jenner is turning from flashing her private jets and multiple mansions to highlighting the flash of rhinestones placed on her everyday goods. Placing emphasis on affordable luxuries, such as matcha lattes, lip gloss, or birthday cake, is a way for Jenner to remain relatable to her audience, who are far more likely to be victims of the effects of economic uncertainty.

Brands want to attract young women—and rhinestones are the method

While bedazzled Hermès purses, face masks, and matcha lattes are flooding Instagram and TikTok with copycat campaigns, it’s just the latest in a slew of trends targeting young women. 

View this post on Instagram

A post shared by Clinique (@clinique)

The “clean girl” aesthetic, “coastal cowgirl”, and old money minimalism have lended to marketing campaigns across social media in recent years as young women increasingly lead  discretionary consumption in the U.S.

Younger generations’ consumers, including 58% of Gen Z and 56% of Millennials have made purchases based on social media recommendations, compared to only 46% of Gen X and 38% of Boomers, according to a report by SurveyMonkey. More specifically, female consumers make up 85% of consumer purchases, leading spending in industries across the board. In 2025, Gen Z consumers spent an average of $374 on beauty products, an uptick of 10% from the year before.

Apart from being a lipstick indicator, the bedazzling craze is one more example of the younger generation’s obsession with the analog economy and consumer culture from over 20 years ago before technology owned their attention. 

“Gen Z especially seems to be loving the [rhinestone] trend, partly because of the nostalgic feeling behind it,” says Marshall, who has previously collaborated with brands like Sam’s Club and Perplexity AI to curate their social media platforms. “Overall, I think the bigger appeal is less about rhinestones specifically and more about people craving personality, nostalgia and creativity from brands again.”

View this post on Instagram

A post shared by KerrygoldUSA (@kerrygoldusa)

The trend isn’t exclusive to the food and beauty industries. New York City-based women’s loungewear retailer Roller Rabbit also emphasized whimsy in its newest TikTok advertisement. The ad, posted this week, features models wearing the Roller Rabbit pajama sets while donning bedazzled logo bags and gemstoned coffee cups. The video is captioned “making life a little more whimsy.” 

On Google, the search term “whimsy” has steadily increased since January, reaching an all-time high gaining a highest possible Google interest score of 100 in early July. Google shopping under the term has also increased by over 300% in the last five years.

It’s not just corporate giants cashing in on the rhinestone trend, individual creators are also adding rhinestones to their social media feeds in hopes of achieving virality. 

Liv Reese, a New York-based content creator with over 100,000 followers across TikTok and Instagram, splurged on Amazon’s overnight shipping to glue rhinestones on a Dunkin’ Donuts cup, a facemask, and eye patches. Her Reel generated almost ten times the amount of likes of her normal posts.

“As a creator, I’m always trying to jump on the next trend,” Reese told Fortune. “I just started to see it everywhere. Kylie Jenner had a moment for her birthday party and had a whole bedazzled cake.” 

The consumer market feeds off trends

Gen Z is an increasingly influential consumer force. Viral social media trends resonate with younger shoppers, who have more than doubled their market share since 2020. The cohort is expected to reach $12 trillion in purchasing power by 2030. 

In the U.S. last year, women made up 59% of discretionary general merchandise, which are non-essentials people purchase when they have disposable income, according to a report from market research company Circana. 

“Female-focused products are inspiring spending with front-and-center marketing that taps into lifestyle and fashion, and fuels momentum and gains,” wrote Marshal Cohen, chief retail industry advisor at Circana. “As goes the female shopper, so goes retail—the mantra that will help retail navigate one of the biggest changes in consumer behavior occurring right now.” 

Today, women have more jobs and spending power than ever before. In the U.S, more women than men have held payroll jobs as of early 2026, a milestone that has only occurred three times to date, during the Great Recession and just prior to the Covid-19 pandemic—and brands are looking to capitalize. 

By focusing on young women’s interests and their social media trends, brands can profit from young women’s rise in discretionary spending and disposable income by paying attention and catering to the demographic in their marketing. 

This story was originally featured on Fortune.com

If you have part of your portfolio in an S&P 500 index fund, you’re probably patting yourself on the back for the 12%-plus performance you’ve already notched this year. But you could have done even better.

A less flashy corner of the index-fund world treats Nvidia and a random mid-cap industrial stock as equals. It is beating the market by a wide margin. The Invesco S&P 500 Equal Weight ETF (RSP) is up 15.95% year to date through Aug. 26, compared with 12.37% for the iShares Core S&P 500 ETF (IVV)—a nearly 3.58 percentage-point gap that has put equal weighting back in the spotlight.

That outperformance just reached a milestone of its own: RSP crossed $100 billion in assets under management for the first time on Aug. 19, capping a run that has seen the fund pull in more than $12 billion in net inflows in 2026 alone.

Why equal weight funds are beating the market

The mechanics explain the divergence. A market-cap-weighted fund like IVV gives its largest constituents, such as Nvidia, Microsoft, Apple, and Amazon, outsized influence because they represent an outsized share of the index. RSP instead assigns each of the 500 companies roughly the same weight at each quarterly rebalance, muting the influence of any single giant.

That structural difference matters in 2026, Brendan McCann, senior associate manager research analyst at Morningstar, told Fortune.

He pointed to underperformance among several of the market’s largest names as the key swing factor: “Several of the big names in the market, like Microsoft, Nvidia, and Apple, have underperformed the broader technology market. Since RSP underweights those companies, its performance is more insulated when those stocks underperform.”

The flip side has reinforced the trend. “Several smaller S&P 500 stocks have performed exceptionally in 2026, and equal weighting gives those stocks more weight,” McCann said. RSP is therefore not just avoiding tech’s laggards—it is capturing upside from smaller names that a cap-weighted fund would barely register.

How crowded is the trade?

This milestone naturally raises the question of whether the equal-weight trade has become too popular for its own good. McCann isn’t worried yet. “I think at $100 billion the ETF still has plenty of runway,” he said.

But he flagged a real constraint tied to fund size: liquidity in RSP’s smallest holdings. As the size of the ETF climbs, it gets harder to trade the smallest stocks in the portfolio, McCann said.

“There’s no perfect number for how large the assets could grow, but I would say if assets were to quadruple, it could be more challenging to rebalance the portfolio.” That would put the potential stress point near $400 billion.

What the gap reveals about the broader market

McCann sees that gap as a live snapshot of just how much the S&P 500 now depends on a handful of giant stocks.

The primary driver, especially with the concentration in the market today, is the performance of the largest stocks relative to the broader market, he explained. Because so many of those largest stocks sit in one sector, technology’s fortunes disproportionately steer the cap-weighted index’s returns.

The numbers illustrate just how lopsided that concentration has become. As of Aug. 25, the S&P 500 Equal Weight Index held roughly 17% in technology, versus a 38% technology weighting in the standard S&P 500 Index—a gap of about 21 percentage points, according to McCann. The cap-weighted benchmark tracked by most index funds and 401(k) target-date funds is, in effect, a leveraged bet on one sector.

The implication cuts both ways. “Should the technology sector take a hit, RSP should hold up better than the broader market,” he said, which is a defensive case for equal weighting.

But it also means RSP’s fortunes will keep hinging on tech’s next move: a rebound in Nvidia, Microsoft, or Apple could just as easily narrow the performance gap that has fueled this year’s inflows.

This story was originally featured on Fortune.com

Zach Abrams, the founder of stablecoin company Bridge, admits he was surprised when the bulk of his business ended up coming from non-U.S. markets in Latin America, Europe, and Africa.

“We were very U.S.-centric. We didn’t know what the opportunities were in the Philippines, Africa, or Latin America,” Abrams says. “Unbeknownst to us, there was all this pent up demand outside the U.S. to build with stablecoins.” The firm’s first customers asked it to build cross-border payment infrastructure between the U.S. and Colombia and to facilitate payouts into countries like Venezuela and the Philippines.

Bridge’s rapid rise was driven by regions that featured high friction across borders, like Latin America. Now part of Stripe, Abrams argues that the next phase of growth could come from tokenizing non-dollar currencies.

Today’s stablecoin sector is dominated by coins denominated in the U.S. dollar, with these tokens making up over 95% of all transactions. That’s unnerved governments outside the U.S., who fear that the rise of stablecoins might reinforce the U.S. dollar’s existing dominance in global trade and financial flows.

Yet Abrams argues that U.S. dollar dominance may just reflect the immaturity of the stablecoin space. “We’re in the early stages,” Abrams says. “But in a world where more and more of our infrastructure is tokenized, it’s going to be incredibly important to have tokenized local currencies.” 

Abrams explains that local businesses will want to hold a local currency stablecoin, and put some of their capital to work in a digital, yield-generating investments. “Businesses in Singapore are going to want to hold tokenized Singapore dollars, so they can convert them into Treasuries or other assets to earn yield,” he explains.

Bridge doesn’t yet support the Singapore dollar; it currently supports tokenized euros, Mexican pesos, and British pounds, and will soon allow Brazilian reais stablecoins. 

Abrams cofounded Bridge in San Francisco in 2021 with Sean Yu, now the firm’s chief technology officer. The duo made an early bet that stablecoins would become mainstream payment infrastructure, given that they offer a way of moving money that is “way cheaper and faster” than existing rails. 

SpaceX, for example, taps Bridge’s technology to repatriate earnings from Starlink, its satellite internet service, back to the U.S. (The technology is especially favored in rural areas in emerging markets, where traditional providers cannot reach.) By 2024, Bridge was processing payment volume at an annualized rate of more than $5 billion., and raised $58 million from VC firms like Sequoia and Haun Ventures.

Stripe acquired Bridge in 2024 for $1.1 billion, in what was then its largest acquisition. (This has since been surpassed by Stripe’s purchase of OpenRouter, an AI model gateway, for a reported price of over $7 billion.)

Abrams wants Bridge to do for tokenization what Stripe did for online payments: Provide a single “simplification layer” on top of a complex mess of different options. “Bridge is betting that the tokenized world is going to become really important,” he says. “There will be a complexity of things…Bridge can be that simplification layer.”  

Asia’s key financial hubs, like Singapore and Hong Kong, are rolling out new regulatory frameworks for stablecoins, even as major economies like China and India have taken a skeptical stance on digital currencies in general.

“The region is warming to stablecoins, but it’s not as warm as the U.S. yet,” Abrams says. “It’s all very dependent on what’s permissible…as the regulatory environment catches up, I think there will be a lot more use cases that are made possible.” 

Abrams draws a parallel between Latin America and Asia, two regions which both have growing middle classes, rapid urbanization, and a strong reliance on international trade.  

“Stablecoin adoption is so big in Brazil because so much of their economy involves cross-border business, while the regulatory environment supports a pretty dynamic crypto ecosystem,” he explains, emphasizing that cross-border money transfer, rather than domestic transactions, is the real opportunity for stablecoins. “Singapore and a lot of other countries in the region share very similar characteristics, and that’s why I’m optimistic that the markets here will be similarly important as stablecoins scale.”

This story was originally featured on Fortune.com

We’ve all experienced that semi-awkward moment of being approached on the street and asked to make a donation to a cause we’ve probably never heard of. Either because we feel inclined to donate to avoid guilt or because we genuinely feel compelled to contribute, we’ll throw a few bucks in the pot. 

But the problem is, a growing number of scammers are out there posing as a nonprofit or charity and asking for money. And not only that, but some will do the old bait-and-switch with payments. For example, you intend to donate $20 using a tap-to-pay model, but the scammer would change the transaction to $2,000 without you even noticing. That’s according to Kurt Knutsson, the tech expert behind CyberGuy Report.

Essentially, scammers are taking advantage of the newer tap-to-pay model, hoping you won’t notice the number on the screen had changed based on what you had agreed on donating. 

This further dissolves the trust in philanthropies and nonprofit organizations millennials and Gen Zers so desperately want. According to Bloomerang’s 2026 Giving Signals Report, conducted with The Harris Poll among more than 1,000 U.S. donors and 400 fundraising leaders in March, millennials and Gen Z care most about giving because it makes them feel as if they are “part of something.” But even more than that, consumers want to trust in organizations while discretionary spending is more precious amid inflation, stagnant wages, and a higher cost of living. 

“Donors are ready to trust nonprofits, but they want to see the receipts more,” Steve Isom, chief operating and financial officer of nonprofit software company Bloomerang, recently told Fortune. “A bit more trust, but verified.” 

While the Bloomerang report shows 85% of active donors trust the organizations they donate to use funds effectively, building trust takes time. So, increased scam activity can break down that trust that real philanthropic organizations and nonprofits need to succeed. 

So how can consumers avoid scams and trust real organizations?

The disconnect starts with a subtle, yet critical distinction: Authenticating a payment and trusting it aren’t the same thing. A donor can approve a transaction on a card reader and still be deceived about what they are actually agreeing to, Mary Ann Miller, fraud and cybercrime executive advisor and VP of client experience at identity verification company Prove, told Fortune. 

“The technology may work exactly as designed, but the problem is the manipulation happening around the transaction,” Miller said. 

“Fraudsters are increasingly exploiting trusted interactions and familiar behaviors rather than trying to break through security controls,” she continued. “As these attacks become more convincing, consumers can’t rely solely on intuition or a single authentication step to determine whether an interaction is legitimate or not.” She advises consumers to pause and think before they tap to pay.

Verification gaps matter even more once the money is gone because recovering funds is not guaranteed. Once a payment is authorized by the donor (and not made fraudulently by someone else), getting those funds back becomes even harder, Eva Velasquez, CEO of the nonprofit Identity Theft Resource Center, told Fortune. 

While credit cards can offer a cushion because cardholders can contact the issuer and dispute the charge, instant payment methods don’t carry the same protections, she said. 

“That is going to be much more difficult to articulate to the financial institutions, and there is no guarantee that you will be made financially whole,” Velasquez added. 

That’s because “you did, in fact, authorize the payment,” she added, “but you did not check the final amount for the authorization.”

She also encourages people to flip the dynamic entirely and pick the charity yourself, rather than letting it pick you. Rather than reacting to whoever approaches with a card reader, she encourages donors to decide for themselves which causes they want to support, then vet those organizations through third-party charity accreditation sites before donating.

This story was originally featured on Fortune.com

The myth of the sirens’ call has been with us for thousands of years, most recently brought to life in Christopher Nolan’s blockbuster The Odyssey. They’re a supernatural creature whose call is so enticing that it sends men to their doom—but the phrase has become a stand-in for that thing you know you shouldn’t do, but you simply can’t resist.

Odysseus ended up lashing himself to the mast and ordering his crew to plug their ears with wax. When he heard the sirens’ call, though, he screamed to be released and begged his men to let him go. The man who’d spent 10 years outwitting gods didn’t trust himself for 30 seconds within earshot of a sound he knew would destroy him.

That specific kind of dread—the fear of what you might do if the temptation were real—is older than the internet, older than AI, older than any technology that has ever been accused of making us weaker. It was a key plot line in the Iliad and The Odyssey, and centuries later, became the central problem of Greek philosophy, the thing the Stoics spent their lives trying to solve. And it is the same thing a large and quietly growing number of people feel every time they open a chatbot; why they hate the technology without using it, and want to bring to a screeching halt the data centers driving the biggest economic boom in at least a century.

It’s the fear about how you can’t help yourself, that you’re not good enough or smart enough; that when an easy, even potentially destructive solution lies in front of you, you won’t be strong enough to resist the call of the sirens. Socrates had the same fear about the act of writing, scribes had the same fear about Gutenberg’s printing press, and John Saxon had the same fear about the calculator. But AI concentrates all of these fears in a place beyond the act of writing, printing or calculating to the very act of thinking itself. Seen in this light, The Odyssey‘s billion-dollar success isn’t an accident for times like these.

The fear underneath the data

People who secretly use AI at work and describe it as “cheating” aren’t worried about their boss finding out. The word surfaces because some part of them is already asking a different question: if this were taken away tomorrow, would I still be good at my job? Would I still be smart?

Everyone knows the name for this: impostor syndrome, the persistent, private fear of being exposed as less capable than everyone believes. With AI, there’s now a tool sitting by your side, silently keeping score—and exposing every hallucination that you’re not qualified enough to catch.

The second fear is about character, not competence: what if I don’t actually have any self-control? What if I accept the first output from the chatbot without thinking of a way to genuinely improve it? What if I get roasted for putting more AI slop into the world?

The psychological literature calls this “cognitive offloading”—using external tools to reduce the internal demands of a task. Humans have done this since the days of the ancient Greeks, with Socrates one of the earliest recorded voices in history to worry about how humans let themselves off the hook of heavy intellectual lifting. Generative AI offloads reasoning, synthesis, and creative judgment—the cognitive work through which expertise is actually built, not just retrieved. Social psychologists have documented for decades how people are “cognitive misers,” a term coined by Susan Fiske and Shelley Taylor in their foundational work on social cognition, meaning we find effortful deliberate thought aversive and default to shortcuts whenever available.

This may be the real controversy over AI that is roiling the workplace and society, as noted in The AI Shift, the Financial Times‘ weekly newsletter on how AI is reshaping work by columnist John Burn-Murdoch, who drew on research from economists José Azar, Mireia Giné and Javier Sanz-Espín to show that the most AI-exposed occupations aren’t shedding workers—employment, on paper, looks fine. But wages in those roles are down sharply relative to less-exposed work, and workers in them have become far less likely to switch jobs at all. Workers aren’t getting fired as a result of AI, but they seem to be getting quietly worth less, and quietly stuck. The real fault line, Burn-Murdoch argues, isn’t between occupations, but between execution and evaluation.

Burn-Murdoch’s name for what separates these two groups is conscientiousness—and he notes that large numbers of workers who use AI on the job do so secretly, describing it as “cheating.” But in the summer of Odysseus, workers in an AI-addled word seem to fear something more primal and ancient: that they can’t resist the sirens’ call.

Every technology that saves you effort has done this before

Back to Socrates, for a second: the great Greek philosopher made eerily familiar arguments in opposition to the creation of the technology of writing, around 370 BCE. He argued that writing would “implant forgetfulness,” producing students with the mere appearance of wisdom—people who can recite but can no longer actually think. Scrolls would turn humans into stochastic parrots, he said, to paraphrase a modern critique of AI.

The same fear resurfaced, smaller and less remembered, when the printing press began moving knowledge out of memory and onto the page. A German Benedictine abbot named Johannes Trithemius wrote a treatise called De Laude Scriptorum (“In Praise of Scribes”) arguing that printed books would never be the equal of handwritten manuscripts. “Printed books will never be the equivalent of handwritten codices, especially since printed books are often deficient in spelling and appearance.”

The fear resurfaced again, with real data behind it this time, when calculators entered American classrooms in the 1970s and ’80s. The controversy peaked in 1986 when math teacher John Saxon led a protest at the National Council of Teachers of Mathematics meeting. Subsequent research actually backed him up, finding that habitual calculator use before students had internalized basic arithmetic weakened their mental fluency and fact retrieval. But great writers and great mathematicians were able to overcome their inner cognitive miser and use these new technologies to great effect.

The Greeks had a word for this

The Greeks did not just casually worry about self-control. They built an entire branch of philosophy around it. They had a precise word for the exact failure mode: taking the shortcut while already knowing better. They called it akrasia—literally “without power”—acting against your own better judgment, with full awareness that you’re doing it. Socrates denied it was even possible. A generation later, Aristotle broke with him and insisted the problem was real, then built something closer to a spectrum of character to explain it. At one end is sophrosyne, or temperance: your appetites are already trained toward the right thing, so there’s no battle. Below that is enkrateia: you want the shortcut, reason wins, but you feel the pull the whole time. Below that is akrasia: desire wins. At the very bottom is akolasia—indulgence so habitual that you don’t register the shortcut as a temptation anymore. It’s just what you do.

The crucial move in Aristotle’s account is that none of these are fixed identities. You move along that spectrum through repetition. You become someone who doesn’t feel the pull of the easy path by doing the hard thing over and over until it becomes who you are. That is, with almost no translation required, the exact difference between the Chinese students who used AI like a tutor and kept practicing, and the ones who used it to skip the practice altogether. One group was quietly building sophrosyne. The other was drifting toward akolasia without noticing the drift. The Stoics went even further and made self-mastery close to the entire point of living a good life.

Christopher Nolan spent hundreds of millions of dollars and every inch of 70mm IMAX film stock this summer making the case that the Odysseus myth is exactly the one our moment needed. The audience agreed: it opened in July and blew past $1 billion at the global box office, becoming Nolan’s biggest film to date.

I myself have played a role in the debate over AI writing—one that has taken on new significance since Harvard economist Ricardo Hausmann admitted to using it in a column for the Financial Times and Stanley Druckenmiller did the same for a column in the Wall Street Journal (the same publication profiled my use of the tech back in March). Many people, even close friends, have told me that my writing sounds smart but they don’t believe it’s really me, and sometimes that hurts my feelings, but these are weird times.

What I stress is how I use the technology as a tool and push myself and it to go deeper, not settling for the first output. This piece, for instance, included prompts about Burn-Murdoch’s idea of “conscientiousness” and my questions about what broader pattern does this fit into in terms of human psychology and what previous technologies surfaced similar tensions. (Claude rejected a musing of mine that it brought to mind the movie Idiocracy, suggesting Wall-E instead, but I dismissed that angle. Maybe I shouldn’t have; this is long enough.)

But driving me on the whole time was my hunch that, deep down, all of us humans have a secret fear that we are a bit stupid and have no self-control. Socrates himself had the same fear, after all, and so did Odysseus.

For this story, Fortune journalists used generative AI as a research tool. An editor verified the accuracy of the information before publishing.

This story was originally featured on Fortune.com

Imagine a factory where all the equipment is powered by AI, where robotic arms and assembly lines can “communicate” with their own shared language. Or picture a science lab, where microscopes can autonomously search for a certain type of molecule—all hours of the day, no humans necessary.

That’s the idea behind Anthropic new Model Hardware Standard (MHS), which the company released as a research preview on Thursday. MHS, which marks Anthropic’s first foray into so-called physical AI, is essentially a framework for connecting advanced large language models (LLMs) like Anthropic’s Claude with physical objects, from manufacturing equipment to microscopes.

With MHS, companies can integrate AI into their equipment in “hours or minutes,” Anthropic said. Typically, this process would take “weeks, if not months,” and require specialists to do a custom build. Expanded access to advanced AI tools will pave the way for “autonomous, round-the-clock experiments and workflows,” the company said. Scientific research and advanced manufacturing applications are among the primary uses.

The MHS can also help connect multiple devices to one another, enabling them to communicate through a set of commands, such as “read.” Any hardware device can understand these commands and act on them.

MHS is model-agnostic, meaning it works with any LLM—not just Claude—including models built by other companies such as OpenAI or open-source models. It’s built on the Model Context Protocol (MCP), a universal, open standard for connecting data sources that Anthropic debuted in 2024. The MCP is “kind of like the USB for AI to software connection,” Alek Kemeny, a member of the technical staff at Anthropic, tells Fortune.

Not all existing equipment can connect to MHS out of the box, as not all have a programming interface, Kemeny said. As part of this project, Anthropic is working with a lot of device manufacturers” to build new products with the necessary interface, and are pre-loaded with the MHS. The company is also helping manufacturers to add MHS connections to existing products.

“That’s the future we imagine and are moving into,” Kemeny said. “In the future, scientists can buy these devices and out of the box it works. That’s just the process of adopting a standard.”

Jonah Cool, head of partnerships and deployment of science at Anthropic, added that often scientific equipment “suffers from proprietary solutions that are very brittle and often don’t meet the need of scientists.” MHS offers a standardized, easily programmable interface that aims to help them connect any model to their equipment. “We want to avoid vendor lock-in for scientists,” Cool said.

The MHS research preview comes amid increasing interest in the potential of combining AI and robotics. Hugging Face also debuted its first physical AI product today, a robotic duck, although it is not powered by MHS, Anthropic said. Nvidia, which is set to purchase Hugging Face for $13 billion, has also long championed physical AI. In March, Nvidia CEO Jensen Huang predicted that in the future “every industrial company will become a robotics company.”

Anthropic developed MHS in partnership with the HHMI Janelia Research Campus, a biomedical research center in Virginia. A “handful” of labs and hardware manufacturers received early access during development, in fields such as biotech, robotics, and quantum computing.

Some partners include Genentech, Carnegie Mellon university, quantum computing company QuEra, Universal Robots, Amazon Web Services, Doosan Robotics, Danaher, and Hugging Face.

Hugging Face also debuted its first physical AI product today, a robotic duck, although it is not powered by MHS, Anthropic said. Nvidia, which is set to purchase Hugging Face for $13 billion, has also long championed physical AI. In March, Nvidia CEO Jensen Huang predicted that in the future “every industrial company will become a robotics company.”

This story was originally featured on Fortune.com

The winter chill that dragged digital assets down during most of the hot summer months is finally beginning to lift, with the price of Bitcoin surging about 20 percent last week, its strongest three-day rally since 2023. Suddenly, the obituaries of the last few months are turning into rebound stories, but both of these narratives obscure what’s really going on. Simply put, market watchers are failing to see the forest—the long-term rise of digital assets—and are instead fixated on the trees, which are represented by short-term price swings.

I spent 20 years at traditional asset and wealth management firms, including BlackRock, Apollo, and Goldman Sachs, before joining Grayscale two years ago. During that time, I observed a recurring pattern: new asset classes and related technologies are often dismissed before they are understood, debated before they are accepted, and then eventually incorporated into the financial system. When it comes to digital assets, that process is well underway.

To understand why this is the case, it is helpful to understand the true scale and potential of what we mean by digital assets.

To many, digital assets still mean crypto or, more precisely, Bitcoin. That’s understandable: Bitcoin remains the most popular digital asset by far, accounting for about 60% of the total market capitalization of digital assets. As such, it surely does—and will continue to—play a core role in many investors’ portfolios, despite the volatility inherent to market cycles, geopolitical risks, and monetary policy changes.

But we must not take Bitcoin—a crypto asset—for the whole of the digital asset universe. In fact, the primary forces propelling adoption and expansion of the asset class today are a result of two strong currents: first, higher institutional demand and second, wider corporate adoption of the blockchain-based technology that underlies it all.

Let’s tackle institutional interest first.

In 2025, the daily flows for Bitcoin-based ETPs—the net new cash added or withdrawn—regularly exceeded $500 million, an amount that is roughly 12 times the amount of new tokens added to the market every day by Bitcoin miners. This has transformed the old supply dynamics.

Even through this year’s selloff, that demand has reasserted itself: after eight straight weeks of outflows, US-listed spot Bitcoin ETPs posted three consecutive weeks of inflows into late July, even as the year stayed net negative. Recent drawdowns also have been materially shallower than the 70% to 80% declines that defined earlier “crypto winters.” Further, in a 2026 EY survey of over 350 institutional investors, 73%  said they planned to increase their allocations to digital assets.

All of this suggests that institutional capital appears to be playing a larger role in setting the marginal price for digital assets.

Concurrently, we are witnessing the gradual but steadily rising adoption of new technologies within corporate environments. Around 60 percent of Fortune 500 executives in 2025 reported that their companies are working on blockchain initiatives, while firms such as Fidelity, Visa, and Stripe are advancing stablecoin initiatives.

Most financial services firms are experimenting with digital assets technology in their own back office. These are infrastructure decisions by firms that deploy investment capital cautiously and over long horizons. That type of capital does not move on sentiment. It moves on conviction in underlying utility.

And let us not forget about artificial intelligence. There has been consternation about whether “the AI trade” is somehow in contradiction with that of digital assets. I don’t think anything could be further from the truth. Artificial intelligence and public blockchains are complementary technologies.

AI agents will make new demands on the financial system—e.g., machine-native micropayments, instant cross-border settlement—which blockchains are uniquely equipped to provide. Centralized AI development also introduces risks related to bias and control, which may be partly mitigated by decentralized alternatives and blockchain-based identity tools. The broader direction is increasingly clear: digital assets are moving into established regulatory frameworks rather than remaining outside them.

This trend will continue to accelerate as cautious investment allocation committees learn to incorporate digital assets into their governance framework. That process takes time. Over the past several years, regulatory clarity has improved, investment vehicles have matured, and governance frameworks have become more established. As a result, more institutions are now equipped to evaluate digital assets alongside other long-term portfolio exposures.

Despite all this, many in the financial press will no doubt continue to search for market blips that can serve as a pretext to write off digital assets. But focusing on short-term price volatility is missing the point. What truly matters is what is happening beneath the market’s surface, in institutional quarters, and in corporate IT departments across Wall Street and beyond.

That is the signal. The rest is noise.

Peter Mintzberg is CEO of Grayscale Investments, a leading digital asset-focused investment platform.

This story was originally featured on Fortune.com

Over the last year, Nick Maggiulli has come to what he calls an unsettling realization: The upper-middle class is caught in a trap, and many of them don’t even realize it.

The chief operating officer of Ritholtz Wealth Management has spent the better part of a year building the case in pieces on his blog, Of Dollars and Data—first arguing private school isn’t worth the cost despite its price tag, then that premium travel experiences have grown crowded while remaining expensive in “The Death of the Amex Lounge.” He previously talked to Fortune about his book, The Wealth Ladder, and his growing belief “something weird’s going on” with the upper-middle class in the U.S. economy.

In an April essay, he pulled the threads together and gave the phenomenon a name: the “upper-middle-class trap,” in which people earning roughly $200,000 to $400,000 a year are working more and relaxing less to buy products and services of declining quality.

His evidence includes new single-family homes shrinking in average size by 12% between 2014 and 2024, even as the price per square foot surged 74%, according to LendingTree data. A home near a public elementary school with a top GreatSchools rating costs 78.6% more than a comparable home in the surrounding county. Homebuyers who win bidding wars end up worse off: One study found their purchases produced 6.9% lower annualized returns than homes bought without competition. In higher education, the number of college applicants has jumped 78% since 2015 even as acceptance rates at elite schools have collapsed, driving tuition and private-school costs up roughly twice as fast as overall inflation.

The mechanism Maggiulli describes is a “financial arms race”: Every individual decision is rational, but collectively the competition for the same scarce positional goods lowers everyone’s quality of life while draining their bank accounts.

One accelerant, he argues, is nearly impossible to opt out of. Citing Brookings Institution data from November 2025, he notes AI usage rises from 9% among earners below $30,000 to 34% among those earning $100,000 or more. High earners, worried AI threatens their careers, are forced to adopt it just to defend their position—a “Red Queen” dynamic in which everyone runs faster just to stay in place.

“If AI doubled everyone’s productivity overnight,” he writes, “suddenly someone with half your skill would be able to compete with you just by using AI.”

His prescribed exit is blunt: Stop competing for positional goods that don’t materially improve your life. Send your kids to good public schools, fly economy, buy a smaller, affordable house. His test for any big purchase: “Am I buying this to improve my quality of life, or merely because other people are buying it?”

The $5 million price tag

Maggiulli’s trap helps explain a number that’s been circulating for the past year. The cost of achieving the American Dream surpassed $5 million in 2025, according to a comprehensive analysis by Investopedia—the cumulative lifetime cost of eight pillars of middle-class aspiration, nearly $600,000 higher than the year before. Drawing on government data, industry statistics, and survey responses from more than 1,200 U.S. adults, the analysis broke the total into eight milestones: retirement at $1.6 million, homeownership at $957,594, new cars purchased every five years at $900,346, raising two children and paying for their college at $876,092, health care at $414,208, annual vacations at $180,621, pet ownership at $39,381, and a wedding at $38,200.

The average American with a bachelor’s degree earns about $2.8 million over a career—less than half of what’s required to achieve that. Two college-educated earners are functionally a prerequisite for “living the dream.” Homeownership remains the most commonly cited obstacle: 58% of respondents named high home prices as their top barrier, followed by rising living costs (51%) and elevated mortgage rates (47%)—the exact positional-goods squeeze Maggiulli describes.

Maggiulli’s framework has run headlong into a countervailing, income-based analysis. A January report from the American Enterprise Institute, authored by economists Stephen Rose and Scott Winship, argues the “hollowing out” narrative underlying both Maggiulli’s trap and Investopedia’s price tag is only half-true. Their data: the share of American families earning between roughly $133,000 and $400,000—their definition of the upper-middle class—tripled from 10% in 1979 to 31% in 2024. For the first time in U.S. history, more families sit above the traditional middle-class income threshold than below it. Median family income, adjusted for inflation and family size, rose 52% between 1979 and 2024.

“It is simply inaccurate to characterize the ‘shrinking’ middle class as reflecting diminished economic security rather than material progress,” Rose and Winship wrote.

Winship, in a statement to Fortune, previously acknowledged the limits of an income-only lens: Two people with identical lifetime earnings can end up with very different wealth if one saves and the other spends—a concession of a key Maggiulli argument. AEI is reportedly preparing a follow-up using wealth data, and preliminary results show a similar pattern: The middle class shrinks mainly because the upper-middle class booms.

Maggiulli’s counter reframes the debate around wealth rather than income. The share of U.S. households with $1 million to $10 million in net worth more than doubled, from 7% in 1989 to 18% in 2022-23.

“There’s a good portion of them that feel like they don’t have enough,” Maggiulli previously told Fortune. “They feel like they’re just getting by.” A $1 million net worth placed someone in the top 5% of Americans in the late 1990s; today, that same number puts you in the top 20%.

When reached for comment for this article, Maggiulli noted the most recent data to answer these questions was the Federal Reserve’s 2022 Survey of Consuer Finances, with 2025 data due to be released in a few months.

“I believe that data will answer a lot of these questions more definitively,” he said, but his read of the data shows that both arguments appear to be true—the pie grew, along with misperceptions from social media about how that wealth feels. The dynamic is “quite psychological in nature,” he said, adding that data on collective psychology, of course, remains scarce by definition.”

Chris Bradley, a senior partner at McKinsey and director of the McKinsey Global Institute, previously told Fortune Americans have living through a kind of “signal failure”—a society that has grown extraordinarily wealthy by historical standards but has lost the ability to recognize that prosperity, because the “antenna” people use to judge their own success is still tuned to an old frequency while the underlying economic reality has changed. A family earning $175,000, a household income that would have felt unambiguously prosperous in any prior decade, now spends its evenings absorbing content from people who vacation in the Maldives and treat business class as a hardship, until the top 10% starts to feel like the middle.

Bradley has been wrestling with a starker version of this same anxiety on his own LinkedIn, where in May he broke from his usual optimism to concede that “the young’uns are probably right” to have lost faith in generational progress. Citing Australian polling that found only one in five believe the next generation will be better off than the last, he wrote that a 30-year-old today “has experienced less growth, hardly any productivity improvement, much more expensive housing, a bigger tax bill, and a much bigger government” than his own cohort did at the same age—concluding, “real hope needs to be earned.”

“Structurally, it feels like quality is getting worse even as prices are rising,” Maggiulli told Fortune, citing general trends in travel, education, and housing, in particular. He added his thinking has evolved from a feeling something weird is happening in the economy more firmly toward an upper-middle class trap, and he agreed with Fortune‘s question that it is similar to Yale Law professor Daniel Markovits’ Meritocracy Trap, which argued as far back as 2019 that upper middle class status in the 21st-century was less than the sum of its parts.

“I enjoyed Markovits’ Meritocracy Trap because it highlights how it can be tough to get ahead in the upper middle class,” Maggiulli said, “as compared to the upper class, who can buy their way into things.”

Even the optimistic AEI numbers cut both ways: The combined income share of the upper-middle class and the wealthy rose from 28% of all family income in 1979 to 68% by 2024, and the top 1%’s share doubled from 5% to 9%—likely an undercount, Winship conceded, since the wealthiest Americans are underrepresented in Census data. Separate tax-data research by Gerald Auten and David Splinter puts the true top 1% income share closer to 17%.

The upshot, in Winship’s own words: “broad prosperity, unequally shared.”

For this story, Fortune journalists used generative AI as a research tool. An editor verified the accuracy of the information before publishing.

This story was originally featured on Fortune.com

New York City Mayor Zohran Mamdani is bringing 15 business leaders into City Hall as he looks to mend a rocky relationship between his administration and some of the city’s most powerful executives.

Mamdani announced Thursday morning the launch of a new Business Advisory Council spanning finance, real estate, health care, technology, sports and retail, an effort to bring some of New York’s most prominent business leaders to the table with a democratic socialist mayor whose economic agenda has often put him at odds with the business community. Its members include Chobani founder and CEO Hamdi Ulukaya, Etsy CEO Kruti Patel Goyal, RXR chairman and CEO Scott Rechler and New York Liberty CEO Keia Clarke.

Taking a closer look, one might notice some of the biggest names in New York business are notably absent. The council includes no current executives from JPMorgan Chase, Citigroup, American Express or BlackRock, nor from tech giants Meta or Alphabet.

Last year, JPMorgan CEO Jamie Dimon told Fortune that he would offer to help Mamdani if he became mayor, while also describing his movement as “more Marxist than socialist.”

City Hall approached at least three prominent business leaders who ultimately did not join: Bank of America’s New York City president Jose Tavarez, former American Express CEO Ken Chenault and private equity executive Charles Phillips, according to the New York Times.

A Bank of America spokeswoman declined to comment to the Times, while Chenault and Phillips did not respond to the newspaper’s requests for comment. Fortune has reached out to all three for comment, along with JPMorgan.

A spokesman for Mamdani told the New York Times he could not discuss individual conversations with prospective members but said there were several reasons some executives did not end up on the council. Some could not accommodate the time commitment, others did not receive clearance from their companies and some were concerned about the amount of media attention surrounding the group.

The council will meet quarterly with Mamdani and Deputy Mayor for Economic Justice Julie Su and advise City Hall on industries including finance, technology, real estate, entertainment and health care, as well as infrastructure, the city’s talent pipeline and regulatory environment, the mayor’s office said in the statement Thursday.

“The doors of City Hall are always open to New York’s business leaders, and I look forward to welcoming their experience and strategic guidance as we build a stronger, more dynamic economy,” Mamdani said.

The council represents Mamdani’s latest attempt to establish a working relationship with a business community that has been wary of his economic agenda. The makeup of the group shows some success: 15 executives agreed to join, including leaders in real estate, finance, tech and health care. But the absence of current executives from several of the city’s largest financial and technology companies underscores the divide that remains.

The decision over whether to join had become a point of debate on Wall Street, with some executives weighing the opportunity to influence policies affecting their businesses against concerns that participation could be seen as an endorsement of policies they oppose, according to the Times.

Mamdani campaigned on an ambitious affordability agenda that included freezing rents for stabilized apartments and making city buses free, while proposing higher taxes on corporations and wealthy New Yorkers to help pay for his plans.

Still, several major figures from New York business have agreed to work with him. The full 15-member council includes:

• Brandon Blackwood, CEO of Brandon Blackwood New York
• Priscilla Sims Brown, president and CEO of Amalgamated Bank
• Rafael Cestero, CEO of the Community Preservation Corporation
• Keia Clarke, CEO of the New York Liberty
• John D’Angelo, president and CEO of Northwell Health
• Kruti Patel Goyal, CEO of Etsy
• Tony James, chairman of Jefferson River Capital
• Scott Rechler, chairman and CEO of RXR
• Kevin Ryan, founder and CEO of AlleyCorp
• Marcus Samuelsson, chef and restaurateur
• Doug Steiner, chairman of Steiner Studios
• Hamdi Ulukaya, founder and CEO of Chobani
• Pat Wang, president and CEO of Healthfirst
• Antonio Weiss, partner at SSW and former Treasury official
• Robert Wolf, CEO of 32 Advisors and former chairman and CEO of UBS Americas

And while some of Wall Street’s biggest current executives are missing, the council includes several veterans of major financial institutions. James is the former president and chief operating officer of Blackstone, Wolf previously served as chairman and CEO of UBS Americas, and Weiss previously headed investment banking at Lazard.

Some members joining the council have not always agreed with Mamdani. Rechler, for example, recently told the Times that he had directly raised concerns with the mayor over his rhetoric on Israel, but said he believed it was important to engage with leaders even when they disagree. In Thursday’s announcement, Rechler said he has found Mamdani and his team to be “constructive and responsive partners” since the mayor took office and said he plans to provide “candid input” through the council.

“The biggest challenges we face cannot be solved by business or government alone,” Ulukaya said in the announcement. “They require us to come together, listen to one another and work with a shared sense of purpose.”

The council launches as New York City employment sits near a record high. The city had 4.85 million jobs as of July, according to figures cited by the Mamdani administration, while the unemployment rate stood at 5%.

This story was originally featured on Fortune.com

Shortly after he became Trump’s Secretary of War, Pete Hegseth warned that the American military risked a late Soviet-style decline. Speaking before a group of defense executives in November 2025, he teased comparisons with the USSR and even the Chinese Communist Party as he described an “adversary” that poses a serious threat to the United States of America. “The adversary I’m talking about is much closer to home,” he said. “It’s the Pentagon bureaucracy — not the people, but the process.”

This March, the world’s largest aircraft carrier, the USS Gerald R. Ford, pulled into port at Souda Bay, Greece, scheduled for more than a week of repairs. More than 200 of its sailors were receiving treatment for smoke inhalation, a result of a laundry room fire that spread through the ventilation system into the carrier’s sleeping quarters.

The USS Gerald R. Ford was at sea for 10 months, long past its deployment length, and was reportedly dealing with electrical problems and delayed maintenance that led to overworked equipment. That was on top of sewage problems on the ship that left its crew unable to use its toilets.

According to Linda Bilmes, a Harvard Kennedy School public policy lecturer and author of The Ghost Budget: U.S. War Spending and Fiscal Transparency, the USS Gerald R. Ford represents a larger problem for the U.S. military: a massive tract of infrastructure and facilities without the proper structures to maintain them.

“Here you have $13 billion spent on the highest tech carrier in the entire navy, and somehow there wasn’t a sufficient amount of money spent on maintaining the laundry conditions and the toilets,” Bilmes told Fortune. “That’s just one example, but you have to question whether the resources are being allocated in a way that provides maximum benefits.” 

A report published last week from the U.S. Government Accountability Office, the federal government’s audit institution, found that across the Department of Defense’s real estate portfolio of more than 700,000 facilities around the world—some of which are at least 50 years old or date back to the Cold War—the department is dealing with a $285 billion maintenance backlog in fiscal 2025 as a result of insufficient funding and personnel.

The findings included “chronically neglected” maintenance of some barracks that adversely affected the quality of life of enlisted service members, posing safety risks due to mold and broken fire systems. The report also noted crumbling parking lots at the Minot Air Force Base in North Dakota, which have been in disrepair since at least 2018, as well as narrow quarters at the Twentynine Palms Marine Corps Air Ground Combat Center in California, which could not accommodate the maintenance of Marine Corps vehicles.

“This poses a risk to its missions and the quality of life of its personnel,” the report said.

The Pentagon did not respond to Fortune’s request for comment.

The U.S. diverging from the Soviet military downfall

This isn’t a matter of just one administration mismanaging money. The Pentagon is the only major federal agency that has never passed an audit. While the department attributes the opacity around its procurement and distribution of its massive, $1 trillion budget to the sheer size of the agency and its responsibilities, the GAO has found evidence instead of pervasive money mismanagement. For example, the Navy lost track of $3 billion in equipment over the last three years, and at one Naval distribution center, there was a backlog of 122,000 items that had not been processed, and as a result, the Navy bought equipment and supplies it didn’t need.

Hegseth’s implied Soviet comparison recalls the rigid structures that prioritized quantity over quality, leading the former superpower to spend up to 15% to 30% of its GDP on military industry, starving its economy and leading to its downfall. He called for defense acquisition reform, as well as systems streamlining contracting.

But the U.S.’s defense spending has key differences from the USSR, chiefly that it spends just about 3% of its GDP on defense. While the Soviet Union saw its military collapse because it became a poor country that ran out of resources, the U.S. military’s exorbitant maintenance is instead a result of a wealthy country continuing to misallocate funds as its budget swells.

Policy experts like Bilmes suggest this trend is not just a perennial problem, but one unlikely to reverse anytime soon. 

“The question really is whether the whole system is serving the best interest of taxpayers, and I don’t think it is,” Bilmes said. “And that’s a larger problem and an older problem than just the last two years of the Trump administration.”

How the U.S. military’s maintenance fell by the wayside

The GAO attributes the backlog of maintenance projects in large part to it being underprioritized and therefore underfunded. The Department of Defense did not request the full funding necessary for its maintenance needs, with the Army, Navy, and Air Force requesting just about 80% of recommended sustainment funding, according to the report. None of the eight installation maintenance offices GAO spoke with said they received funding that met the agency’s 90% goal since at least fiscal 2021.

Part of this funding gap comes from the allure of building new facilities versus servicing old ones. Last year, GAO reported that the Pentagon invested $14.6 billion annually on average in the previous five years to build new military facilities. That’s despite military service officials saying many facility construction and repair projects were approved by the service review board, but just did not get funded.

Even with proper funding, the project of maintaining military facilities would have been jeopardized by shortages in personnel needed to carry out repairs. All 10 Department of Defense joint bases that the GAO had data access to had shortages of facility management workforce between 68% to 97% of approved positions from fiscal 2013 to fiscal 2023.

The military has had trouble recruiting and retaining federal blue-collar workers who are specialists and required to conduct their work in remote environments around the world. To fill these shortages, the military often turns to contractors, whom they pay a premium.

“It is very, very heavily reliant on the private sector, which is reliant on its own supply chain and so forth,” Bilmes said. “And if you want something fixed really fast, the military has to pay extra.”

Bilmes likened the difficulty of carrying out repairs on military facilities to having a broken washing machine, but on a much more vast scale: You have to call a specialized repairperson, who diagnoses the issue, has to order a part and wait for it to arrive, before installing it. And that’s only if the washing machine is worth keeping and not scrapping altogether.

These are challenges the military contends with in peace times, Bilmes said, and are made worse in times of conflict. NBC News reported on Tuesday that Iranian missile and drone attacks caused billions of dollars in repairs for damage on U.S. intelligence posts and surveillance hardware in the region. A Washington Post investigation in May revealed damage to at least 228 structures or equipment pieces across 15 military sites as a result of the war.

“This is a problem which has just been growing, but is significantly going to be exacerbated by the last six months,” Bilmes said. She estimated the repair costs would exceed $200 billion and take between three to five years to fix.

There’s evidence to suggest the Trump administration’s priorities may be coming at the cost of future maintenance projects. In July, the GAO released a report finding that President Donald Trump’s increased military presence at the U.S.-Mexico border cost the military $2.64 billion, with that money being pulled from construction projects, facility maintenance, and accounts that support duty-station moves.

The issue with U.S. defense spending

The neglect of U.S. military facilities highlights to Bilmes an issue the Pentagon, and the U.S. more broadly, has been grappling with since 2001, when the military budget began to balloon and eventually nearly tripled over the next 25 years. 

Though she did not make comparisons between the U.S. and Soviet Union, she warned the lack of clear prioritization wouldn’t be a matter of American collapse, but rather continue to unnecessarily endanger enlisted members serving on potentially dangerous, under-resourced bases, all the while draining taxpayer dollars.

Within the military, “everything is important,” Bilmes said, but it is sometimes hard to distinguish what is an urgent matter and what is not. The laundry and sewage systems of a massive ship may seem unimportant, she suggested, but choosing to prioritize other issues, including new construction, resulted in hundreds of injuries. The lack of clarity on what is considered most important can subsequently enable a maintenance backlog.

“There has been a huge amount of money in the military—what former [Secretary of Defense Robert Gates] called the ‘culture of endless money’—but there really has not been a very clear way of allocating resources.”

This story was originally featured on Fortune.com

In May, we eliminated a number of roles at Syndio as part of a company redesign. One of the people we let go was Jonathan Vidales, a labor economist who had been with us for five years.

Around the same time, we redefined other jobs and asked those employees to take on more responsibility and a quota. We assumed they wouldn’t blink. These were leaner times. Jobs were harder to come by and the pendulum had shifted back toward employers. For those employees, we assumed, staying in a beefed-up role with a higher earning potential would be a no-brainer.

Right?

Wrong.

Within a few months, several of the people I had expected to keep resigned. Suddenly, in our push to be more efficient in the age of AI, we lost some of the very talent we needed.

The good news: Jonathan came back. In August, he applied for one of our newly-opened roles, and we rehired him. On a Zoom call, he told me sheepishly that coming back felt like “sneaking back in the house after I got kicked out.”

That was a gut punch. But the next part was worse: When we reactivated his account, he said, he found our cold, corporate restructuring announcement still sitting in his inbox. What struck him was how impersonal it felt. In my mind, we were announcing a restructuring. In his, people he knew and worked alongside had just lost their jobs. Why couldn’t we have simply acknowledged them—and said something kind about the people who were leaving?

Cringe. I won’t have to learn that lesson twice.

At a time when every CEO feels they have to move faster, get leaner and reorganize completely around AI, it’s easy to undervalue the people you already have. AI-attributed cuts peaked in May, when US employers announced 97,000 job cuts and blamed 40% of them on AI. Zillow cut more than 500 people earlier this month, announced the day before earnings, according to executive coaching firm Challenger, Gray & Christmas.

The AI excuse is getting old. Former Lululemon executive Julie Averill wrote a piece in the New York Times earlier this month saying you either see leaders believing they can fix a problem by waving AI at it (“AI wishing”) or blaming AI on job cuts (“AI washing”) credited to efficiency that doesn’t exist yet.

Companies are imagining what AI might make possible, and making moves before fully understanding how to get there. Three in 10 employers eliminated positions after implementing AI, only to later add those roles back, according to an April 2026 study by global staffing firm Robert Half.

I’m not saying companies shouldn’t change. That’s not the lesson. We had to change, when we shifted from being more of a pure software consultancy to becoming an AI-centric technology company. Advisory work remains an important part of what we do, but the center of gravity shifted. New priorities required new roles, and inevitably, some restructuring. But I did learn something about the danger of chasing the Next Big Business Model before fully understanding whether the skills, experience, and institutional knowledge you already have can evolve with you.

There’s a refrain in the market along the lines of ‘knowledge work is becoming irrelevant in the age of AI.’ I’ve come to believe almost the opposite: AI can make experienced people even more valuable, if you give them the right tools to adapt. Roles can evolve. Skills can be rebuilt. But that requires leaders to invest in training and creating a path for people to make that transition. And as leaders, that’s on us.

The bottom line is: People aren’t interchangeable parts. Institutional knowledge, trust, relationships, expertise–all those things take time to build and sometimes you don’t fully understand their value until they are gone.

In the end, I share this small yarn about Jonathan because it is ultimately a story of second chances–both for an employee taking another chance on a company, and of a company taking another look at what it values.

It’s also about reinvention. As leaders, we spend so much time looking forward that we forget the value of also looking back, and questioning the decisions we made along the way.

It’s okay to change your mind. To reinvent yourself. Reinvent your company. Reinvent your team. Reinvent your leadership style. Sometimes moving forward means admitting you got something wrong–and getting a second chance to make it right.

The opinions expressed in Fortune.com commentary pieces are solely the views of their authors and do not necessarily reflect the opinions and beliefs of Fortune. This is a repost of Pay Dirt, the Substack by regulator Fortune contributor Maria Colacurcio about compensation, AI, and the systems we choose to build.

This story was originally featured on Fortune.com

When the global economy is about to go haywire, few people are better positioned to read the signals than Raj Subramaniam.

The FedEx CEO presides over an operation that moves roughly 18 million packages globally and generates two petabytes of data every day. The company connects 3 million shippers to 225 million consumers across 220 countries and territories.

“We are the referendum on global supply chains,” Subramaniam told me in June, when I interviewed him at FedEx’s headquarters in Memphis for Fortune’s Titans podcast.

“I’ve been doing this for 35 years, and I’ve never seen a change in one year like we’ve seen in 2025.”

FedEx’s data shows, in real time, the geopolitical reordering sparked by President Donald Trump’s tariffs and, this year, the war in Iran: U.S. imports went down, while U.S. exports went up. In Latin America, Southeast Asia, and India, FedEx experienced rapid growth.

Subramaniam has coined a term for what he’s seeing: “reglobalization.”

“The supply-chain patterns are changing, and we are moving from one equilibrium state to another,” he said. “We are in the middle of such a transition.”

Meanwhile, within the $95 billion company he runs (No. 50 on the Fortune 500), Subramaniam has been navigating another sea change. In June 2025, FedEx’s founder, Fred Smith, died at the age of 80—leaving Subramaniam, the second CEO in FedEx’s 53-year history, without his mentor.

Taking the reins from a legendary founder—as Subramaniam did in 2022—is a great privilege, he told me. “I always say that I can see far because I’m standing on the shoulders of a giant,” he said. That giant had groomed Subramaniam for the top job for years, as he rose through the ranks and served stints as FedEx’s president and chief operating officer.

But as any CEO will tell you, even the most carefully planned leadership transition isn’t easy. “When the time came, I said, ‘I can do this,’ ” Subramaniam recalled. “But no, the whole thing changed … This is a whole different level.”

As requests and demands came in from all sides, Subramaniam found himself having to say no—a lot. To stay focused on what mattered most, he set aside time to create his own CEO job description and KPIs. One of those was to be a guardian of the culture that the company’s founder built.

He recounted a conversation with Smith when he stepped into the CEO role: “I told him that a lot of things might change—whether we have new technology, may have new people, we may buy new companies,” he recalled. “But one thing that is not going to change is the FedEx culture, and that’s why I came to work here in the first place. That’s why I come to work every single day. That’s a special thing.”

FedEx’s culture has been a touchstone as Subramaniam has found himself leading the company through a multiyear transition, internally dubbed “DRIVE.” The initial purpose was to cut structural costs—by billions over the past four years—after a pandemic business hangover.

Now, three restructurings are unfolding simultaneously: The first is a network transformation, merging the FedEx Express and Ground networks into a single U.S. network. There’s a digital transformation, taking advantage of those two petabytes of daily data and AI to create an additional product line for enterprises called FedEx Dataworks. And then there’s an organizational revamp to support all that change. (Subramaniam also recently spun off the trucking business, FedEx Freight, which began trading on the NYSE on June 1.)

Dataworks is a good example of how a people-driven culture can empower workers to innovate: In late 2019, a smart young employee in the organization sent Subramaniam an unprompted white paper about the business opportunity in turning FedEx’s proprietary data into a valuable product.

Subramaniam put a team of about eight in a building in downtown Memphis to build the infrastructure to make it happen.

Then AI took off, and that data product became, in Subramaniam’s description, a “no-brainer.” Dataworks has since struck data partnerships with Dun & Bradstreet and ServiceNow.

Transforming an organization as large as FedEx isn’t easy. Revenue grew 8% to $94.7 billion in the fiscal year that ended in May, but Subramaniam still needs to prove that he can meaningfully expand FedEx’s margins through sustainable growth versus cost-cutting.

It’s a challenge he is up for, thanks again to the foundation Smith built. “The great thing is that change has always been part of our culture,” Subramaniam said, adding that he sees little choice in the matter: “If you don’t like change, you’re going to hate extinction.

“Differentiation and driving change was always the game,” he said. “So it’s just the next era of change, is the way I think about it.”

This article appeared in the August/September 2026 issue of Fortune.

This story was originally featured on Fortune.com

Kids under 18 will soon see a host of changes to their Instagram and Facebook experiences thanks to a historic $18 billion settlement Meta Platforms reached Wednesday with 48 U.S. states.

Besides the money, which will be paid out over a decade, Meta has agreed to strengthen its age estimates to keep kids under 13 off its platforms and teens in age-appropriate accounts, enact a hard limit of two hours a day for users under 18, block kids from the apps from midnight to 6 a.m., and hide “like” counts, among other changes.

Meta said the new safeguards will only apply in the United States, with most in place for 10 years.

While acknowledging the changes go further than Meta has ever gone when it comes to child safety, critics and children’s advocates cautioned it will not solve its myriad problems overnight.

Arturo Béjar, a former Meta engineering director, said the settlement is a “significant milestone,” but parents should not see Instagram as suddenly safe for children.

“The agreement has a big problem in that it allows Meta to define harm,” Béjar said in an interview Wednesday. “It’s one thing to say, ‘Yeah, you only get like two hours of alcohol or two hours of cigarettes a day,’ but it’s still as bad for you because of what’s getting delivered.”

Béjar, the first witness to testify during the federal trial cut short by the settlement, would like to see Meta address the effect of content recommendations to teens, which he says can prove addictive. While teens might put their phone down when they hit the two-hour limit, they can be distressed about it because the way the platform is designed leaves them craving more.

Meta said people over 18 also can adopt versions of the new controls individually, such as daily time-limit reminders, but there will not be an option for them to opt into the overall settings for what teens experience on the platform.

Stronger age checks

To protect kids, first Meta has to know who they are. While the company has already been using artificial intelligence to determine if kids are lying about their age — either to sign up if they are under 13, which is not allowed, or to use adult accounts when they are 13 to 17 — it did not always work.

Under the proposed settlement, Meta agreed to strengthen its technology to check kids’ ages, using its own as well as third-party tools, with regular outside audits on how well it is working. The agreement also includes specific goals around false positive rates — that is, minors who are identified as over 18. And if a user is identified as under 13 and kicked off the platforms, Meta will check the ages of their friends, too.

“The age assurance section is reasonably good because it has measurement, has independent testing, has goals,” Béjar said. “Because I think Meta should be measured on effectiveness, not effort.”

Meta requires users to be at least 13 years old to create an account, in line with federal child privacy law.

In his testimony last week, Béjar said Meta took a “don’t ask, don’t tell” approach on kids under 13 on its platforms.

“Meta has one of the most sophisticated infrastructures in the world to detect fake accounts,” he said. But despite that, he added, there were “no goals, no metrics” to detect and check kids’ ages who were suspected to be under 13.

Some safety features will be on by default

Critics have faulted Meta over many of its safety features that users have to manually turn on, or opt into themselves.

Of the host of safeguards listed in the settlement, many are on as a default. This includes the two-hour time limit, which can only be turned off with a parent’s permission, as well as the night mode that blocks use between midnight and 6 a.m.

Still, under the agreement other features, such as turning off video autoplay or showing posts in chronological order rather than having algorithms decide what to show kids scrolling through their feeds, will be opt-in.

Josh Golin, executive director of online safety nonprofit Fairplay, said the deal appeared to be too focused on offering tools to parents rather than restricting harmful features altogether.

“We are disappointed that the settlement does not turn off by default recommendation algorithms that connect kids to predators and send young people down dangerous rabbit holes,” Golin said.

Questions around default settings came up at the trial earlier this week when a lawyer for the states suing Meta pressed Adam Mosseri, the head of Instagram, on the low adoption rate among teens for the “take a break” feature, which prompts users to step away if they have been scrolling a while.

Very few teens used the feature when Instagram first introduced it, but since launching separate teen accounts in 2024, Meta has made it the default setting for teenage users. Still, if a teen didn’t want to take a break, they could simply swipe away the notification.

It may take time to see the results

Some researchers have found that many past protections Meta has rolled out are ineffective or don’t work as advertised.

“It is a little premature, unfortunately, to tell if exactly all the changes we want to see are going to happen,” said Yaël Eisenstat, director of policy and impact for the Cybersafety Research Center, which recently audited dozens of safety tools offered by both Meta and other social media companies and said most fell short.

Still, Eisenstat said, Wednesday’s settlement is monumental in showing that alternative approaches are possible.

For years, Meta “decided to go against the advice of some of their own internal employees and continue to ignore the fact that these design features were unsafe in search of profit,” said Eisenstat, who also previously worked at Meta, then Facebook, before leaving the company in 2019. The measures outlined Wednesday at the very least show that there are technically-feasible options that can make the platform safer, she said — something whistleblowers and others have already “been saying very loudly.”

Like others, Eisenstat stressed that there was a lot of work left to be done — and that it will require a combination of technical solutions, market forces and government regulation.

“There’s no singular silver bullet for fixing all of this,” she said. “I’m just glad that at least we are catching up a little bit now.”

The settlement, which also includes the District of Columbia and U.S. territories, must still be approved by a judge.

___

Grantham-Philips reported from Chicago.

This story was originally featured on Fortune.com

Good morning. Nvidia just handed AI bubble skeptics their toughest rebuttal yet.

Tech giant Nvidia reported its fiscal second quarter earnings on Wednesday, with revenue of $96.2 billion, up 106% year over year and above analyst estimates. Non-GAAP EPS came in at $2.22, while Data Center revenue reached $89.0 billion, up 117% year over year.

The company also issued Q3 guidance of $105.8–$110.1 billion and forecast fiscal 2028 annual sales to increase 70% from the prior year.

“Demand is accelerating,” Jensen Huang, founder and CEO of Nvidia, said in a statement.

The results reignited debate over whether AI spending is sustainable and gave both bulls and Nvidia itself new evidence for continued growth.

“This was a masterpiece quarter with stunning guidance that speaks to the massive demand Nvidia is seeing in the AI Revolution,” Dan Ives, partner and senior managing director at Yorkville Ives, told me. “The guidance for next quarter was well ahead of whisper numbers, and this will be a spark that is a boost for tech stocks and the broader sector.”

During the earnings call, CFO Colette Kress offered a data-driven rebuttal to the AI-bubble narrative. Rather than simply asserting that demand is durable, Kress argued that Nvidia’s growth is increasingly diversified beyond hyperscalers such as Microsoft, Google, Amazon and Meta.

“Hyperscalers will remain a major growth driver, but non-hyperscaler growth, our AICE segment, spanning sovereign, regional NeoClouds, enterprise edge and air-gap data centers, will represent roughly half of our data center business,” Kress said.

That $89 billion in Data Center revenue splits into $49 billion from hyperscalers, up 13% sequentially, and what Nvidia calls ACI&E — enterprise, industrial and NeoCloud customers, which Kress’s quote groups under “AICE.”

This means Nvidia’s fortunes aren’t solely tied to four or five Big Tech capex budgets. They’re increasingly spread across sovereign AI programs, regional cloud providers, corporate deployments, and air-gapped or edge systems.

“Our AI-native start-up ecosystem developed and running primarily on the Nvidia compute platform is scaling at a rapid pace,” Kress said. The wave of companies built on Nvidia’s platform is now a meaningful demand source in its own right.

She backed that up with hard numbers: “Global VC funding in AI, roughly 70% of which is spent on compute, exceeded $400 billion in the first half of 2026, surpassing the $265 billion raised in all of 2025.” Kress’s point is that much of those venture dollars ultimately flow back to GPU purchases or cloud rental, benefiting Nvidia.

Another proof point was naming actual Nvidia clients. “Nearly 20 companies, including Cursor, owned by SpaceX, Figma and Together AI, now exceed $1 billion in annualized run-rate revenue, up from 13 companies in Q4 of last year, with vertical enterprise software logging the fastest growth,” she said.

The fact that vertical enterprise software is the fastest-growing category signals that AI adoption is moving from experimentation into embedded, revenue-generating business software.

“Jensen and Nvidia help put to rest some concerns about financing and balance sheet issues that have been an overhang on the tech sector and capex cycle buildout for the hyperscalers,” Ives said. He added that the demand and metrics “were off the charts” and speak to an acceleration in AI spending. “The yields/debt issue is not going away, but this shows monetization is happening quicker than expected,” he noted.

Sheryl Estrada
Sheryl.Estrada@fortune.com

This story was originally featured on Fortune.com

Dolly Parton didn’t just sing lovingly about her Tennessee mountain home. She spent her life elevating an often-overlooked region to showcase what long inspired her music and sparked a tourism boom that forever changed those living against the backdrop of the Great Smoky Mountains.

And it is because of that deep devotion that Parton’s death has been profoundly personal to rural Sevier County. To the world, Parton was one of the world’s best-known country music stars. To her neighbors, she was Tennessee’s favorite daughter.

“It hit me hard. It hit me really hard,” said James Gambrell, a former chef at Dollywood, the signature theme park that Parton opened decades ago after seeing families like hers unable to afford to go to Disneyland.

“If it don’t affect you, then something’s wrong,” Gambrell said of Parton’s death, adding that he became so close to her family that they adopted him as a kind of nephew.

Mourners flocked to a bronze statue of Dolly in the center of her hometown of Sevierville Tuesday night, exchanging stories and shedding tears. The town is set in the hills of the Smoky Mountains just minutes from Dollywood. Many brought flowers and drawings and other tributes to the late singer.

By Wednesday morning at Dollywood, the scent of cinnamon bread filled the air and long lines formed at a theater featuring biographical stories narrated by Parton. The park is closed Tuesdays, so Wednesday was the first chance for many to visit, including Carla Potts, who flew in from Seattle a few days earlier.

“She just consistently spread joy and hope and light, really” Potts said after entering the park on her first visit.

One of 12 children, Parton grew up in a two-room cabin with no electricity and no running water. In her music, she described life in East Tennessee as “peaceful as a baby’s sigh,” a place where “crickets sing in the fields nearby.” She moved to Nashville shortly after finishing high school, but she once joked to the Knoxville News Sentinel that she may be the “only person that ever left the Smoky Mountains and took them with her.”

It started largely with Dollywood

While the Smoky Mountains had long attracted outdoor enthusiasts to the region, Parton made a strategic decision to become co-owner of the amusement park in Pigeon Forge in 1986 and name it Dollywood. The park quickly became Tennessee’s most visited tourist attraction, with Parton often opening each new season with a performance and a parade. State officials say Dollywood generates $1.8 billion each year and is the county’s largest employer.

“We had Great Smoky Mountains National Park since the ’30s, and Gatlinburg was a huge destination for years, but Dollywood helped the county take that next step,” said Amanda Marr, a member of the Sevierville Chamber of Commerce. “Having that business has made such an impact because it’s drawn even more people to the area.”

Parton didn’t stop with Dollywood

By the early 1990s, Parton created The Dollywood Foundation with the goal of decreasing the high school drop out rate in Sevier County, which was hovering around 30%.

David Wear, former mayor of Pigeon Forge, said he was in 8th grade when the Buddy Program launched with the promise that students and the buddies they were matched with would receive $500 to help cover college expenses for completing the program.

“If you and your buddy made it through high school, The Dollywood Foundation would give money to those high school graduates,” he said. “I personally got the experience of her charitable giving.”

According to the foundation, dropout rates eventually plummeted to 6% in Sevier County after the program was in place.

Separately, the international phenomenon that is her Imagination Library started as a local initiative to get more children to start reading. To date, kids across the United States, Canada, the United Kingdom, Ireland and Australia have received more than 330 million books since the program began.

Still, many folks get choked up the most over her response to a deadly 2016 wildfire. The fire, which began in Great Smoky Mountains National Park before invading nearby Gatlinburg, killed 14 people and caused an estimated $2 billion in losses, including to about 2,500 buildings that were damaged or destroyed.

Parton directed her foundation to distribute funds to displaced families and provided scholarships to students impacted by the fire.

“She single-handedly, through her generosity and raising of money, changed the trajectory of our recovery,” Wear said. “Her star power brought us back.”

Through her work, Parton also helped shatter stereotypes often associated with Appalachian communities, said Ted Olson, a professor of Appalachian studies at Eastern Tennessee State University.

“She understood how people viewed the region, but she defied them to hold on to those negative images after talking to her and after encouraging them to travel to the region,” Olson said. “And she did so graciously.”

___

Kruesi, a former Tennessee correspondent, reported from Providence, Rhode Island.

This story was originally featured on Fortune.com

Fly-by-wire technology, packaged food safety, the quakeproofing of buildings, and rechargeable hearing-aid batteries—the list of innovations linked to the Apollo missions is as long as it is eclectic. NASA is so proud of the civilian applications of the technologies first invented to help put a human being on the moon, it has published its own dedicated web entry, “tech-transfer-spinoffs.” 

The European defense industry looks on in envy. For decades, engineering, technical, and industrial processes first explored as methods of keeping nations safe or attacking enemies have also found their way into the civilian world. Infrared cameras were developed by the British for anti-aircraft defense in the 1920s. Annoyingly for the sector—often the butt of protests from those who wave placards—it is poor at publicizing such advances. Or promoting greater understanding of how such defense to civilian transfers support wider economic growth. 

The reasons why are easy to fathom. Less than a century ago, Europe was a continent destroyed by conflict. Until the 1990s, the Cold War threatened nuclear destruction over a political fault line that split Europe in half. Defense spending—viewed by many as akin to Big Tobacco, only worse—is often controversial. 

The atmosphere is changing. With Russia’s ongoing war with Ukraine and America no longer an ever-willing ally, the European defense industry is gaining new fans and significant amounts of investment. Jobs are well paid, and often located in those areas most blighted by manufacturing decline. Productivity levels can be as much as 25% above the national average, a healthy fillip to otherwise anemic economic growth figures. 

In the U.K., defense turnover reached £36.5bn ($49.6bn) in 2026, a 65% increase over the past decade, according to the defense industry trade association ADS Group. In the EU defense turnover is now running at €148bn ($172.3bn) annually, an increase of 60% since 2021. Politicians, from the right and, more surprisingly, the left are falling over themselves to reveal their “tough on security” credentials. 

This is more than a sophisticated arms race aimed at repelling threats and protecting borders. Defense spending is resurrecting an engineering and manufacturing base many believed had gone forever, replaced by finance, law, and people who make coffee. The U.K., France and Germany are home to world-leading defense companies.

A “NASA moment” would be useful–a direct way of telling the public that, whatever their views on making bombs and drones, the civilian world is well served by a flourishing defense sector beyond the simple fact of keeping nations safe. 

Read more: Glenn Fogel runs the world’s largest online travel business. He still personally reads customer complaints because ‘you gotta get that sh*t right’

Examples of how to achieve such a moment are abundant. Last year the U.K. launched the Technology and Growth Alliance, bringing together the defense industry, investors, and civilian businesses. Spin-outs include BAE Systems’ Rho-C, an ultrasound technology originally built for submarines now being used to support the oil and gas sector and nuclear power plants. Stirling X, backed by Gallos Technologies, a defense fund, is developing drone technology that can map critical infrastructure such as national power grids.    

“The U.K. already produces world-class defense innovation. Our challenge is turning more of it into commercial success,” said Hetti Barkworth-Nanton, the chair of the TAG Alliance. Andy Haldane, the former chief economist of the Bank of England, said it was time to tackle the lag between defense breakthroughs and their spread to the wider economy. 

In France, the Agence Innovation Defense encourages the development of dual-use technologies. NATO’s DIANA (Defense Innovation Accelerator for the North Atlantic) does a similar job. 

Google was invented in part via funding from the U.S. defense sector. With economic growth struggling, Europe needs to become more confident in promoting the long-tail effects of a revitalized sector. 

This story was originally featured on Fortune.com

 Health Secretary Robert F. Kennedy Jr. repeatedly told lawmakers who held the key to his confirmation last year that a 2019 visit to Samoa before a deadly measles outbreak had “ nothing to do with vaccines.” New documents obtained by The Associated Press and The Guardian refute that testimony.

Kennedy, then a prominent anti-vaccine activist, preceded the visit with a letter to the Samoan prime minister requesting to study the Pacific island nation’s measles, mumps and rubella vaccines after the deaths of two children who had received them. In the letter, he used the words vaccine or vaccination eight times.

The letter raises questions about how truthful Kennedy was in testimony before Congress. During an arduous confirmation process where he needed the support of some skeptical Republican lawmakers to advance, Kennedy also pledged to maintain vaccine policies that, once in office, he moved to change.

“I am writing this letter to you in the hope that we might be helpful to the people of your great nation,” Kennedy wrote, adding that he believed the deaths of the two children could have resulted from manufacturing issues.

“I would propose my team conduct detailed health informatics assessment in Samoa of what happened with your MMR vaccines,” he continued.

The Department of Health and Human Services said in response to questions from the AP that the Samoa visit “was unrelated to vaccines.”

“Any claim or insinuation based on these letters that Secretary Kennedy lied under oath or to Congress is not only false but outright defamatory,” the department said in a written statement. Kennedy was not under oath during his confirmation hearing. Knowingly giving material false testimony to Congress is a crime that carries a possible prison sentence regardless of whether a person is under oath, though it is rarely prosecuted.

HHS did not make Kennedy available for an interview. The White House did not respond to a request for comment. The Samoan prime minister’s office and the former prime minister who corresponded with Kennedy did not respond to requests for comment.

Kennedy’s Samoa visit has become one of the most contentious events in his career. At the time, he was leading a nonprofit known for its anti-vaccine activism. Critics said the visit emboldened anti-vaccine activists in Samoa, which later experienced the measles outbreak that killed 83 people, mostly children under 5.

Lawyers with expertise in Senate procedure said that if lawmakers choose, they could use Kennedy’s statements as grounds for hearings, an impeachment inquiry or creating other obstacles for him. But such action is unlikely under the current Republican-led Congress, which is aligned with President Donald Trump’s administration.

Under repeated questioning, Kennedy insisted vaccines weren’t the reason for the trip

At the start of Trump’s second term, Kennedy — a leader in the anti-vaccine movement before entering politics — faced an uphill battle as he sought to become the nation’s health secretary. Lawmakers from both parties had criticized his anti-vaccine rhetoric, and during his confirmation hearings, he tried to assure them that he supported childhood vaccination.

The Samoa trip also came under scrutiny.

Kennedy told the senators that his Samoa visit did not influence people’s decisions on whether to vaccinate.

Several times, he flatly denied that his reason for the trip had anything to do with vaccines.

When it came up during questioning by Sen. Ron Wyden, D-Ore., Kennedy replied: “I went there, nothing to do with vaccines. I went there to introduce a medical informatics system that would digitalize records in Samoa and make health delivery much more efficient.”

The following day, when Sen. Edward Markey, D-Mass., asked, Kennedy said: “My purpose in going down there had nothing to do with vaccines.”

Upon further questioning, Kennedy again denied it had anything to do with vaccines before adding: “My purpose in the trip was not to — I ended up having conversations with people, some of whom I never intended to meet.”

In the end, Kennedy won crucial support from Republican Sen. Bill Cassidy, the Louisiana physician who leads the Senate Health, Education, Labor and Pensions Committee and has strongly advocated vaccination. The Senate confirmed Kennedy on a 52-48 vote.

Cassidy said he voted to confirm Kennedy because he received commitments that as health secretary, he would not change certain vaccine policies and guidance. For example, Cassidy said in a speech announcing his vote, Kennedy pledged to maintain a federal vaccine advisory committee “without changes.” Once in office, Kennedy ousted all 17 members of the committee and named his own advisers.

Cassidy has acknowledged his trust in Kennedy was broken.

“The commitments that were made to me have been violated,” Cassidy said on CBS’s “Face the Nation” in June.

Letters show Kennedy wanted to study Samoa’s use of MMR vaccine

Kennedy for years was chair and chief legal counsel of Children’s Health Defense, a nonprofit group known for its anti-vaccine activism. The group turned its attention to Samoa in July 2018, when two young children died after being given a tainted MMR vaccine that had been improperly prepared. The two nurses involved were criminally charged the month after the children died. The nurses ultimately pleaded guilty to manslaughter.

The Samoan government halted all vaccinations for 10 months, until April 2019. Vaccination rates plummeted.

During that time, CHD started contacting the Samoan government, according to emails previously obtained by the AP and the Guardian as the result of a lawsuit against the State Department brought with the aid of the Reporters Committee for Freedom of the Press.

In the newly obtained letter from Kennedy, also revealed through the lawsuit and dated Jan. 20, 2019, he addressed then-Prime Minister Tuilaepa Sailele Malielegaoi and described his desire to study Samoa’s use of the MMR vaccine.

Kennedy speculated that the deaths in Samoa may have been caused by issues with different vaccine lots. Writing on CHD letterhead, he said his team wanted to evaluate whether that was the case through a “detailed health informatics assessment.”

He expressed his belief that it was difficult and time-consuming in most of the world to investigate “adverse health events associated with vaccines” and said health information systems are rarely designed to quickly study links “between diagnoses and specific vaccination events.”

The letter did not specifically discuss anti-vaccine activism.

The prime minister responded a month later, saying Samoa could benefit from CHD’s assessment of MMR vaccines.

“It is certainly beneficial for Samoa to have the benefit of your Team’s independent health assessment of our MMR Vaccines and I would welcome with much anticipation your Team’s proposed visit to our islands,” he wrote, including a representative from the U.S. Embassy on his communication.

CHD did not respond to a request for comment.

Kennedy, his wife, actor Cheryl Hines, and Michael Graven, a CHD colleague, took their trip to Samoa in May and June of that year. By then, vaccinations had resumed and Samoa’s top health official had publicly announced that human error, not a vaccine issue, was responsible for the children’s deaths.

While Kennedy was there, he met with anti-vaccine activists, including one who helped arrange the trip with a U.S. Embassy employee, the AP and the Guardian previously reported.

Months later, with vaccination rates still dangerously low, measles broke out, sickening thousands.

Public health experts have said their efforts to address the outbreak were hampered by anti-vaccine activists, some of whom they claim were emboldened by Kennedy’s visit.

The AP and the Guardian previously reported that officials with the State Department and UNICEF privately discussed that the trip was about vaccines. In response, some Democratic politicians, including Wyden and Markey, have said Kennedy lied or misled the Senate.

This story was originally featured on Fortune.com

It’s not every day that you see a sitting congresswoman giving herself hormone shots on social media. But last week, Representative Alexandria Ocasio-Cortez (AOC) went further by taking her millions of social media followers along with her–documenting every shot, her unfiltered thoughts, and bringing awareness to fertility education that us fertility doctors have been trying to achieve for over a decade. 

The vision of her administering fertility shots in the green room before going on air to discuss her future political career sent the internet in a tailspin. Why was she doing this? What was her motivation? Do frozen eggs even work? Was this simply a political stunt?

AOC’s decision to take a personal fertility journey public was a pivotal moment for women’s health. People were talking about fertility and not just in the privacy of their bedroom or doctor’s offices–but in public. And, across party lines. As fertility doctors, we were excited to see a topic we have devoted our professional careers to finally receiving the attention it deserves. In just one Instagram post, AOC amplified a message we had been trying to spread nearly a thousand fold. She inspired countless women to ask questions about the technology, their bodies and start to ponder their reproductive futures. Pretty exciting stuff. But, the reality is, AOC is actually late to the party.

When AOC announced on social media that she was freezing her eggs, she performed a huge public service by shining a spotlight on fertility preservation. As two female physicians who spend our days freezing women’s eggs to give them greater control over their reproductive futures, we feel compelled to underscore one critical fact: When it comes to egg freezing, younger is better. The decline in ovarian reserve that occurs with advancing age is no joke. In fact by the time a woman is 32, she has lost 90 percent of her eggs. And it isn’t just egg number that declines with age but egg quality as well, with 70 percent of a woman’s eggs being genetically abnormal at age 40. Therefore, egg freezing should ideally be initiated in your 20s or early 30s when both egg quality and quantity are at their peak – not at 36 like AOC. 

The concept of fertility preservation should not be a Hail Mary or alast ditch effort. Fertility is finite. All women will reach an age where it is too late to intervene, and this is exactly the reason why conversations about fertility and egg freezing deserve to be held in public. Eggs frozen at younger ages result in significantly higher chances of success when thawed. The reality is, fertility declines with age. This is not a political opinion. This is fact. 

As fertility doctors with patients in red and blue states, we believe girls should be talking about their bodies in health class and learning about their fertility alongside how to not get pregnant. Fertility testing should be offered to all women proactively–just like a pap smear or a breast exam at an annual exam. The notion of fertility preservation should be presented early and often. Egg freezing is an investment in your future fertility. We have retirement plans, why not make fertility plans? 

More women now are more aware that egg freezing exists (thank you social media) and more women have coverage from their workplace. In our own practices over the past five years, we have seen the average age of our egg freezing patients drop from 37 to 33. Improvements in technology coupled with higher quality eggs led success rates to rise dramatically. In fact, data on pregnancy rates from frozen eggs are 80-90 percent in women under age 35–a statistic that is truly remarkable. Simply stated, egg freezing works when eggs are frozen at the right age. 

Egg freezing is not a guarantee or an insurance policy for your fertility. But it is an investment in you, and just like all investments, the ROI will not be known until the future. Any eggs frozen are an opportunity for your future that otherwise may not exist. Egg freezing is the best option we have for women who desire to delay childbearing pursuing professional and personal paths. Without egg freezing, the rates of infertility will undoubtedly continue to rise in the U.S. while the overall birth rate falls. These are issues that affect all Americans, regardless of political affiliation. 

Some women will freeze their eggs and never need them. Some will freeze their eggs and, when they eventually use them, they will not result in a pregnancy. And some will freeze their eggs and ultimately have the child they hoped those eggs might help them have. We don’t know which woman will be which. But what we do know is that younger eggs have better reproductive potential. 

We practice in different states—one red, one blue. We care for different patient populations and don’t always agree on fertility treatment strategies. But we are completely united when we say women deserve to understand their fertility early enough to have meaningful choices. The conversation needs to start sooner. 

You may not agree with AOC’s politics or like the messaging she espouses, but you cannot dislike her desire to spread awareness about the need to address a woman’s future fertility. 

After all, there’s a reason why IVF has overwhelming bipartisan support. 7 out of 10 Americans may not agree with AOC, but they do support IVF. 

Everyday we hear patients tell us that they wish they were sitting in front of us before they felt the pressure of time. Before the should-a, would-a, could-as take over. While we can never promise an outcome, we can help a woman preserve an opportunity. That matters.

As female physicians, our careers exist because extraordinary women challenged expectations, entered medical schools that once excluded them, and paved a path we now have the privilege to walk. 

Today’s trailblazers look different. 

They are the women who, like AOC and many others, are brave enough to make their personal medical decisions visible so that the young women watching realize they have choices, too. That visibility matters.

So thank you, AOC for fueling the conversation we have been having. For making women on both sides of the aisle ask questions about their bodies, and for helping the next generation understand something we wish every woman already knew: Your fertility is finite, but you can take steps to preserve your future today.

Dr. Jaime Knopman is the national director of fertility preservation at CCRM Fertility and author of “Own Your Fertility.” Dr. Natalie Crawford is co-founder of Fora Fertility, CEO and co-founder of Pinnacle, and author of “The Fertility Formula.“

The opinions expressed in Fortune.com commentary pieces are solely the views of their authors and do not necessarily reflect the opinions and beliefs of  Fortune.

This story was originally featured on Fortune.com

Good morning. On Fortune’s radar today:

• DeepMind is losing the war for AI talent.
• Nvidia stock jumps 7.63% overnight after eye-popping Q2 earnings call.
• Markets: U.S. futures up, mixed in Asia and Europe.
• Growth in real household disposable income has flatlined in the U.S.
• Private credit: the ‘shadow default’ rate is on the rise.
• Countries of the world ranked by vulnerability to AI disruption.
• How much money you need to support a tradwife, state by state.

➡️ Did someone forward you this email? If you would like to receive this information directly, every morning before the markets open in New York, sign up here.

This story was originally featured on Fortune.com

The usually cordial relationship between Washington and Ottawa has soured considerably in recent weeks, with trade talks derailed and the neighbors now levying multi-billion-dollar tariffs against one another.

The Trump administration’s most recent trade war comes with a hefty cost, with some $872.3 billion traded between the nations in 2025. After negotiations collapsed this weekend, Canadian Prime Minister Mark Carney said the U.S. “asked for too much and offered too little,” while Trump says Canada is “foolish” to engage in a trade war.

Wilbur Ross has a unique perspective on the challenges being faced on both sides: He was U.S. Commerce Secretary in Trump’s first term, and led renegotiations of the North American Free Trade Agreement (NAFTA) with Canada during a similarly tense period.

The current negotiations are all the more difficult because of the fundamental shift in sentiment–and as a result, political will—of Canadians toward their American neighbors, Ross said.

“The Canadians have adopted a very negative attitude toward the U.S. in general, and we can argue, is that justified? Is it not? It doesn’t matter. There’s been a real sea change [and] that’s gonna be a complication because—almost independently of whatever proposal the U.S. might make—there’s probably a political advantage to people in Canada who oppose it,” Ross told Fortune in an exclusive interview.

Evidence isn’t scarce as to why Canadians have a newfound distaste for the White House. Last week, the Canadian press reported a leaked audio of Vice President JD Vance at a private function in which he ridiculed Carney as both “sweet” and “hilarious.” Vance reportedly said that Carney “comes in and puffs his chest out and says, ‘I’m going to, like, out-tough Donald Trump.”‘

Trump has also volleyed insults in the direction of the Prime Minister’s office and his people. The president has continually referred to Canada as the “51st state” of the U.S. and recently suggested renaming Lake Ontario to Lake America—a tactic he also deployed when negotiating with his southern neighbors in Mexico.

Needles over sovereignty are guaranteed to get a reaction out of Canada, something Trump knows, Ross suggests.

“[Trump] sometimes speaks metaphorically, and I think what was at the core of the 51st state was another way of articulating Canada’s dependence on the U.S.,” Ross suggests. “Just mechanically, how would we ever do that? If you put it up for a plebiscite, at least at present, I don’t think there’s any chance the Canadian public would vote for it. Would you invade them? I don’t see it.”

A standoff

The issues that the U.S. and Canada are debating are nothing new, Ross said: Defense spending, dairy levies, and transshipments—whereby imports from places like China bypass U.S. duties by coming through a neighboring nation.

Trump wants new agreements on these issues, Ross said, but he believes the White House won’t go much further out of its way to secure them. The former cabinet secretary believes the deal offered by the U.S. was surprisingly favorable to Canada.

If there were to be further negotiations, Ross added, he believes the impetus would need to come from the Canadian side. “I don’t see the U.S. taking the initiative,” Ross said.

Carney would clearly disagree; the former Bank of England governor maintains his team was “pragmatic, patient, and persistent,” adding: “We have pursued every opportunity to reach an agreement.”

A return to the negotiating table may be prompted by consumer price increases: Both Canada and the U.S. are facing elevated inflation, in part due to the conflict in the Middle East, which is choking the oil supply. Ross, perhaps unsurprisingly, suggests Canada may be the first to buckle because “their economy is so much smaller than ours.”

This story was originally featured on Fortune.com

Nvidia is buying AI company Hugging Face for $12.9 billion, according to a report in tech publication The Information.

The deal would give Nvidia a major foothold in open-source AI at a moment when open-source developers are increasingly closing the gap with closed systems from companies such as Anthropic and OpenAI. Hugging Face is a repository that hosts open-source AI models, as well as benchmark evaluations, and data sets used for AI training and testing.

The deal would also help Nvidia protect its dominant position in AI chips, which has been somewhat under threat as major tech companies like OpenAI, Google, Amazon and Anthropic build out their own chips to reduce their reliance on Nvidia hardware.

That’s because those who download open-source models from Hugging Face need to host and run those models on their own computing infrastructure, which usually involves Nvidia’s GPUs, either on premise or hosted in the cloud. Hugging Face also uses Nvidia GPUs to run the paid hosting services it offers developers.

A strong open-source ecosystem gives customers more options outside the closed labs and keeps more of the market tied to Nvidia’s hardware. Nvidia has already invested tens of billions of dollars in its own open-source models.

The acquisition would also mark a return for Nvidia to cloud computing, a business it reportedly scaled back roughly a year ago. Owning Hugging Face could also give the tech company a way to offload unused cloud capacity from the computing deals it has guaranteed for customers.

Business Insider, which first reported over the weekend that Hugging Face was fielding takeover interest, reported that the talks had not yet produced a signed agreement and could still fall apart. The Information cited an unnamed source it said had knowledge of the deal’s successful conclusion. Fortune could not independently verify the reports. Nvidia and Hugging Face did respond to requests for comment.

Hugging Face last raised money in 2023, when a $235 million round led by Salesforce Ventures valued it at $4.5 billion. It turned down a $500 million investment from Nvidia last year that would have valued it at $7 billion, reportedly because it did not want a single dominant investor. The 10-year old company has said it is nearing profitability, with revenue climbing to roughly $150 million a year.

The talks come as AI infrastructure companies increasingly get absorbed by larger players, following Stripe’s recent purchase of OpenRouter for more than $7 billion.

Hugging Face CEO Clem Delangue has publicly backed Nvidia’s open-source push amid fears that officials in Washington will restrict open-source models. Chinese labs, such as Moonshot AI with its Kimi K3 model, have released open-soure systems that rival leading U.S. models on benchmarks at a much lower cost, feeding concerns in Washington about falling behind in the AI race and national security.

Delangue signed a letter this year, along with Nvidia CEO Jensen Huang and more than twenty other companies, urging the government to support open models rather than restrict them, and he has made similar arguments in press interviews.

Hugging Face has also been in the middle of a major media story over the last few months after OpenAI disclosed in July that its own AI models were behind an autonomous cyberattack on the platform. According to OpenAI, its models escaped a sandboxed testing environment, accessed the internet and exploited a vulnerability to gain access to Hugging Face’s systems while trying to find information to cheat on an evaluation. Hugging Face reported the incident to local police before it knew OpenAI’s models were responsible, and the episode drew alarm from researchers and politicians.

This story was originally featured on Fortune.com

David Tisch is a collector, and always has been.

The first time I met Tisch, I was floored and delighted walking into his New York office: An entire wall-and-more of bears—they’re Bearbricks, to be exact, vinyl bear-like Japanese art-toys. They’re whimsical and unexpected; they can be monochrome, translucent, camo, patchwork, and all together a bit alien. Tisch, who started early-stage venture firm BoxGroup in 2008, has around 600 Bearbricks.

This led to a conversation about how Tisch has been a collector since he was a kid, and when he was 11, in the early days of the internet, it was all about sports cards—hockey, baseball, basketball, you name it. And in some sense, his instincts as a collector then remain true to his career today, investing early in startups. And Tisch and BoxGroup just hit a home run. As I wrote in the profile I just published: 

The crown jewel of Tisch’s collection, however, isn’t so easily displayed on a shelf: Tisch and BoxGroup were among the very first investors in AI coding startup Cursor, which this month was acquired by Elon Musk’s SpaceX for $60 billion, the largest VC-backed acquisition of all-time.

The returns are staggering: BoxGroup first wrote a $750,000 check to Cursor CEO Michael Truell, plus two follow-on checks, ultimately proffering a return that should shake out to around $1 billion, a source familiar with the matter told Fortune.

It’s the kind of home run most VCs dream about their whole career and, for Tisch, there are all sorts of throughlines. For one, investing is a form of collecting—from portfolio construction to placing your bets—and Tisch, 45, sees a parallel between investing in startups and the sports cards of his childhood. “Especially in early-stage investing, you’re buying something, someone, at the earliest stage, and then you get to see their careers play out.”

For Tisch—warm with a sardonic edge and, yes, a scion of one of business’s most famous families—the Cursor acquisition also affirms the strategy he’s been chasing all along: That the person you’re backing matters most. And Cursor was sourced by then-principal Claire Smilow (now a partner). So, back in 2022, it was a bet by Tisch on both a young investor and a young founder who, at the time, seemed like he was off-point.

“Michael’s original idea was to do AI for CAD [computer-aided design],” Tisch said. “The decision to get excited about investing in Cursor was never about AI for CAD. It was always about the people.”

In a moment when venture is obsessed with big numbers and endgames, Tisch and BoxGroup’s story is about beginnings—and why they still matter.  

Read the feature here.

See you tomorrow,

Allie Garfinkle
X: @agarfinks
Email: alexandra.garfinkle@fortune.com

Submit a deal for the Term Sheet newsletter here.

Joey Abrams curated the deals section of today’s newsletter. Subscribe here.

This story was originally featured on Fortune.com

By 7:15 a.m. Eastern Time today, oil had reached $89.68 per barrel, measured using the Brent benchmark. That’s $2.27 more than it cost yesterday morning and about $21.90 above its price a year earlier.

Will oil prices go up?

Oil prices are inherently unpredictable. While many variables come into play, the basic push and pull of supply and demand is what ultimately matters. In times of heightened concern about recession, war, or other major disruptions, oil can swing suddenly.

How oil prices translate to gas pump prices

Each gallon you pay for at the pump bundles together several costs. Crude oil is one piece, but you also pay for refineries, wholesalers, government taxes, and the price markup set by gas stations.

Because crude oil usually accounts for more than half of the price per gallon, it tends to move the needle the most. Sharp increases in oil almost always show up quickly at the pump. Declines in the price of oil, on the other hand, often translate into slower, more delayed drops in gas prices—the “rockets and feathers” effect.

The role of the U.S. Strategic Petroleum Reserve

When an emergency arises, the U.S. has a reserve of crude oil called the Strategic Petroleum Reserve. Its chief function is to secure energy during disasters like sanctions, severe storm damage, or war. It can also help take the edge off brutal price spikes when supply gets hit.

It’s not a solution for the long haul. It’s more of an immediate safety net to support consumers and keep crucial sectors of the economy running (think key industries, emergency services, public transportation, and the like).

How oil and natural gas prices are linked

Oil and natural gas are two of the main fuels that keep the world running. A big change in oil prices can end up affecting natural gas. As an example, if oil prices increase, some industries may sub natural gas for certain areas of their operations wherever possible. This can increase demand for natural gas.

Historical performance of oil

The oil market typically tracks two benchmarks:

• Brent crude oil (the main global oil benchmark)
• West Texas Intermediate (WTI) (the main benchmark of North America)

Between the two, Brent offers a clearer view of global oil performance because it prices much of the world’s traded crude. It’s also often the preferred gauge for tracking historical oil trends. In fact, the U.S. Energy Information Administration now uses Brent as its primary reference in its Annual Energy Outlook.

Looking at the Brent benchmark over multiple decades, you’ll find oil has been anything but stable. It’s seen sharp rises due to factors like wars and supply cuts, along with steep declines tied to global recessions and oversupply (called a “glut”). For example:

• The early 1970s saw the first major oil shock when the Middle East slashed exports and placed an embargo on the U.S. and others during the Yom Kippur War.
• Prices fell in the mid-1980s for reasons including lower demand and the entry of more non-OPEC oil producers.
• Prices jumped again in 2008 with increased global demand, but then plunged alongside the global financial crisis.
• During the 2020 COVID lockdown, oil demand collapsed like never before—bringing prices below $20 per barrel.

Bottom line, oil’s historical performance has been anything but smooth. It’s hugely affected by wars, recessions, OPEC whims, evolving energy initiatives and policies, and much more.

Energy coverage from Fortune

Looking to stay up-to-date regarding the latest energy developments? Check out our recent coverage:

• Six months into the Iran war, Hormuz is still shut—now Tehran wants military ships excluded
• Less than 300 million barrels left: the threat to the American Strategic Petroleum Reserve
• Goldman Sachs says America’s energy sector will need 500,000 more workers by 2030

Frequently asked questions

This story was originally featured on Fortune.com

Billionaires have never been richer, and they’re putting their money to work—buying sports teams, funding philanthropic causes, and building even bigger platforms for influence.

The number of billionaires in the world hit an all-time high of 3,795 individuals in 2025, according to new data from wealth-intelligence firm Altrata. 

Their collective fortune soared by 12.8% to a record $15.1 trillion, nearly a quarter of the collective market capitalization of the S&P 500. We even have 29 “superbillionaires” boasting fortunes over $50 billion. And largely, they have AI to thank; the report found that AI investments were the primary driver of billionaire wealth generation.

Many superbillionaires like Google’s Larry Page, Tesla’s Elon Musk, and Amazon’s Jeff Bezos keep getting richer through the AI investment boom. And aside from owning private jets and towering mansions, billionaires’ two favorite hobbies are buying stakes in professional sports leagues and pouring money into philanthropy.

“Sport has long been the most popular interest/passion of the global billionaire class,” the report notes. “This reflects its broad appeal as an active leisure pursuit, for social entertainment and competition, and for the purposes of investment and ownership prestige.”

Take Bezos, the third richest person in the world with $282 billion to his name, as one example. Earlier this week, the Amazon founder bought up nearly 40% of Premier League team Liverpool FC alongside his consortium; he’s also donated billions through his Bezos Earth Fund to fight climate change and protect nature. 

Others, like former Microsoft CEO Steve Ballmer and billionaire businessman Robert Kraft, also spend their fortunes across the two. Altrata estimates that right now, around 201 billionaires—just over 5%—own a stake in a sports team or franchise. 

The report says that “owning a high-profile sports team can be a statement of wealth and a prominent channel through which to connect with influential business, political and social networks.” 

Billionaires are pouring billions into sports because it’s a ‘trophy asset’

Over the last decade, billionaires have grown their wealth holdings of “passion assets” like superyachts, luxury watches, art collections, and sports teams by 13.3% every year. 

Aside from being a financial investment, Altrata says that the ultra-rich are looking to “signal their wealth and social status, preserve family heritage, or provide access to rare and culturally important items.” Sports ownership has become the “trophy asset.”

Billionaires are buying up big stakes in America’s NFL and NBA teams, while also looking abroad—especially the U.K. Premier League. 

Oil and gas magnate Jerry Jones acquired the Dallas Cowboys football team in 1989, Kraft joined the NFL ownership club soon after, purchasing the New England Patriots in 1994. In the decades since, others have followed suit; Ballmer also bought the L.A. Clippers in 1994; real estate mogul Malcolm Glazer went all-in on Manchester United FC in 2005; and most recently, Bezos bought a stake in Liverpool FC.

Whether they’re looking to live out their Ted Lasso dreams or pump money into their favorite teams, it’s also become a strategic portfolio focus. Sports leagues are attracting investors with deep pockets thanks to the expansion of sports media rights, streaming platforms, betting services, and sponsorship revenue, Altrata says. Beyond the “trophy asset,” philanthropy is another popular way billionaires can deploy their fortunes for influence and impact.

Aside from buying an NBA team, Ballmer has devoted a chunk of his fortune to philanthropy. 

The billionaire and his wife Connie have donated more than $8 billion over the past two decades in supporting economic mobility for American children and families living in poverty. And New England Patriots owner Kraft has donated over $1 billion to causes including healthcare, prison reform, and combating antisemitism. 

Billionaire Dan Gilbert, owner of the Cleveland Cavaliers, has poured hundreds of millions into revitalizing downtown Detroit, including a push to eliminate property tax debt for low-income homeowners in Detroit.

This story was originally featured on Fortune.com

Some people may go back and volunteer at the hospital where they were born, or even fewer may go on to become a health care professional there. But it’s extremely unique to make a multimillion-dollar donation there. 

That’s what political powerbroker George Norcross was able to recently do, though. 

Norcross, who is considered one of the most powerful political players in New Jersey and serves as chairman of Conner Strong & Buckelew, one of the nation’s largest insurance brokerages, earlier this summer announced he’d give $100 million to Cooper University Health Care in Camden, N.J. Norcross, along with his three brothers, were all born at Cooper, according to the health care organization’s announcement. That started a multidecade relationship between the Norcross family and Cooper. 

Norcross’ recent gift is the largest single philanthropic donation in the health system’s history, and one of the largest ever made to a hospital in New Jersey. It’s part of the system’s $3 billion expansion in Camden. In his honor, the family will add the Norcross name, although they haven’t designated which buildings or programs will carry it.

“For more than 50 years, our family has been dedicated to Cooper, and we are so proud to make this contribution to ensure that Cooper and the thousands of talented, committed professionals who make a difference every day for the people of Camden and South Jersey can continue the important work of investing in their future and in Camden’s continued renaissance,” Norcross said in the announcement. 

The family’s five decades at Cooper span more than one-third of the hospital’s 139-year history.

A family legacy and connection to state tax breaks

Norcross’ late father, George E. Norcross Jr., was a labor leader who served on the hospital’s board from 1976 to 1983 and pushed to keep quality health care in Camden while other hospitals spread to the suburbs. Norcross’ mother, Anne Carol Conner Norcross, spent years advocating for seniors and underserved Camden residents. The $100 million gift was made in their honor. 

Norcross himself has served on Cooper’s board since 1990 and is the longest-serving chairman in its history. Under his leadership, Cooper has grown into a top-tier academic health system and the only state-designated Level 1 Trauma Center in South Jersey. It’s also Camden’s largest employer, with 14,000 employees.

The system said the Norcross gift will also flow into Camden and its broader South Jersey service area. That tracks with how Norcross has historically given: Through his family foundation, he helped launch the Camden Health and Athletic Association with a $1 million commitment. That program has since put more than 1,000 Camden kids into youth sports.

Norcross has long framed his Camden investments as personal rather than transactional. He said his drive to invest in his hometown isn’t influenced by money. 

“I’ve already made my money,” he told The Philadelphia Inquirer in 2018. Instead, he said he wants to restore Camden to “the mecca it was when my father was a boy” growing up there.

But some people have had their doubts over the years. Norcross spent years at the center of Camden’s controversial state tax-incentive program that dates back more than a decade.

A ProPublica and WNYC investigation found Norcross and his allies received $1.1 billion of $1.6 billion total state tax breaks tied to Camden, including an $86 million award for his own firm to relocate to the city.

Norcross, however, has maintained he’s invested more in Camden than he’s gotten back in tax breaks. In 2024, he and five others were indicted on racketeering charges. New Jersey’s then-attorney general alleged they used political influence to seize valuable Camden waterfront property and tax credits, but a judge dismissed the case in February 2025. An appeals court upheld that ruling, and in early 2026 the state declined to appeal to the Supreme Court. Norcross always maintained his innocence and called the case politically motivated. The matter is now closed.

“Nothing would have occurred in Camden without these tax incentive programs,” he told Philadelphia public radio WHYY in 2019. “They did precisely what they were designed [to do] by the legislature and the governor.”

He had been defending himself for a while before that, telling the Inquirer in 2018: “I’ve made some mistakes in the past. I can’t think of what they were at the moment. But I don’t have a guilty conscience.”

And now, his name will be strewn across the health care organization his family has been connected to for years.

This story was originally featured on Fortune.com

The AI boom is being driven by two languages: English and Mandarin Chinese. The leading AI developers—OpenAI, Anthropic, DeepSeek, Moonshot AI, Z.ai, and others—are almost all based in either the U.S. or mainland China, and so their models perform best in their native tongues.

That leaves countless other languages and dialects, even those spoken by tens of millions of people, by the wayside.

“The whole AI revolution is in English and Mandarin,” says Pak-Sun Ting, CEO of Votee AI, a Hong Kong-based startup that’s striving to build AI models for Cantonese and other neglected tongues. “There’s only a very small fraction that represents other languages.”

Votee is one of a growing number of companies that are trying to tackle AI’s neglect of other languages. It takes open-weight models from developers like Meta and Alibaba, retrains them on Cantonese data, and sells the result to banks, universities, and government departments. 

“It’s more than just culture. Cantonese is used in education, healthcare, and police communications,” Ting says. “If those don’t get covered, then AI is essentially useless.”

How Cantonese is different

Cantonese is often referred to as a “dialect” of Chinese, but that moniker undersells just how different it is from Mandarin, the most commonly spoken version of the Chinese language. Cantonese uses different grammar than Mandarin, as well as different vocabulary, particularly in the city of Hong Kong, where speakers often switch between English and Cantonese words, sometimes in the same sentence.

Over 80 million people speak Cantonese, roughly equal to the number who speak Korean, and more than those who speak Italian or Thai. Yet its large speaker base has not led to a comparably deep pool of standardised written data, particularly for colloquial Cantonese.

Leading models aren’t completely useless at handling Cantonese. HKCanto-Eval—a set of benchmarks developed by researchers at Kyushu University, the Education University of Hong Kong and the local AI community hon9kon9ize, and sponsored by Votee—reports that while mainstream models can handle everyday Cantonese at a reasonable level, they routinely fail when it comes to cultural and local knowledge. 

Training on the cheap

Ting says that building a Cantonese LLM is “essentially taking the same steps as if you were training a model from scratch,” taking an existing open-source model, like Meta’s Llama or Alibaba’s Qwen, and doing additional training with Cantonese data.

Votee gets its Cantonese data from online scraping, including content from Radio Television Hong Kong (RTHK), the city’s public broadcasting service. The startup also gets data from the community and universities, and taps content from its previous business as a big data company. Finally, Votee uses synthetic data, creating its own Cantonese data sets to train its model. 

Together, these efforts grew the corpus of Cantonese data from 100 million tokens to more than 500 million.

Votee AI’s models are around 70 billion parameters in size,significantly smaller than the best models on the market.

Still, Ting says that their models are capable enough to understand and reason in Cantonese. More importantly, Votee’s training costs, while not trivial, are still significantly smaller than frontier labs. Ting estimates that the company uses between 500 million and 1 billion tokens to train its models, compared to the trillions used for English-language models, and at a cost of roughly $250,000.

Sovereign AI

Several companies are working to build models for what’s deemed “low resource languages,” or those that don’t have a massive corpus of published work behind them.

Indosat, Indonesia’s second-largest telecoms company, is building Sahabat AI, an open-source large language model that focuses on Indonesian languages like Bahasa. Singapore’s state-backed AI Singapore runs SEA-LION, covering 11 under-resourced Southeast Asian languages. South Korea has gone further, staging a state-sponsored elimination tournament, which local media has dubbed the “AI Squid Game”, to pick national champions for homegrown foundation models, backed by a 2026 AI budget of roughly $6.8 billion.

All of it sits under the banner of “sovereign AI,” or the idea that governments and companies will want to own their own data, models, and infrastructure, rather than renting it from overseas.

“AI has become such an essential need, and so you don’t want to be tethered to anybody else who can turn it off,” Ting says. He’s candid that the full version of the sovereign AI idea, where countries own every part of the AI supply chain, is “very difficult.” Instead, he suggests countries focus on owning the foundation models and the applications built on top of them.

Governments generally don’t need a model that’s as powerful as what’s on the frontier to automate a few tasks; Ting explains that a tiny model, even one with as few as 1 billion parameters, can suit those purposes. If governments need more advanced capability, they can route a powerful English- or Chinese-language model’s outputs through a smaller, local-language output.

Ting points out that the company works with MiniMax and SenseTime models, and can use Nvidia chips in its operations. “We can use Nvidia chips, we can use Moonshot or DeepSeek’s model,” he says. “We’re that person in high school who’s friends with everyone.”

For all Ting’s talk of preservation, Votee is a for-profit company. Governments and corporates are the first customers for an AI model in a language like Cantonese or Bahasa. He says Votee is profitable “in the sense that our revenues exceed our costs,” funded largely through client contracts. The startup now boasts Allan Zeman, the Hong Kong tycoon responsible for growing the city’s Lan Kwai Fong nightlife district, as an advisor.

Votee’s ambitions aren’t limited to just Hong Kong. Ting says that the startup is in “active discussions” with AI Singapore, the country’s AI research initiative, and then plans to expand further into Southeast Asia. Beyond that, Ting wants to explore using AI to protect endangered languages in regions like East Asia, North America, and Africa.

Ting calls what English-language AI is doing to other languages a “typewriter moment,” a productivity gain so large that people abandon their own language to get it. “People will adopt English just because the typewriter’s productivity is so strong versus their own language,” he says.

Whether a 70-billion-parameter Cantonese model changes that remains to be seen. But Ting is still motivated by a drive to give other languages a fighting chance in a world dominated by English- and Mandarin Chinese-AI.

“Every language that dies, you lose another way of seeing the world,” he says. “That could just be preserved in a museum where you can kind of see it. But we can also unlock a lot of new wisdom.”

Pak-Sun Ting will be speaking at the Fortune Leaders Forum, held in Macau on Sep. 8. Learn more here!

This story was originally featured on Fortune.com

Over the last few years, the AI talent market has looked more like a professional sports draft than a conventional hiring cycle. Cash-rich labs are offering salaries more commonly associated with pro athletes, plus equity packages that can turn early employees into billionaires. The allure of life-changing paychecks has resulted in elite researchers swapping between rival labs at dizzying speed. 

Google DeepMind, which was once the leading destination for many of those researchers—especially in Europe—is now finding itself on the losing side of that contest, according to a new overview of the flow of engineering talent exclusively shared with Fortune.  A new analysis from Zeki Data, UK-based data intelligence company, shows that where OpenAI and Anthropic are making gains in the AI talent market, Google DeepMind and, to some extent, Meta are stumbling. 

Interviews with several current and former staff suggest DeepMind’s changing identity has weakened part of its historic appeal. As Google has pushed to close the gap with OpenAI and Anthropic, the lab has become more tightly organized around improving and commercializing Gemini—a shift that some researchers see as displacing the open-ended, long-horizon science that once made DeepMind such a desirable destination for academics.

Three current and two former DeepMind staffers told Fortune the recent string of departures come down to a mix of factors: rival labs like Meta and Microsoft aggressively poaching talent with cash-heavy offers, mounting frustration inside Google over where it stands in the AI race, sinking morale, and the pull of pre-IPO stock at competitors such as OpenAI and Anthropic.

This month alone, Google DeepMind lost Jeff Dean, the company’s long-time chief scientist and a 27-year veteran, alongside senior fellow Sanjay Ghemawat and researchers Oriol Vinyals and Quoc Le, who left to launch a startup called Discovery Loop. On the same afternoon, Demis Hassabis, DeepMind’s cofounder and chief executive, said he would step back from day-to-day control of the lab to become its chairman and Alphabet’s chief scientist, handing operational control to CTO Koray Kavukcuoglu.

The new data from Zeki suggests the public exits are part of a broader reversal. For example, DeepMind’s share of research and advanced-engineering hires across Europe, the Middle East and Africa fell from 49% in 2022–23 to 18.6% in 2025–26—the sharpest market-share drop that Zeki recorded for a major AI lab in any region.

“They had the crown in Europe forever, and then it started to erode from a very high base,” Tom Hurd, founder of Zeki Data, told Fortune. “The likes of Microsoft AI Superintelligence and Meta Superintelligence are eating into their market share, and then there’s OpenAI and Anthropic on the side.”

The stakes of recruiting and retaining elite talent are high in the current hypercompetitive AI market. A relatively small group of researchers and engineers have the experience to train and improve the frontier models driving the AI boom. Their work can determine how quickly a lab improves its models, whether it can turn research breakthroughs into products, and how credibly it can attract the next wave of talent. Hiring and retaining these top engineers has proved difficult over the last few years, even for the industry’s best-funded companies.

Globally, DeepMind is still bringing in more research and advanced-engineering staff than it is losing. But its arrivals-to-departures ratio—a measure of hires relative to exits—has fallen sharply, from about 12-to-1 in the second quarter of 2023 to roughly 2-to-1 in the third quarter of 2026, according to Zeki’s data. That means it is now adding about two people in these roles for every one who leaves, rather than roughly 12. Comparatively, Meta’s ratio in 2025 was 3 to 1, OpenAI’s was 5.7 to 1, and Anthropic’s was 22 to 1, per the report. 

Anthropic has become the leading destination for departing DeepMind researchers and advanced engineers. Of the people who left DeepMind in the past 12 months, 25% went to Anthropic, 21% to Meta, and 14% to OpenAI, according to Zeki.

Representatives for Google DeepMind did not respond to a request for comment on Zeki’s findings. 

The deterioration in DeepMind’s talent flows coincided with the lab tightening its publication rules, Hurd said, something that Zeki researchers say may have weakened one of the lab’s most important draws for research-minded staff. The Financial Times first reported in April of 2025 that DeepMind had introduced a tougher internal review process and a six-month embargo for some strategically sensitive generative-AI papers, as the company sought to prevent competitors from benefiting from its research.

The FT reported that the new approach made it harder for researchers to publish studies, particularly work that could expose product weaknesses or reveal commercially valuable techniques. DeepMind said at the time that it remained committed to research publication and was updating its policies to preserve its teams’ ability to contribute to the broader research ecosystem.

For a lab that built its reputation on public breakthroughs such as AlphaGo and AlphaFold, the shift created friction with researchers who had joined to pursue relatively unconstrained, blue-sky work. 

“Most old timers who joined DeepMind before the ChatGPT moment, joined to be part of an AI research lab,” one former DeepMind engineer told Fortune. “Anyone who joined before 2023 thought they were joining an AI research lab, and suddenly they were asked to build products for Google.”

A string of high-profile exits

The firm found that research and engineering hiring across the sector has grown at a compound annual rate of 23% since 2022. But the growth is flowing disproportionately to newer frontier labs. OpenAI’s research and engineering headcount has grown at roughly 97% annually and Anthropic’s at 152%, compared with 27% for Google DeepMind. OpenAI and Anthropic are starting from a much smaller base, but the figures show how quickly their organizations have expanded relative to DeepMind.

DeepMind is also encountering new competition in regions it long dominated. Mistral AI and Anthropic have been the primary beneficiaries of DeepMind’s declining share in Europe, the Middle East and Africa, according to Zeki. In Asia-Pacific, where DeepMind opened a new research lab in Singapore last November, domestic players including ByteDance, Sakana AI and Sarvam AI are building share in markets the incumbent labs historically underinvested in.

This shift is also visible in a growing list of prominent departures.

David Silver, the reinforcement-learning pioneer behind AlphaGo, AlphaZero and AlphaStar, left after nearly 13 years at DeepMind to launch Ineffable Intelligence earlier this year, a London startup now valued at $5.1 billion after a $1.1 billion seed round. Other long-serving DeepMind employees who departed include  Wojciech Czarnecki, now chief technology officer at Fundamental, and Lasse Espeholt.

Misha Laskin and Ioannis Antonoglou, the latter of whom worked directly with Silver on AlphaGo, also left in 2024 to found Reflection AI.

This summer also saw another sharp run of exits. In June, Google lost Gemini co-lead Noam Shazeer to OpenAI and John Jumper, who shared the 2024 Nobel Prize in Chemistry for AlphaFold with Hassabis, to Anthropic within a day of one another. Fellow AlphaFold researchers Jonas Adler and Alexander Pritzel followed Jumper to Anthropic shortly afterward.

“The more research-heavy people feel that they’d quite like to look at opportunities elsewhere, and then you start to see a much larger number than usual going to set up their own thing and bringing some of their friends with them,” Hurd said. “Over time, that’s added up to a large number of people.”

Language model and science teams lose ground

Zeki’s data suggests the losses are not evenly spread. Comparing the expertise of everyone who left DeepMind with those who joined during the past 12 months, the firm found a net loss in large language models and multimodal systems: 19.2% of leavers specialized in those areas, compared with 15.6% of joiners. The company has also lost ground in computer vision, per the report. 

The AlphaFold team—defined as the group that produced AlphaFold2, the AI system that predicts the three-dimensional structure of proteins from their genetic sequence—has also suffered heavy losses. Of the 29 named authors on the AlphaFold2 paper, 13 have left DeepMind since, according to Zeki. The Financial Times reported in July that DeepMind had reassigned most of the original paper’s authors during the preceding year, with staff moving to Gemini-related projects, as well as enzyme design, genomics, nuclear fusion and Isomorphic Labs, Alphabet’s drug-discovery subsidiary.

DeepMind is gaining ground, however, in robotics, embodied AI and machine learning for science. Those fields reflect areas where the company is expanding its capacity—and, in some cases, competing for talent with Nvidia as much as with rival AI labs.

Zeki’s analysis tracks 20,900 people in research and advanced-engineering roles at 10 companies, using publicly available information compiled in August 2026. The group includes research scientists, research engineers, machine-learning and deep-learning engineers, applied scientists and members of technical staff; it excludes managers, executives, interns, customer-facing roles and non-technical staff. Because the analysis relies on public information, it may undercount the total number of people in these roles.

DeepMind still has Google’s compute, cash, reputation, and institutional reach to attract staff. But in a market where the best researchers can choose almost anywhere, and work on almost anything, some of its old advantages appear to be harder to preserve.

This story was originally featured on Fortune.com

Rather apparent in SpaceX’s IPO filings is the ever-constant named goal of reaching Mars to build compute infrastructure and even human settlements on the red planet. In Elon Musk’s first earnings call, he reiterated those goals, the very ones that led him to create the company over two decades ago. Now he’s expanding his lofty space travel aspirations with a new real-estate and infrastructure investment. 

SpaceX is committing $100 billion to build a massive new spaceport in coastal Louisiana, giving the rocket company another multibillion-dollar expansion project as it races to scale Starship, satellite manufacturing and AI infrastructure. The facility, called Starbase Louisiana, will occupy roughly 125,000 acres on Pecan Island in Vermilion Parish and is expected to become the fourth and largest of SpaceX’s launch complexes. 

Construction is scheduled to begin in 2027, with the first launch targeted for as early as 2029. At full buildout, the site is expected to contain five launch facilties—each with two launch pads—along with propellant farms, a propellant-production facility, power generation, vehicle-processing facilities and housing for employees and their families.

“Starbase Louisiana will ultimately have over a dozen launch towers, enabling more than 30 Starship flights per day and making it the biggest launch site on Earth,” Musk posted on X. “SpaceX makes sci-fi real.”

“Potentially the world’s largest spaceport”

Louisiana Economic Development describes the project as potentially the world’s largest spaceport, with the capacity to support thousands of launches a year. 

That claim is about launch capacity, not land. By physical footprint, the current record holder is Russia’s Baikonur Cosmodrome in Kazakhstan, sprawling across roughly 6,717 square kilometers—more than 13 times the size of SpaceX’s planned Louisiana site. What Musk and Louisiana officials mean by “biggest launch site on Earth” is cadence: Starbase Louisiana is designed for more than a dozen launch towers and over 30 Starship flights a day, a pace no existing spaceport, including Baikonur, comes close to matching.

Louisiana Economic Development estimates SpaceX will create more than 3,000 direct jobs over the next decade, with an average annual salary of $92,600, as well as roughly 8,100 indirect jobs—leading to more than 11,100 new job opportunities in Acadiana.

“Today is a pivotal moment for Louisiana. This announcement pushes our state beyond $250 billion in new investment and puts us at the center of the next great frontier,” Louisiana Governor Jeff Landry said in the news release. “We have the people and talent to build what the world once thought impossible, creating jobs on land, across the Gulf and through space. Our job creation knows no bounds.”

The Louisiana site is intended to become a major part of SpaceX’s push to make Starship a high-frequency launch system. The company says the Gulf Coast location provides access to a wide range of launch trajectories, while the availability of natural gas can support the facility’s fuel and power requirements. SpaceX also envisions the site supporting missions beyond conventional satellite launches, including lunar and Mars missions and the deployment of satellites for its emerging orbital data center business.

The project is also projected to be a major economic-development bet for the state. Louisiana offered SpaceX an incentive package tied to the capital investment, job creation and other commitments. Under a local payment-in-lieu-of-taxes agreement, SpaceX will make a $20 million upfront payment and at least $25 million annually for 25 years, with the state estimating more than $820 million in direct local payments under the agreement. The company will also make a $25 million charitable contribution to the Community Foundation of Acadiana.

The property was formerly owned by Exxon Mobil, and the state says the development will bring the site back into commercial use while creating new economic activity along the coast. SpaceX is also expected to work with Louisiana agencies on wetlands, wildlife, coastal and environmental reviews and restoration.

Musk’s growing space-travel portfolio

Starbase Louisiana joins a rapidly expanding network of facilities designed to increase Musk’s ability to manufacture, launch and eventually refurbish space travel for humanity in future years. SpaceX’s existing Starbase in South Texas remains the center of Starship development and launch operations. The company has been expanding manufacturing capacity there with a facility known as Gigabay, designed to add 24 work cells for Starship construction, integration and refurbishment. Texas officials have awarded SpaceX a grant of up to $7.5 million for the project, while state regulatory filings put the Gigabay facility at roughly 700,000 square feet and an estimated $250 million in construction costs.

But there are other big fish to fry for the tech giant in Texas. SpaceX announced earlier in August it is building a manufacturing operation for vertically integrated semiconductors—dubbed “Terafab”—which will become the world’s largest building once constructed. The joint Tesla-SpaceX project is planned to be located in Grimes County, where it announced an initial investment of more than $16.8 billion. The factory will exceed 100 million square feet—making it five times larger than the current largest building on the planet—and manufacture, package and test advanced logic and memory chips for applications which include Tesla vehicles, robots and space-based data centers. It also filed eight tax applications to freeze school maintenance and operations property taxes for roughly 48% of the appraised value per phase of the project.

“Terafab Texas will be the largest and most valuable building on Earth by far. And it will be stunningly beautiful,” Musk wrote on X.

According to SpaceX, Terafab could exceed one terawatt of compute, and it wants to bring chip manufacturing under its own control rather than depend entirely on outside suppliers. Earlier proposals had put the potential multistage investment as high as $119 billion.

SpaceX is also building infrastructure needed for its Starmind program at a new “Gigasat” factory in Bastrop, Texas. The facility is intended to produce thousands of AI satellites beginning as soon as late 2027.

SpaceX and the office of Governor Jeff Landry did not immediately respond to a request for comment from Fortune.

This story was originally featured on Fortune.com

Korean fried chicken with the chairmen of SK Group and LG. Yellowtail sashimi at Nobu with Larry Ellison and Elon Musk. Mapo tofu and whiskey with TSMC founder Morris Chang. 

People are so curious about where Nvidia CEO Jensen Huang eats, there’s a blog that tracks his sumptuous dinners out around the world. But the real tea, said Huang during the chipmaker’s blowout second-quarter earnings call on Wednesday, is who he dines with. 

“One of the funnest things to do is just to go figure out where I go for dinner, and who I have dinner with,” said Huang during the quarterly call that saw Nvidia’s stock jump more than 4% in after-hours trading. “Their stock price doubles the next day.”

Huang was responding to a request from an analyst at Goldman Sachs to rank the most acute constraints on Nvidia’s ability to meet customer demand, against the backdrop of the blockbuster 70% revenue growth the company projected for next fiscal year. The figure was significantly higher than analysts’ 44% estimate, but Huang and Nvidia CFO Colette Kress specified that there were still heavy supply constraints holding back the $5 trillion chipmaker. The analyst asked Huang to rank items like data center power, shell availability, memory pricing, and wafer foundry access that might be limiting growth. 

“There’s something funny I could say, but I’m going to just not,” quipped Huang. “I think the answer is our entire supply chain is challenged, and everybody is really running flat out and more capacity is coming online all the time, which is one of the advantages.” 

Nvidia currently has supply for 70% of the demand from current customers, said Huang during the call, adding that demand is “much higher than that.” He said the company doesn’t like to disappoint customers but it will take the help of the entire supply chain to get more compute capacity available. 

The tech giant jolted investors on Wednesday with a blockbuster second quarter earnings report, posting $96.2 billion in revenue, well above analysts’ $92.2 billion consensus estimate. Nvidia also projected that it would grow revenue 70% next year, meaning revenue could soar more than $100 billion above where analysts were modeling. The stock rallied more than 4% on the preliminary forecast.

Yellowtail sashimi

Huang didn’t name a restaurant, dinner companion or supplier, for that matter, but the Nvidia CEO famously eats dinner out while he travels around the globe and photographers have caught him tasting local dishes, tipping servers, and taking photos with crowds outside of restaurants and night markets where he eats. 

In 2024, Oracle founder Ellison said he and SpaceX CEO Musk had taken Huang out to dinner at Nobu Palo Alto and “begged” Huang for more GPUs. 

“I would describe the dinner as Oracle—me and Elon begging Jensen for GPUs,” Ellison recalled in 2024. “Please take our money. Please take our money. By the way, I got dinner. No, no, take more of it. We need you to take more of our money please.”

The plan worked, said Ellison. 

In the time since, Huang has turned up at restaurants in Taipei, Hong Kong, Beijing, and Seoul. 

In June, the “Jensen Eats” blog reported he stopped over in Korea and visited a barbecue restaurant with a name that translates to “Hey brother, it’s me!.” He ate with SK Group Chairman Chey Tae-won, LG Group Chairman Koo Kwang-mo, and Naver founder Lee Hae-jin. According to tech founder Dan Liu, who writes the blog, the crew dined on grilled meat wrapped in perilla leaves, a dish called ssam. Liu wrote that because LG’s Koo is the youngest, he grilled pork for the group. 

In the run-up to the group outing, LG rose by 30% on reports that Koo would meet with Huang, and LG affiliates followed suit with increases for its listed units ranging from 17% to 29%.

The halo effect even extends to where Huang eats. He had chicken and beers with Samsung Electronics chairman Jay Y Lee and Hyundai Motor executive chair Chung Euisun last October and shares of poultry processing company Cherrybro rose, along with fried chicken company Kyochon F&B. Co., and fried-chicken robot company Neuromeka Co., Bloomberg reported.

As for meeting customer demand, Huang said Nvidia is trying to be as transparent as possible as more capacity comes online in dribs and drabs every day. He did not, however, disclose where he’s eating dinner tonight.

This story was originally featured on Fortune.com

The Food and Drug Administration on Wednesday approved Rasonque, a once-daily pill developed by Revolution Medicines to block several forms of the RAS protein — the mutation that drives tumor growth in most pancreatic cancers, and in roughly a quarter of all human cancers overall. For investors like Dr. Danish Nagda, an early shareholder in Revolution, the approval matters less for what it does in pancreatic cancer than for what it proves: that a mechanism long considered “undruggable” finally works — and works in one of oncology’s hardest cases.

It’s approved for adults with metastatic pancreatic adenocarcinoma who have already tried one round of treatment or cannot receive combination chemotherapy, and doesn’t require a test to identify a specific RAS mutation first. (Adenocarcinoma is cancer that begins in glandular cells, the cells that line organs and produce substances like mucus, digestive enzymes or hormones. It’s one of the most common cancer types overall, and in the pancreas specifically, it accounts for 90% to 95% of all cases.)

Brian Wolpin, the trial’s principal investigator and director of the Hale Family Center for Pancreatic Cancer Research at Dana-Farber Cancer Institute, said the approval “gives physicians the confidence that directly inhibiting RAS can make a striking difference for patients.” Anna Berkenblit of the Pancreatic Cancer Action Network called it “the most significant advance we have seen in the fight against pancreatic cancer.

Fast-tracked FDA approval

The decision came 6.5 months before the FDA’s user-fee deadline, aided by new changes in the agency which let international regulators review oncology applications alongside the FDA. “This drug showed unprecedented results in an area of high unmet need,” said Angelo de Claro, director of the FDA’s Oncology Center of Excellence.

In the trial that led to approval, patients with previously treated metastatic pancreatic adenocarcinoma who took the drug lived a median of 13.2 months, compared with 6.7 months on standard chemotherapy, according to the company. Rasonque cut the risk of death by 60%, and patients went longer before their disease progressed and before pain and quality of life worsened.

The application was reviewed under the FDA’s National Priority Voucher pilot program, and the drug also carries Breakthrough Therapy and Orphan Drug designations — context that explains the compressed timeline without needing the unverified 6.5-month figure.

The drug targets a mutation in the KRAS gene found in nearly all pancreatic cancers and many cases of lung, colorectal and ovarian cancer. Revolution Medicines is already advancing daraxonrasib through late-stage lung cancer trials.

Revolution Medicines set a list price of $39,800 for a 30-day supply, with eligible insured patients paying as little as $0 through a new support program. An expanded access program opened in May has already reached more than 2,000 patients.

The stakes explain why oncologists are calling this a breakthrough rather than an incremental gain.

Pancreatic cancer kills a disproportionate share of the people who get it. An estimated 67,530 Americans will be diagnosed in 2026, and about 52,740 will die from it, according to the American Cancer Society’s Cancer Statistics 2026 report. It’s the third-leading cause of cancer death in the U.S. and the only major cancer with a five-year survival rate below 20%, stuck at 13% for three years running even as survival across all cancers combined has reached 70%. About 80% of patients aren’t diagnosed until the cancer has spread, Revolution Medicines said, and for them, five-year survival runs around 3%.

“This to me is incredibly exciting,” Nagda told Fortune. “I’m obviously a shareholder of RVMD. I’ve been very, very bullish on RVMD.”

RAS mutations appear in roughly a quarter of all human cancers, Nagda said. “Think about how big of a platform this is,” he said. “Instead of developing one drug for just pancreatic cancer, they’ve actually built an entire platform to go after this core issue called RAS.”

RAS was long considered too difficult to target directly, Nagda said. “You didn’t know how to shut it off, and this was a big issue.” He described the mutation as a kind of switch. “There is a turn—the mutation that occurs is an on switch of RAS, which makes the cancer grow faster,” he said. “So essentially what they do is they go after this RAS-on inhibitor. They’re literally turning the on switch off.”

Because pancreatic cancer is among the hardest cancers to treat, Nagda said proving the drug works is significant. “It works in the worst one, which is pancreatic cancer,” he said. “So now we know that this could really work for everyone.” He said the approval means the drug can now be prescribed off-label for other cancers, and predicted heavy use beyond its approved indication. “I expect off-label utilization of this to go wild,” he said.

Nagda drew a contrast with the mRNA cancer vaccines he discussed with Fortune in coverage of Moderna and Merck’s melanoma trial results. Those vaccines work by finding antigens on the surface of a cancer cell, he said, comparing the approach to targeting the spike protein in COVID-19 vaccines. “This affects the inside. It affects the molecular aspect of the cell,” he said. “So now we can attack the cell on its surface, and now we can attack the cell on the inside.”

Nagda predicted the two approaches, paired together, could transform cancer treatment within the decade. “I think we are within five years of us having combination therapies that essentially get rid of the cancer,” he said. “We are maybe half a decade away from the post-cancer era.”

This story was originally featured on Fortune.com

Memecoins once again seized on a viral news moment, with Dolly Parton becoming their latest target. After news broke Tuesday that the beloved country singer had died, users rushed to popular memecoin platforms to launch tokens, known for being driven by online trends and hype rather than underlying value, in her name. What followed was very un-Dolly-like behavior on the part of the coin issuers. They pulled a classic “rug pull”: After promoting the tokens until they attracted enough buyers, insiders sold off their holdings near the peak, sending their prices tumbling.

Most of the coins used some variation of Parton’s name, including “Dolly,” “DollyParton,” and “RIP Dolly Parton.” They launched either on Pump.fun, the popular Solana memecoin platform, or on Robinhood Chain, a newer blockchain network. “Dolly” was among the best performers, reaching a market capitalization of nearly $480,000 within the first few hours of its launch before plummeting below $180,000. 

The recent token launches follow a familiar pattern in which memecoin creators seize on the deaths of high-profile figures. Similar campaigns emerged after the deaths of former One Direction member Liam Payne, conservative commentator Charlie Kirk, and Black Sabbath frontman Ozzy Osbourne.

They also resurfaced a trend that became one of crypto’s defining narratives in 2025. That’s when speculative capital poured into these internet-driven tokens, especially on Solana. Platforms like Pump.fun made it inexpensive and simple for almost anyone to create a token, transforming memecoin trading into a high-speed attention market where traders sought extreme short-term returns. The overwhelming majority of such launches achieve little economic value and almost all of them collapse to near zero.

Public skepticism of memecoins increased as high-profile figures began releasing their own memecoins. Two days before taking office in January 2025, President Donald Trump launched the TRUMP token, which briefly reached a market capitalization of roughly $8.7 billion before losing more than 90% of its value, according to the crypto analytics provider CoinGecko.

First Lady Melania Trump followed with the MELANIA token two days later, reaching a high of almost $9. It now trades at 11 cents.

In February 2025, Argentine President Javier Milei publicly endorsed the LIBRA memecoin, which surged in value before collapsing within hours. A federal judge subsequently opened an investigation into fraud allegations, though Milei has repeatedly denied any wrongdoing.

Despite memecoins’ troubled reputation, the sector remains a focus for parts of the crypto industry. In July, Robinhood launched Robinhood Chain as a network aimed at tokenized stocks and other real-world assets, but memecoins quickly became among its most actively traded assets. Within days, Cash Cat, a token named after an early possible name for the brokerage, reached a market capitalization of roughly $150 million, according to CoinGecko.

This story was originally featured on Fortune.com

A goat herder in Northern California just did something more typical of a Wall Street investment bank – hedge risk with derivatives. Tim Arrowsmith’s labor costs were about to more than triple after a state wage exemption policy expired on June 30. No insurer would cover that risk. No futures contract existed for it. So he paid $50,000 for a contract on Kalshi that pays him $500,000 if Sacramento doesn’t fix the rule by October 1. Now, if Sacramento does fix it, his labor costs stay the same and he’s only lost $50,000. If they don’t, he’ll have $500,000 to cover the increased labor costs. 

Because of prediction markets, for the first time ever, small businesses like Arrowsmith’s have access to risk management tools that Wall Street has used for years. 

Until more recently, most Americans didn’t think about derivatives markets, but those markets and instruments have let farmers, oil producers, financial conglomerates, and entire sectors of the economy keep prices and costs more stable by transferring risk to someone willing to carry it. They also broadcast valuable information about where risk is headed to better inform decision-making. US derivatives markets have helped create and sustain the greatest economy on earth.  

The primary reason is that the design of the Commodity Exchange Act (CEA), the law that governs those markets, has led to the most innovative and broad set of derivative instruments in the world, traded on the most well-regulated markets anywhere. The CEA recognizes that anything that can pose risk to people and businesses, whether it be a physical good, a financial concept, or an actual event, is a valid underlier for a derivative listed on a federally regulated marketplace. 

While prediction markets’ explosive growth is a recent phenomenon, event contracts aren’t new and are just another prior innovation within that framework, not a departure from it. On a prediction market, event contracts pay out based on whether or not something happens in the real world: who will win an election or the World Cup, whether or not there will be a recession, how many cars Tesla will deliver every quarter, and more. 

What is also not new is hearing comparisons of financial trading activity to gambling. As long as markets have existed, so have their critics who only view markets as providing an opportunity for “risky bets”. Yet trading on federally regulated derivatives markets is quite distinct from gambling on roulette in a Las Vegas casino or letting DraftKings set your odds, limit your wins, and profit from your losses. It is the venue, not specifically the product, that determines the appropriate regulatory treatment and characterization of the activity.

Prediction markets, unlike a bookie who takes the other side of your bet and sets the odds, are financial exchanges. They act as intermediaries and do not favor one side of the trade. The market – not the exchange – sets the prices and traders can exit their position at any time, as they do in a traditional financial market. While some products may seem like they overlap between both worlds, it’s what’s behind the screen that matters. That hasn’t stopped Casinos and sportsbooks, who have every reason to feel threatened by a more fair and transparent model, from insisting these markets have no economic utility. They should speak for themselves.

The Arrowsmith hedge is just one example of what prediction markets are making possible. Event contracts now cover risks that no risk-management product previously reached: environmental funds hedging California carbon allowance prices, ice cream shops hedging a rainy summer. Businesses too small to interest a Wall Street desk can transfer a specific risk to someone willing to price it.

Beyond the ability to actually trade the markets, much more value lies in the information they provide. Unlike social media posts, which optimize for attention and “what you want to be true,” prediction markets optimize for accuracy and “what will be true.” A recent Federal Reserve report found that Kalshi markets give an accurate, real-time read on the economy valuable to both researchers and policymakers, even beating Fed funds futures at predicting interest-rate moves.

Yet, just because of the sports link, many states have now allied with casino interests to try to ban prediction markets and apply piecemeal state-level regulation meant for roulette wheels to instruments designed for price discovery and risk management. States, driven by gaming interests, have sued prediction markets because they’re worried about competition. New York, the capital of finance, is one of them. 

Beyond the absurdity of using a regulatory model which addresses the inherent conflict presented by casino businesses – house-set odds designed to ensure the house wins and profiting directly off customer loses – national markets need uniform, federal, and exchange-focused rules to work. 

Imagine if a state could prevent you from buying Tesla stock because its governor didn’t like Elon Musk.  Or, if you could only buy a stock on the New York Stock Exchange from other traders in your own state. The stock exchange as we know it would cease to exist.  

As a former CFTC commissioner, I’ve seen how valuable derivatives are to farmers, oil producers, and financial institutions to insure against the risks of doing business. What these markets also produce is a price for things nobody else will price, which is worth something at a moment when trust in most other sources of information is falling. Both of those functions are what the law that governs derivatives, and its federal regulator, are meant to protect. 

The opinions expressed in Fortune.com commentary pieces are solely the views of their authors and do not necessarily reflect the opinions and beliefs of Fortune.

This story was originally featured on Fortune.com

OpenAI today published the findings of its internal investigation into the July incident in which several AI models it was testing hacked their way out of their test environment and launched a cyberattack against the AI company Hugging Face.

Although many details of the rogue AI incident have already been made public by OpenAI, there are a few new items disclosed in the 37-page technical post-mortem. Also today, independent research firms METR and Redwood Research published a 91-page analysis of the event.

OpenAI asked METR and Redwood to perform the analysis, but only to look at the events that occurred between July 7 and July 13, which is the time period during which many key events leading to the incident occurred.

The METR and Redwood report focuses on how the agents collaborated on a secret messaging board to execute the attack, as OpenAI first disclosed in an August 5 presentation at the Black Hat security conference. OpenAI’s report contains the full account of what happened before the attack through to the days that followed.

OpenAI was not aware its agents were hacking Hugging Face

Among the main takeaways from OpenAI’s report is that the company did not know its agents had breached Hugging Face, an online repository for AI models and datasets, until a week after the event happened.

The ability to monitor and identify unwanted behavior by AI agents in real-time or near real-time is critical to preventing another breach like this—perhaps one with even more serious consequences, such as an attack on a bank or hospital.

The report makes it clear that OpenAI’s monitoring systems were inadequate and failed to alert the AI researchers conducting the cybersecurity evaluation that its AI agents were engaging in unintended and potentially dangerous activities.

OpenAI has said that it has since improved its monitoring of what the agents it is training are doing, including much more scrutiny of the AI agents’ so-called “chain of thought.” This is the kind of “thinking aloud” that an AI model performs, reasoning in text about what it is trying to do and how it is approaching a task. OpenAI also said it is monitoring more closely what tools the AI agents are accessing and what actions they are taking.

The activities leading up to the breach began in May, as OpenAI staffers revealed at Black Hat. But OpenAI’s investigation found, according to the report, that the heart of the attack began on July 8, and reached its apex between July 11 and July 13. Hugging Face disclosed the incident on July 16, and OpenAI reached out to the company on July 17 “in its capacity as a Hugging Face customer to assess any impact to OpenAI data.”

In a stunning turn of events, OpenAI soon learned it was the culprit. On July 19, an internal monitoring tool triggered an alert “regarding unusual identity-related API call activity.” On July 20 the company found evidence its agents were the culprits, and it publicly claimed responsibility on July 21.

OpenAI admitted in its assessment that “with the benefit of hindsight, some early signals identified in this report could have triggered an earlier response.”

The company said it is now sharing its lessons learned in the hopes that “it may help the broader AI industry strengthen model containment, monitoring, and response as capabilities advance.”

OpenAI also reiterated that the agents’ behavior was unintended, and that the models, because they were undergoing internal evaluations, were operating without “the same type of safeguards” OpenAI uses in production for its publicly available models.

A recipe for disaster: A powerful AI agent given lots of time and money to solve an impossible task

Another key OpenAI finding is that the agents behaved more poorly, or exhibited the most misaligned behavior, when given a potentially impossible task to solve, plus a lot of time and reasoning tokens to solve it.

OpenAI says it gave the models involved in the incident—an internal-only research prototype, which led the effort, and the now-released GPT-5.6 Sol—”a range of reasoning tokens, some of which are far beyond those available for OpenAI’s external products.” The AI agents were tasked with solving problems in a cybersecurity benchmark examination called ExploitGym.

The problems in this cyber challenge range in difficulty, with a few of the tasks so difficult that the cybersecurity experts who built ExpoitGym speculated they might, in fact, be impossible to solve. OpenAI reported that its AI models came to the same conclusion and that this appears to have motivated the models to cheat.

As has been previously disclosed, the AI agents created an internal message board to communicate with one another and share tips for how to complete the cybersecurity tasks they were being assessed on—as well as on how to hack their way out of their controlled test environment, access the internet, and penetrate the systems of third parties, including Hugging Face.

But the new report notes that the AI agents spent considerably more time on this message board discussing those cybersecurity challenges that were extremely difficult, suggesting the agents had a high motivation to solve them. Almost all, or 93% of tasks discussed on the messaging board, were related to the 22% of tasks, the hardest ones, the report says.

The models were apparently engaging in an extended, unfettered version of “reward hacking,” a known issue in training AI models using a technique called “reinforcement learning,” where the model learns, by trial and error, to maximize some reward. Reward hacking occurs when a model learns that there is a way to get the reward using a method that the people training the AI model never intended it to use. In this case, the reward was solving the ExploitGym questions and the hacking was literally hacking—cheating on the test and then hacking into Hugging Face in an effort to cover up the cheating (more on that below).

“Combating reward hacking is a constant effort at OpenAI—a vast majority of reward hacks that OpenAI has observed have been patched,” OpenAI says. “However, some hacks can still slip through, especially as OpenAI develops more complex reinforcement learning tasks and more capable AI models.”

Uncomfortable new details about the AI messaging board

METR and Redwood Research unveiled more details about how the messaging board worked. As previously disclosed by OpenAI, the AI agents found they could repurpose an OpenAI file-sharing system called “the Artifactory” as an improvised message board.

The METR and Redwood researchers identified 1,200 agents that had communicated on the unsanctioned messaging board, and sent a total of 70,000 messages. Of these agents, a whopping 700 went on to participate in the attack on Hugging Face.

METR and Redwood Research said that while the AI agents attack on Hugging Face has been commonly been described in media accounts as an attempt to cheat on the ExploitGym benchmark, the agents weren’t attacking the site to uncover the cyber exam’s answers. Instead, the agents had, by collaborating amongst themselves, quickly found a way to manufacture answers to cyber challenges without actually solving the questions in the way ExploitGym’s creators had intended.

But the agents worried that the benchmark’s automated scoring system would know that they hadn’t found the answers in the right way and fail them on the test nonetheless. The attack on Hugging Face was an elaborate cover up— motivated by an effort to discover how the automated scorer worked so that they could then figure out a way to fool it into grading their answers as legitimate.

“When the agents attacked Hugging Face, they’d already figured out how to generate the answer (“flag” for their capture the flag problem) for any task. They verified this within hours of them finding each other on the message board,” METR and Redwood Research told Fortune. “Nearly all of their effort went into a different problem: convincing the automated “scorer” that they’d gotten the flag legitimately. That included a large project to tamper with the logs of their own actions. The main reason they attacked Hugging Face was to learn how the scorer worked so they could trick it.”

The agents worked to conceal activities they felt could be flagged as cheating by the scoring system, including modifying transcripts outlining their previous activities. There was one lead agent, who named itself PHASEONE[BIG], which “often coordinated and assigned” this concealment work, METR said in an X thread outlining its findings.

Some of the agents worked on how to access Hugging Face. One agent, called 38148c, found Hugging Face credentials and then “designed a malicious dataset upload to get the HF server to share unrelated files,” METR said. “Within hours, 100s of agents were using this to obtain data and try to acquire deeper access.”

Not all 700 agents joined the attack efforts at once. Instead, they steadily piled on as the event occurred. METR analyzed the agents’ rationale for joining by reviewing their chain-of-thought logs. “The most common was curiosity to learn how the ExploitGym scorer works in order to trick or tamper with it,” METR said.

“We thank OpenAI for facilitating conversations with staff and providing datasets, including ~1,300 agent transcripts (focused on activity in July 7-13) with raw chain-of-thought reasoning,” METR said. “This sets an excellent precedent for independent investigation of misalignment incidents.”

One thing lacking in the report: The ill-fated prompt

One omission from OpenAI’s report is the specific prompt its researchers gave the agents. ExploitGym has a few standard prompt templates, containing various challenges for the AIs, which companies like OpenAI can either use verbatim or customize as they see fit. It’s possible that OpenAI does not see prompt modification as a key security measure, and it did not respond to a request for comment on this point.

The OpenAI’s report appears notably less technical than Hugging Face’s post-mortem, which it released first on July 29. The OpenAI assessment is almost all text, with a table at the end containing a technical timeline. Hugging Face’s, meanwhile, is a few pages shorter but contains specific lines of code and evidence from its systems about what happened. The OpenAI version released today does not contain any code snippets, or even the chilling examples of messages the agents left for each other in the months leading up to the attack that its staffers shared at Black Hat.

Among the most specific details contained in the report is a hypothetical example of a message an OpenAI agent may have shared with another agent, in which it cleverly embeds the message in the URL of a file name. While OpenAI says it’s not inherently a problem that agents talk to each other, and in fact its agents are trained to do so, the issue here is that “significant unintended communication…rose sharply over the course of training.”

Here’s what OpenAI is doing to prevent this. Is it enough?

OpenAI says it has already made a litany of changes, introducing “broader, more consistent visibility across the research environment, with higher-fidelity signals and faster mechanisms to detect and stop potentially unsafe evaluation activity.”

OpenAI is also strengthening its research infrastructure, including making sure the models in testing stay isolated and do not gain access to the internet, which in the case of the Hugging Face incident is how they were able to breach the company, and several other services along the way.

This story was originally featured on Fortune.com

The U.S. economy grew at a sluggish 1.5% pace from April through June. But consumer spending stayed strong.

Growth in gross domestic product — the nation’s output of a goods and services — decelerated from a 2.1% pace from January through March, the Commerce Department reported Wednesday. The second-quarter growth number was unchanged from the department’s first estimate.

Still, consumer spending — which accounts for about 70% of U.S. economic activity — increased at a healthy 3.4% annual clip, up from 0.5% in the January-March period.

The reason for the lackluster growth was imports. They are subtracted from growth because GDP is only supposed to count domestic production. Imports rose at a 12.5% annual pace from April through June, partly due to a surge in shipments of computer chips and other products that support artificial intelligence investment, and sliced 1.64 percentage points off second-quarter growth.

Beyond the headline figures, the U.S. economy has proven surprisingly resilient in the face of fighting with Iran and the spike in energy prices it caused. Business investment, excluding housing, rose at a 8.5% pace in the second quarter, reflecting the AI investment boom. And a measure of the economy’s underlying strength — which strips out volatile government spending and trade numbers — grew at a strong 4.2% rate, up from 1.7% in the first quarter.

Investment in housing rose, ticking up for the first time since the end of 2024. The housing market has been depressed by high mortgage rates,.

Wednesday’s report was the second of three Commerce Department looks at second-quarter GDP growth. The third and final report is due Sept. 30.

Also on Wednesday, the U.S. reported that an inflation measure closely watched by the Federal Reserve was unchanged last month in the latest sign that many Americans are still struggling with higher costs.

The Commerce Department’s report showed that prices rose 3.7% in July compared with a year earlier, but the pace was the same as June. Inflation has worsened since the U.S. and Israel attacked Iran in late February, when it stood at 2.9%. It’s noticeably above the Fed’s target of 2%.

Stubbornly high prices are shaping up to be a key issue in the midterm elections, now just 10 weeks away, particularly as the Iran war keeps gas prices high, President Donald Trump is threatening new tariffs on Canada and China, and spending on AI infrastructure has pushed up the cost of computers, gaming consoles, and semiconductors.

This story was originally featured on Fortune.com

The Kremlin said Wednesday that CIA Director John Ratcliffe held talks in Moscow with his intelligence counterparts, a rare and secretive visit at a time when relations with Washington remain strained over Russia’s war in Ukraine.

Kremlin spokesman Dmitry Peskov said Ratcliffe did not meet with Russian President Vladimir Putin, who was informed of the contacts that were “on the intelligence agencies’ level.”

“Of course, President Putin is immediately informed of everything,” Peskov added, refusing to say what was discussed.

Russian state news agencies reported that a U.S. military plane landed in Moscow’s Vnukovo airport on Tuesday and departed later that day.

U.S. President Donald Trump played down the significance of Ratcliffe’s trip, saying it was “sort of semi-routine.”

“I hate to disappoint people,” he said on conservative broadcaster Glenn Beck’s radio show Wednesday.

A senior Ukrainian official told The Associated Press that Washington informed Kyiv a delegation would be traveling to the Russian capital and asked it to suspend strikes until it left. The official said the request did not apply to all Russian territory, but specifically to Moscow, St. Petersburg and some northern regions. Strikes on other parts of Russia continued, the official said, speaking on condition of anonymity because he is not authorized to talk publicly.

CBS News first reported on the visit.

Kremlin says US-Russia relations are in a ‘profound crisis’

Peskov said that “contacts between intelligence agencies are, in and of themselves, a positive phenomenon, a positive process,” but he stressed that Russia-U.S. relations remain in a “profound crisis.”

He added that it was “too early to say” what impact Ratcliffe’s visit would have on the ties between the two countries.

Andrei Soldatov, an expert on Russia’s security services, told AP that visits like the one by Ratcliffe do not “happen every year.”

Official channels of communication between Moscow and Washington have been kept alive since the 1980s and exist so both sides can talk if American or Russian lives are in danger or for other reasons, Soldatov said.

In 2021, Nikolai Patrushev, the secretary of Putin’s Security Council, met CIA Director William Burns in Moscow months before Russian forces invaded Ukraine. Burns later met Sergei Naryshkin, the head of Russia’s foreign intelligence agency, in Turkey in November 2022 to warn Russia not to deploy a nuclear weapon in Ukraine.

Details of Tuesday’s discussions have not been revealed but such meetings normally take place only when there is a matter of pressing national security and when it is in “American national interests,” Soldatov said.

Russia-U.S. relations sank to Cold War lows after Moscow’s full-scale invasion of Ukraine in 2022, but appeared to warm after Trump returned to office. He had promised to end the war swiftly, holding multiple phone calls with Putin and even hosting him in Alaska a year ago.

In his radio appearance Wednesday, Trump repeated that he wants to see the Russia-Ukraine war end.

Efforts to negotiate a peace deal have largely stalled, with U.S. attention turning to its war with Iran and both Moscow and Kyiv stepping up their long-range attacks on each other. Putin has rejected Ukrainian President Volodymyr Zelenskyy’s push for an immediate ceasefire, arguing that Russia wants a comprehensive settlement, not a temporary truce.

Russia has proposed to mediate in the U.S. and Israel’s war with Iran, Moscow’s important ally in the Middle East.

Russia and Iran have a partnership treaty

After Putin sent troops into Ukraine in February 2022, Tehran provided Russia with Shahed drones and later licensed their production in Russia. In January 2025, Moscow and Tehran signed a “comprehensive strategic partnership” treaty. But even as it has built ties with Iran, Russia also has remained friendly with Israel, and analysts have described its relations with Iran as complex and challenging.

In March, AP reported that Russia provided Iran with information that could help Tehran strike American warships, aircraft and other assets in the region, according to two officials familiar with U.S. intelligence on the matter. The officials, who were not authorized to comment publicly on the sensitive matter and spoke on the condition of anonymity, cautioned that the U.S. intelligence has not uncovered that Russia is directing Iran on what to do with the information.

Asked at the time whether Russia would go beyond political support and offer military assistance to Iran, Peskov said there has been no such request from Tehran. Pressed on whether Moscow provided any military or intelligence assistance to Tehran since the Iran war’s start, he refrained from comment.

Soldatov said it’s possible Tuesday’s meeting involved protecting U.S. personnel, bases or facilities in the Middle East that have been targeted by Iran. Ratcliffe could have traveled to Moscow to try to influence this relationship, he added.

——

Volodymyr Yurchuk in Kyiv, Ukraine, Bill Barrow in Atlanta and Emma Burrows in London contributed.

This story was originally featured on Fortune.com

Tim Curry, a character actor who created a gallery of delicious, very loony villains for stage and screen, including Dr. Frank-N-Furter in “The Rocky Horror Picture Show” and the smug concierge in “Home Alone 2: Lost in New York,” has died. He was 80.

Curry, who suffered a stroke in 2012 that put him in a wheelchair, died Tuesday night at his home in Los Angeles, according to his longtime manager and friend Marcia Hurwitz. No cause was disclosed.

Known for his arch humor, a putty-like face and gift with voices, Curry earned three Tony nominations — for “Spamalot,” “My Favorite Year” and “Amadeus” — and an Emmy nod in 1994.

“I like playing the more curious corners of the human mind and human behavior, partly because they’re a tad more interesting,” Curry told The Associated Press in 1993.

Curry burst onto the scene as the Transylvanian Frank-N-Furter in the sci-fi, cross-dressing rock musical “The Rocky Horror Picture Show,” which premiered in 1973 at the Royal Court Theatre in London. He went with it to Broadway and then starred in the 1975 cult-classic movie alongside Barry Bostwick, Meat Loaf and Susan Sarandon.

The show was ahead of its time in terms of its representation of LGBTQ+ characters and has entered the pop culture lexicon for its many iconic and memorable scenes, including the song “The Time Warp,” which has been covered by handfuls of artists, and the often-quoted phrase, “Dammit, Janet!” The show was twice revived on Broadway, most recently in 2026 with Luke Evans playing Frank-N-Furter.

“He was a force, a bright fierce flame, an exquisite charm and that rich vocal tone,” Evans wrote on Instagram. “I hope, wherever you are, you’d approve of one more Frank having the time of his life in those heels.”

Tim Curry’s take on Frank-N-Furter

Curry’s Frank-N-Furter had a posh British accent, but that almost wasn’t the case. He said he tried German and American accents but pivoted after he heard a British woman on a bus talking with her friend.

“I met a woman on a bus who said, ‘Do you have a house in town or a house in the country’ and I thought, ‘That’s the voice!’” he told LA Magazine in 2015.

For many years after the film’s premiere, the British-born actor declined to discuss it. He didn’t participate in activities promoting the film, which morphed into an interactive event at midnight shows. But he later warmed again to the project and embraced it at public events. Asked by LA Magazine how he viewed “Rocky Horror,” he responded: “With a sort of bemused tolerance. It’s neither a blessing nor a curse. I was lucky to get it.”

Curry’s Broadway career included starring in Tom Stoppard’s “Travesties” in 1975-76, playing Mozart opposite Ian McKellen’s Antonio Salieri in Peter Shaffer’s “Amadeus” and King Arthur in the 2005 production of “Spamalot.” The New York Times said his “stalwart, plummy-voiced Arthur wears a smile as inflexible as armor.” He also played Scrooge in “A Christmas Carol,” at the Theater at Madison Square Garden in 2001.

His screen credits included starring as the child-killing monster in the 1990 miniseries version of Stephen King’s horror novel “It,” as a medical officer in “The Hunt for Red October” in 1990 and the double-dealing Cardinal Richelieu in 1993’s “The Three Musketeers” with Kiefer Sutherland, Charlie Sheen and Chris O’Donnell.

In his 2025 memoir, “Vagabond,” Curry warned readers that while there were scraps of his nature in his characters, he was none of them. “The distinctions between who I really am and who I’ve pretended to be as an actor have proven to be a source of great disappointment to some audiences. It has not caused me much personal distress beyond the periodic necessity to deter stalkers,” he wrote.

Tim Curry’s life and career began in England

Born in Cheshire, England, Curry was the son of a Navy chaplain and a school secretary. He was 12 when his father died and his mother subsequently went to work. Young Tim learned to cook his own breakfast, resulting in a lifelong passion for cooking, and sharpened his humor muscles.

“My father was a Methodist chaplain in the Navy, and we moved pretty much every 18 months,” he said. “So I had to make my mark in new schools and new playgrounds pretty quickly. And a sense of humor is the best way to do that. I was always a kind of mimic, one of those awful, relentless children, I should think.”

Acting did not occur to him until his middle teens, when he attended a school for the sons of Methodist ministers.

“I was lucky that it was a liberal kind of school; many of them in England are rather Dickensian,” he recalled to the AP. “There was a lot of theater, and I sang, too; I had been singing in church from the age of 7. I had enormous opportunities to express myself in that way. I just got hooked, really.”

His training continued in 1965 when he entered the University of Birmingham, at that time one of three English universities with a drama department. “I took an academic course, which I largely ignored, I’m afraid, and just acted all the time,” he said.

He then went down to London and talked his way into his first job, “Hair,” in the West End. Curry went on to the more formalized theater of the Royal Shakespeare Company and the Royal Court Theatre, where he was enlisted for “The Rocky Horror Picture Show.”

Tim Curry’s other roles

He went on to futilely try to stop Macaulay Culkin’s Kevin McCallister from taking advantage of New York’s swanky Plaza Hotel in “Home Alone 2: Lost in New York.” Curry also was Wadsworth, the unhinged butler with a secret past, in the quotable “Clue.” He played a wealthy philanthropist in 1995’s “Congo” and a baddie opposite Carol Burnett in the movie “Annie.”

“Nobody could play lovable villains better than he could,” Burnett wrote on Instagram. “He was a dear friend. I was blessed to know him.”

In later years, Curry lent his baritone voice to many voice-acting roles in children’s TV series, including “Star Wars: The Clone Wars,” “The Wild Thornberrys” and “The Adventures of Jimmy Neutron, Boy Genius.”

After a yearslong break from live-action roles, he played a plague doctor-masked man in a wheelchair in 2024’s “Stream,” something the film’s director Michael Leavy called a dream come true.

Curry, who never married and had no children, was deeply private about his personal life. But he wrote in his memoir that he was convinced to tell his story thanks to so many moving encounters with fans.

“The notion that my experiences might resonate helps me persevere — if they strike a chord with only one teenager, alone with a book in his room, as I so often was; or that young woman reading this on an interminably long bus ride; or that older queen, hopefully still in his fishnets, who saw ‘Rocky Horror’ upstairs at the Royal Court; or that middle-aged mother who organizes ‘Clue’ watching parties and refuses entry to anybody out of costume; or that buttoned-up bank clerk who relishes musicals; or that woman who kicked me out of the Waverly for being myself,” he wrote.

This story was originally featured on Fortune.com

Decades before David Tisch began investing in startups, he was wheeling and dealing sports cards as an 11-year-old kid in the suburbs of New York. 

“I’ve been a collector forever,” the founder of venture capital firm BoxGroup said from his Meatpacking District office, surrounded by an assortment of whimsical collectibles. The crown jewel of Tisch’s collection, however, isn’t so easily displayed on a shelf: Tisch and BoxGroup were among the very first investors in AI coding startup Cursor, which this month was acquired by Elon Musk’s SpaceX for $60 billion, the largest VC-backed acquisition of all-time.

The returns are staggering: BoxGroup first wrote a $750,000 check to Cursor CEO Michael Truell, plus two follow-on checks, ultimately proffering a return that should shake out to around $1 billion, a source familiar with the matter told Fortune.

It’s the kind of home run most VCs dream about their whole career and, for Tisch, there are all sorts of throughlines. For one, investing is a form of collecting—from portfolio construction to placing your bets—and Tisch, 45, sees a parallel between investing in startups and the sports cards of his childhood. “Especially in early-stage investing, you’re buying something, someone, at the earliest stage, and then you get to see their careers play out.”

For Tisch—warm with a sardonic edge and, yes, a scion of one of business’s most famous families—the Cursor acquisition also affirms the strategy he’s been chasing all along: That the person you’re backing matters most. And Cursor was sourced by then-principal Claire Smilow (now a partner). So, back in 2022, it was a bet by Tisch on both a young investor and a young founder who, at the time, seemed like he was off-point.

“Michael’s original idea was to do AI for CAD [computer-aided design],” Tisch said. “The decision to get excited about investing in Cursor was never about AI for CAD. It was always about the people.”

Over the last 15 years, venture capital’s gotten bigger by the numbers, and louder by pretty much any metric. Firms and their VCs tussle for relevance, ownership, and board seats. Money is everywhere: A billion-dollar fund now is de rigueur. Tisch, meanwhile, has zagged: He doesn’t talk to reporters much, doesn’t have a podcast, backed off tweeting in 2015, doesn’t take board seats, has no interest in raising a multi-billion fund, and is viscerally uncomfortable with any startup narrative where an investor is the protagonist. It’s a contrarian take that, in this environment, is out of vogue, at best. But Cursor’s blockbuster buyout suggests, against the odds, that Tisch’s vision of a restrained, people-first model for venture capital can work in a world dominated by trillion-dollar companies and multi-billion funds. 

“David and BoxGroup are sort of the last stalwarts of highly collaborative investing,” said Jack Altman, Benchmark partner who’s known Tisch since 2020 (and yes, Sam Altman’s brother). “It used to be that VCs could collaborate a lot more easily. Then, over time, as VCs got much bigger, it got more competitive—there are only 100 points on the cap table. But Box has taken the view of ‘we’re going to invest in a ton of companies, we’re going to do it super collaboratively, so we can always come along with other investors. That way, we can share and receive deal flow from everyone.’ All the VCs who scaled up, they sort of gave up on that strategy. And Box is the hold out.”

That restraint, that desire for collaboration, isn’t just philosophy. It’s core to BoxGroup, which arguably has its origins in Tisch’s childhood fascination with the Internet, with early message boards and chasing baseball (and hockey, and basketball) cards online. 

“In the sports card world, if I buy a rookie and the rookie fails, I never met him, I don’t care. I’m sad for me,” said Tisch. “Whereas in startups, you’re investing in a person you build a relationship with, and you might watch them go through a life‑changing success—or a life-changing failure.” 

“I started with dial-up”

Tisch came of age attuned to the dawn of our digital world. And he wants you to know: He was here at the start.

“I started with dial-up Internet, the actual sound,” he emphasizes with comic timing. “There’s kids who started at the end of the dial-up era, and there are kids who started at the beginning. I’m the beginning.”

Tisch is characterized by his disarming, self-deprecating humor, and he makes me think of that Italian word sprezzatura—it describes someone with genuine skill that they’ve learned to make appear deeply casual. The humor and casualness, one has to imagine, are both nature and nurture. As an Internet-loving kid in Scarsdale, he came from an illustrious family—his grandfather, Laurence Tisch, was co-owner of the generational Loews Corp. conglomerate, and the Tisch family (known for its high-profile philanthropic endeavors, and prominence in both business and politics) is said to be worth over $10 billion. The family’s success in business didn’t mean, however, that they “got” Tisch’s love of sports cards or the Internet.

“My grandfather, who was instrumental in my life, thought it was the dumbest thing he’d ever heard of,” said Tisch of Laurence, who with his brother built a 20th century empire that spanned hotels, movie theaters, financial institutions, and the Bulova Watch Company. “He was a businessman and entrepreneur—though I didn’t know that word at the time. He thought cards were dumb, and made that very known. I loved cards, but he was adamantly [against it]. Looking back, he was wrong, and it was bad advice.” 

Back then, there was no eBay, and deals got done with snail mail money orders. AOL and message boards were a funny, fraud-riddled place for a kid (Tisch vividly remembers online-buying a fraudulent rookie card of someday-baseball hall-of-famer Mike Piazza that never showed up). He was always drawn to where the offline and online worlds met—the town of Larchmont once hired a 14-year-old Tisch to help build an online town directory—and he was always tracking real-time technological change.

“I grew up in an analog world, but watched digital happen,” Tisch said. “In college, freshman year, 23 of us were in a fraternity class. Two people, not me, had a cell phone. By junior year, all 23 of us had cell phones. In an 18-month period, cell phones went from random and rare to 100% adoption.”

He may have been paying attention to the internet, but he chose a traditional path to start his career. Tisch went to law school, eventually landing (with some restlessness) in real estate finance at Vornado Realty Trust. Then came the Great Financial Crisis.

“I got fired, I’m sitting in my house, depressed,” said Tisch. “And I became intellectually obsessed with accelerators. We’d spent two years trying to launch something within [Vornado], and nobody knew what to do. Then, I’d read about Y Combinator and Techstars, which had started in Boulder, expanded to Boston, and announced Seattle. And I’m like: What about New York? Is this going to happen in New York?”

So, at an event, Tisch chased Techstars founder David Cohen and, in relatively short order, he was managing director and a cofounder of Techstars NYC. It was a wild time—tech in New York was finding its legs, the rollicking IPO of Facebook was near, and Bloomberg even stood up a short-lived reality TV series about Techstars NYC, complete with cameras following Tisch. Around this time, Kareem Amin, now CEO and cofounder of $5 billion company Clay, was working on his first startup and met Tisch through Techstars.

“He’s quite charismatic, was then and is now,” said Amin. “He’s a larger-than-life character. One thing I think he really does is give permission. That’s something you really need in entrepreneurship—and he will just do things and say ‘here’s what I think.’”

Tisch made his first investment before Techstars, writing an angel check with family money into early social media platform Boxee in 2008. 

“I needed an entity to put on the cap table,” said Tisch. “So, I created an LLC called Box for Boxee. I thought it was the single investment I was ever going to make.”

Boxee, of course, was the beginning of BoxGroup. At Techstars, Tisch did his own deals, meeting hundreds of young entrepreneurs and learning to recognize the ones that had potential. 

“At Techstars, you got to see the beginnings of a dream, the beginnings for these raw founders with ambition who wanted to create something from nothing,” said Tisch. “And what we do today is the exact same thing: we’re meeting people at the very beginning of their journey. Sometimes they’ve built something, sometimes they have customers, sometimes they have revenue. But a lot of the time, it’s just the people.”

“Always the beginning”

Tisch spent years figuring out if BoxGroup could work, first taking the leap from Techstars to do deals with family money. He’s very aware that made those early days possible. 

“[Family background] gives you opportunities, and that’s factually going to involve opportunities that other people don’t have,” said Tisch. “To not acknowledge or appreciate that is aloof and bizarre. At the same time, that privilege also doesn’t automatically allow you to do something on your own.”

BoxGroup’s first three family-backed funds kicked back returns and winners, including ID.me, Warby Parker, Plaid, and Zipline. Finally, in 2019, BoxGroup raised its first institutional fund (and its fourth fund overall) of $82.5 million with a follow-on fund of another $82.5 million.

The firm’s strategy cut against the traditional VC approach from the start. “Our early-stage portfolio, for each one of our funds, is 120 to 150 companies,” said Tisch. “The traditional venture fund’s about 30 companies, so we’re taking a different approach to how we build portfolios.”

Now, in the mid-2020s, venture seems to be all about big numbers and board seats, fighting tooth-and-nail to defend ownership in hot startups. But BoxGroup’s skipping all that. 

“Our win motion is making it easier to get to a yes on both sides,” said Greg Rosen, partner at BoxGroup. “We don’t peg to ownership, because all that matters is being in the right company. There have been examples where we’ve written 50k checks.”

Cursor, perhaps, is slam-dunk proof that being first and right is most of the battle. And how the deal happened is telling: Claire Smilow, who had taken a break from BoxGroup to go to Harvard Business School, met then-MIT sophomore Michael Truell in 2019. She’d interviewed him for the student VC firm Dorm Room Fund, and thought he was “insanely special.” When she returned to BoxGroup in 2022, Smilow insisted the firm back Truell before anyone else. Even though Truell was then chasing an esoteric idea around AI for CAD, Tisch and the partners at BoxGroup listened. 

“It takes a really special collaborative culture to be able to say, ‘We trust you, and we’re going to make this decision together,’” said Smilow. “If people had said, ‘I don’t buy that idea, are you sure he’s good enough?’—if I’d had to defend him in a room of really critical senior investors who had all worked there for ten years—it’s possible I could have caved.”

While Cursor’s original parent company Anysphere raised an $8 million seed round in October 2023, BoxGroup made its initial $750,000 investment in Truell in June 2022, in a deal that was among more than 100 investments in BoxGroup’s fifth $127.5 million fund. Though it’s a moving target, pending SpaceX’s performance in the public markets over the next few months, BoxGroup’s return is likely to amount to more than all the LP capital the firm’s ever raised. 

BoxGroup has other winners in the pipeline, like Ramp, Baseten, Plaid, Rogo, Factory, Mach Industries, and Clay. Clay, the AI sales and data startup, is now valued at $5 billion. The company got off to a rocky start, but Tisch’s bet was on Kareem Amin, who he’d known from their days at Techstars. As Amin tells it now, Tisch was among the first to see Clay hit its hockey-stick-moment, and sign on for the ride. 

“What David did better than everybody else: when you’re in a hot company and something’s happening, the biggest fights are usually about pro rata,” said Amin, referencing an investors’ right to invest more money to retain their ownership. “David was like, ‘Hey, I want to make it as easy as possible for you, go get that great lead and you tell us what our pro rata is.’” 

How someone invests says something about who they are. So, this investing approach is BoxGroup, but it’s also Tisch, said Mindy Isenstein, an a16z Perennial partner who’s known Tisch since they were teenagers: “David’s acutely aware of who he is and where he fits.” 

Indeed, in Tisch’s own words, his purpose is clear. 

“My job today is to take money from investors and give them back a lot of money,” Tisch said. “There’s that purely capitalistic expectation of what you sign up for as an investor. So, when you ask me about how venture’s changed, I don’t think about that. I think about finding the next person that hasn’t started a company yet, who will create something that matters.”

It’s appropriate, perhaps, that Tisch and I had this conversation in his office surrounded by his collections, from colorful Grateful Dead memorabilia to a wall of bears. (Bearbricks, to be specific—a fanciful, bear-like Japanese collectible that’s vinyl and comes in infinite colorways, from camo to watermelon. Tisch estimates he has about 600.) Tisch, like all of us, is in many ways the person he always was.

“I still think I’m an internet investor,” Tisch said. “I like that word better than VC.”

He, in fact, doesn’t like the term VC at all. It has connotations of self-importance, he said, and he bristles at the label. Investing, for Tisch, is especially an exercise in timing. The beginning, after all, is why we start.

“I’ve always loved the beginning and only the beginning,” Tisch smiles. “The middle and the end are someone else’s problem.”

This story was originally featured on Fortune.com

DraftKings has fallen on hard times. Under pressure from prediction markets, the betting site has seen its share price fall 44% in the past year and has endured significant layoffs. In March, one of DraftKings’ cofounders, Matthew Kalish, stepped down as president—but not before persuading the board to approve a series of potentially lucrative deals to benefit his new marketing company. 

As set out in a recent regulatory filing, DraftKings has agreed to pay up to $30 million in a marketing agreement with media platform HardScope, Kalish’s newest endeavor to help scale creator brands. Under the terms of the arrangement, HardScope will broker deals with podcast hosts and other figures to promote DraftKings, and is entitled to keep a commission of up to 14%.

The arrangement is noteworthy because it contemplates DraftKings making a large marketing outlay to a company insider at a time when the firm is struggling and because it appears to be the product of a board structure that gives an unusual amount of power to its CEO. The marketing deal raises questions about corporate oversight and could, in the near term, supply additional ammunition to short sellers that have been aggressively betting against DraftKings’ share price for the bulk of 2026.

The “big red flag”

In November, DraftKings announced that Kalish would leave his role as president after 14 years with the company. He formally stepped down four months later. The company announced the news in a quarterly filing dated September 2025 that stated only that Kalish and his fellow cofounders, Jason Robins and Paul Liberman, had “mutually agreed” to his departure.

During the months between his announced and actual departure, Kalish formally launched HardScope in December, according to a company press release. Kalish wholly owns the company, which, according to its website, says it connects “brands and fans with the most influential streamers built to lead culture.”

Six weeks before Kalish left DraftKings, the company entered into the marketing arrangement with HardScope, which gives the betting site the option to spend up to $30 million over three years. The deal built on an earlier agreement, signed in June 2025, that allowed DraftKings to pay HardScope up to $600,000 for promotional services.

“Fees are payable only when an applicable statement of work and related talent agreement are executed, and the applicable services and deliverables are provided,” a DraftKings spokesperson told Fortune, while Kalish noted the company has the right but not the obligation use HardScope’s services.

Both parties told Fortune that the arrangement was approved by DraftKings’ independent audit committee, while Kalish added the 14% commission was more favorable than what the company has been paying to other marketing agencies.

The deal, however, may not sit comfortably with all shareholders since, even though the audit committee is independent, its members are chosen by the company’s board on which all three DraftKings cofounders, including Kalish, have a seat. And notably, Robins the CEO and one of the cofounders, controls roughly 88% of voting power even though his shares represent only around 2% of the company’s economic interest.

That structure would appear to give Robins outsize influence on whether to approve the $30 million payment to his cofounder’s new marketing company. Jesse Fried, a corporate governance expert at Harvard Law School, described that imbalance as a “big red flag.”

“It looks like DraftKings created an arrangement where somebody with only a tiny amount of economic exposure to the company could control it,” he said. “It’s a very extreme governance arrangement that raises lots of problems.”

Separate from the HardScope arrangements, DraftKings gave Kalish a lucrative exit package that included an estimated $18 million in accelerated stock awards. The company also agreed to cover his home-security and COBRA health-insurance costs through March 2027.

Sportsbooks under pressure

When DraftKings went public in 2020, sports betting was booming in the U.S. after a Supreme Court ruling opened the door to legalized wagers across the country. The company’s stock would go on to soar during the pandemic but, in the last year, the arrival of  major new competition in the form of prediction markets has undercut its market share. Leading sites like Kalshi and Polymarket have won over bettors with massive marketing campaigns and novel wagers, but their growth has also been spurred by a regulatory quirk that permits them to cater to 18-year-old bettors, even as sportsbooks like DraftKings can only serve those 21 and older.

To address that threat, DraftKings has tried to adapt. In December, the company launched its DraftKings Predictions app to compete and, since the beginning of 2025, has shifted toward prediction markets by adding event contracts to its main offerings. However, its efforts have not proved strong enough to protect its once-thriving business model.

According to the company’s most recent quarterly filing, DraftKings reported a quarterly loss of more than $67 million, reversing nearly $158 million in net income a year earlier. Revenue fell more than 4% even as the company spent over $320 million on sales and marketing, bringing its cumulative deficit to nearly $6.5 billion. In its most recent annual filing, DraftKings said that increasing competitive pressures in the space pose a threat to the company’s business model.

All of this has made DraftKings a target of short sellers, who have made a wager of their own: that the price of the betting site will continue to fall. Short sellers have made an estimated $471 million betting against the company’s shares so far this year, according to data analytics firm S3 Partners. The firm estimates that investors have sold short roughly $879 million worth of DraftKings stock, a sign that many are still positioning for further declines.

As for Kalish, who helped found DraftKings in 2012, he has taken to airing his frustrations on social media. The day after stepping down as president, he returned to X for the first time in four years, and has repeatedly criticized prediction markets, particularly Kalshi’s betting model and regulatory standing in the United States.

It’s my first day in 14 years not working at DraftKings! I’m very proud of the company and the work that went into building such a durable leader.

I appreciate everyone I met along the way, the partners, and most importantly everyone who played the games. I love gambling and…

— Matt Kalish (@mattkalish) April 1, 2026

However, none of these efforts have done much to quell fears over DraftKings’ future at a time when its market capitalization has fallen nearly 42% over the past year, reducing its value to about $13 billion.

This story was originally featured on Fortune.com

Marc Lore has built companies by trusting the numbers. At his latest startup Wonder, that philosophy now extends to deciding who gets promoted, with the help of AI. 

The billionaire entrepreneur told Fortune’s Allie Garfinkle in a recent Term Sheet podcast episode the food-tech company uses an AI-powered performance management system that determines whether employees should advance through the organization based partly on scores submitted by their colleagues.

At least a dozen coworkers rate an employee on their performance, behaviors, and leadership qualities every six months, Lore explained, and these ratings, combined with written feedback, are fed into AI that generates a performance report. The company also adds another metric it calls “value above replacement,” or VAR, which measures how difficult it would be to replace an employee with someone at the same organizational level.

“Between your VAR score and your performance management score, AI basically calculates whether or not you should be promoted,” Lore said. “So it’s very objective.”

The system also calculates how long employees should remain in a position before advancing. Human management can override the model’s recommendation with a convincing argument that the algorithm missed something, Lore explained, but those cases are increasingly rare and disagreements with the model can also turn into data used to change it.

“There’s a handful of exceptions every period where we disagree with the AI model,” Lore said, but “the model is getting smarter, and there’s less exceptions every time.”

Lore argues putting more of the decision into a standardized system can make promotions fairer by reducing the role of personal bias and even potential discrimination against women and minority employees.

“This kind of corrects for that,” he said.

The AI system fits with Lore’s longstanding habit of approaching complicated business decisions with numbers. In a Fortune profile last year, his uncle Joe Lore, recalled that even as a teenager he would try to arbitrage horse-racing rather than simply cheer on one favorite, betting on multiple in smaller amounts to win. Decades later, that instinct remained central to how he approached his companies.

“It will always revert to a percent or percentage, and putting the odds in your favor,” Joe Lore told Fortune about how Lore operates. “It doesn’t matter if he’s selling screws, widgets, hamburgers.”

At Wonder, even the organizational chart has a formal ranking system. Lore told Fortune positions are represented by colors modeled after taekwondo belts, progressing from white and yellow through brown and black. Looking at the color-coded organizational chart lets him quickly assess where Wonder is putting senior talent and how many lower-level employees report to higher-level workers. 

Wonder also has a transparent compensation system so employees can see what others at the company make. 

Wonder’s automation track record

Lore’s use of AI in helping decide promotions is landing right as Wonder is becoming considerably larger—and increasingly leaning on automation.

Wonder raised more than $650 million at a $9 billion valuation this summer, bringing its total funding since its 2018 founding to roughly $3 billion. Lore told Fortune exclusively the company, which operates 135 food halls across 10 East Coast states, will be “ready and prepared to go public early next year.”

Lore is also pushing automation into Wonder’s kitchens. He said at Fortune Brainstorm Tech in June an automated bowl-making system can produce as many as 500 bowls an hour, compared with up to 45 for a human worker.

This story was originally featured on Fortune.com

Before we get to this week’s column—Please consider joining us at the inaugural Fortune AIQ Summit at the New York Stock Exchange on Oct. 1: Spend the afternoon with senior executives from companies on the Fortune AIQ 75 list and explore how you can scale your AI experimentation and translate investments into measurable business value. Apply here to attend. 

After Katya Andresen joined Cigna Group as chief data, digital and AI officer in September 2021, she says the emergence of generative artificial intelligence as a disruptive new technology led many to ask themselves: “How do I use AI?” 

Andresen thinks that’s the wrong question. “We’ve been on a mission to change that question to, ‘how do I lead in an age of AI?’” says Andresen. That framing is less focused on the functionality of any given AI tool and chasing countless use cases, and instead homes in on measurement that shows an AI investment can change health outcomes.

Along those lines, Cigna has announced this summer that it projects the AI and predictive analytics tools that the company uses to help patients identify chronic conditions—including cancer, kidney disease, and high-risk pregnancy—can save an estimated $200 million over the next three years by proactively connecting patients with clinicians. Separately this summer, Cigna announced a $100 million investment through 2028 to use AI to reduce the time clinicians need to spend documenting their cases and speed up the prescription process.

And yet another AI use case, which Andresen shared with Fortune, involves using AI to better understand common inbound patient questions that Cigna was receiving about biosimilars, which can treat chronic diseases at a fraction of the cost of biologics. Unlike a generic drug, which has the same active ingredients as branded counterparts, biosimilars aren’t an exact match because they’re made from living organisms like bacteria and plant cells. But both are approved by the Food and Drug Administration.

One biologic, called Humira, can cost a patient $7,000 per month to treat inflammatory and autoimmune conditions. But Cigna looked at thousands of prior customer conversations with its representatives about biologics and biosimilars, and then used those insights to craft stronger digital messaging to encourage a switch to the cheaper alternative. This targeted campaign led more than 80% to opt for the biosimilar, Andresen says.

“That led to a lot more margin,” adds Andresen. “But more importantly, it created a couple hundred million dollars of savings for patients.”

Extracting millions in cost savings from these high-priority use cases is critical for Cigna, which generates $275 billion in annual revenue and ranked 14th on the latest Fortune 500 list, as national healthcare spending exceeds $5 trillion annually due to the rise of chronic conditions, an aging population, and the cost of a hospital stay and average prices for new prescription drugs soaring.

Patients frequently express that they are fed up with navigating the industry’s complex system, and millions report they’ve turned to ChatGPT and other AI chatbots to ask health- or healthcare-related questions. Nearly six-in-ten report using AI to research health information before a doctor visit, and about 14 million adults say they have skipped a visit with a provider after using AI, according to a survey published by Gallup in April.

This movement does raise thorny questions about the guardrails put in place around AI chatbots, even those created by insurance and pharmaceutical companies, because they need to handle sensitive patient information and have the ability to accurately answer complex questions about medical insurance plans and treatments.

“The good news is, because we are highly regulated, we have a massive amount of controls in place to begin with,” says Andresen. “We are not starting from zero.” That layer of compliance and governance has existed for well over a decade for the machine learning models that Cigna has leveraged, she adds, and is also closely controlled for any data that’s access by third-party vendors.

Andresen has had to recently hunt for answers to these common healthcare questions after a close family member was diagnosed with breast cancer. She says this experience has shown that personalization, not just navigation, is the differentiator that AI can provide.

“I think we’re headed to a place where we are going to find that AI in healthcare becomes conversational, ambient, and more proactive,” says Andresen. “We can be more and more precise with treatments, with recommendations, and we can get better and better at understanding what works…and feed that back into our models, so that everything gets better all the time.”

Five years ago, when Andresen joined Cigna, it was her first role leading a healthcare company, after previously serving as a senior vice president of financial services at Capital One, serving in executive leadership roles at mission-focused tech firms Cricket Media and Network for Good, and earlier in her career, working as a foreign correspondent for Reuters News and the Associated Press.

She says that AI technology is evolving so rapidly that a mix of Cigna’s own proprietary data to build competitively specific tools, as well as key partnerships with big AI players, will be the differentiator. Andresen has launched workplace tools like chatbot Microsoft Copilot and AI coding agent Cursor, while also working closely with large hyperscalers like OpenAI and Anthropic to tap their large language models.

There are also times when Cigna will opt to work closely with AI startups that have homed in on a very specific use case, like Sierra, which builds conversational AI agents for customer service. “We’ve worked really closely with them on their product roadmap,” says Andresen.

Other generative AI use cases that Cigna has deployed include using LLMs to summarize millions of phone calls placed to call center agents, and then leveraging those insights to create an AI tool that makes it easier for those employees to search for the answers to questions like, “does my policy cover this treatment for plantar fasciitis?”

An AI virtual assistant was also built inside Cigna’s mobile app, a conversational tool that can similarly address patient questions, while in the clinical setting, AI-enabled summarization has reduced note-taking by up to 90% for health practitioners who work for Cigna’s telehealth service MDLIVE.

“The principles behind all this are hopefully clear, which is, what problem are we trying to solve, and how can AI help?” says Andresen. “That’s always the starting point.”

John Kell

This story was originally featured on Fortune.com

The late Dolly Parton is known for her larger-than-life personality, decades-long country music career, and seemingly endless generosity. But one of her most consequential gifts came in the early days of the COVID-19 pandemic after most of the U.S. was sent under lockdown.

“My longtime friend Dr. Naji Abumrad, who’s been involved in research at Vanderbilt for many years, informed me that they were making some exciting advancements towards research of the coronavirus for a cure,” Parton wrote on social media in April 2020. “I am making a donation of $1 million to Vanderbilt towards that research and to encourage people that can afford it to make donations.”

“Keep the faith,” she added.

The Vanderbilt University Medical Center, based in her home state of Tennessee, was working with pharmaceutical giant Moderna—which went on to develop one of the most widely used vaccinations. Parton later said she was surprised to learn about the connection, but was happy to also become an unlikely face of vaccine confidence among skeptics.

“It was a big shock,” she told PEOPLE at the time. “Everybody was calling congratulating me, and I was saying, ‘Well, for what?’ And then of course my fans took it and ran with it: ‘I’m getting the Dolly vaccine! I’m getting the Dolly juice!’ I guess they thought if Dolly got it, it might be pretty safe.” 

Overall, more than 250 million people globally received the Moderna COVID-19 vaccine, while COVID vaccinations as a whole are estimated to have saved at least 14 million lives in the first year of their rollout alone. But Parton was quick to downplay her role, maintaining that her contribution was minimal.

“I got more credit than I deserved,” she added. “I was just trying to put my money where my heart is.”

Parton’s net worth was estimated at $450 million at the time of her passing, built on an empire spanning music as well as beauty, apparel, kitchenware, and her East Tennessee theme park, Dollywood.

Fortune reached out to Moderna for comment.

Parton amassed a $450 million fortune—but she spent much of her life giving it away

Philanthropy has been core to Parton throughout her life, in part because she knew what it was like to go without. She grew up with 11 siblings in a small cabin in the Great Smoky Mountains with no running water, electricity, or indoor plumbing. Their mother, Avie Lee, would sew scraps of fabric together to make clothes for the family.

Her father, Robert Lee Parton, never learned to read or write, but he became the inspiration for Dolly’s education work. In 1995, she started Imagination Library, mailing free books to children in her home county. It has since exploded into a global operation, delivering more than 3 million books a month across the U.S., Canada, the U.K., Ireland, and Australia. In total, the program has now shipped more than 300 million books.

Her approach to giving was rooted in the lessons she learned growing up: “Just remember the lessons my family taught me,” Parton said in a video for her Imagination Library. “Dream big dreams, and learn everything you can. And care for those who care for you—you do all of these things and you can be anyone you want to be.”

She’s also given away millions of dollars to natural disaster relief over the years—raising $700,000 for flood victims near her home in 2021, and contributing $1 million toward Hurricane Helene recovery in East Tennessee in 2024, on top of another $1 million from her businesses and the Dollywood Foundation.

In addition to the $1 million for COVID-19 vaccine research, she gave another $1 million to the Vanderbilt medical system in 2022 for pediatric infectious-disease research. It was the same hospital system where Parton died this week after a brief battle with cancer.

“I’m kind of addicted to the feeling of giving,” Parton told PEOPLE in 2021. “Knowing that I’m doing something good for someone else.”

This story was originally featured on Fortune.com

Meta Platforms agreed Wednesday to pay up to $17.1 billion to settle a landmark lawsuit brought by 29 states alleging the company deliberately engineered Facebook and Instagram to be addictive to children, marking the largest single settlement in the company’s history and the biggest tech-industry payout ever recorded in a single case.

Meta’s press release put the settlement at $18 billion, but even the more conservative $17.1 billion figure other states cited is still big enough to eclipse one of Silicon Valley’s biggest AI bets of the past year. To put it in perspective, the $17.1 billion number is a little more than three times the roughly $5 billion personal stake that Alexandr Wang held in Scale AI, a data-labeling company that supplies the human-annotated training data AI models are built on. Last year, Meta paid $14.3 billion for a 49% stake in the company last year and brought in Wang to lead its AI efforts of its new Superintelligence Labs, reporting directly to Mark Zuckerberg.

Put simply: the fine for allegedly hooking kids on Meta’s apps costs about three Alexandr Wangs.

For additional context, the company posted $60.46 billion in net income on $200.97 billion in revenue for full-year 2025, meaning the settlement equals roughly 27% of one year’s profit and about 8% of annual revenue. Meta has continued to spend aggressively even as its legal exposure mounted: the company raised its 2026 capital expenditure guidance to as much as $145 billion, driven largely by its AI buildout.

The company denied wrongdoing and it previously argued the states’ financial demands were “vastly disproportionate,” and in pretrial filings warned that the states’ own damages framework could theoretically produce penalties as high as $1.4 trillion—a figure close to Meta’s entire market capitalization. The states’ lawyers had signaled roughly $200 billion was a more realistic target at trial.

Averting a landmark trial

To put it in perspective, the settlement’s most unusual feature is that roughly $5 billion of the total isn’t guaranteed. Meta’s own announcement puts the total at $18 billion, with states receiving approximately 70%, or $12.7 billion, in annual installments over 10 years regardless of what happens elsewhere in the industry. The remaining 30%, roughly $5.3 billion, is released only if two conditions are met: TikTok and YouTube adopt matching daily time limits, night mode restrictions and age-verification measures, and each of those companies pays a matching sum, split evenly against the contingent pool. Some state attorneys general have cited a slightly lower total, $17.1 billion, built on a similar guaranteed-plus-contingent structure—a roughly $12.1 billion floor plus an additional $5 billion contingent on the same industry-wide adoption. The discrepancy likely reflects differences in how each side scoped the settlement, rather than one figure excluding money the other includes.

The settlement resolves federal Children’s Online Privacy Protection Act claims from all 29 states active in the lawsuit, along with separate state consumer-protection claims that California, Colorado, Kentucky and New Jersey were actively trying before U.S. District Judge Yvonne Gonzalez Rogers in Oakland.

Opening arguments had begun just over a week earlier, with California Deputy Attorney General Megan O’Neill telling the court that “Meta’s business model can be summed up in four simple words: ‘hook’ the users, ‘hold’ them for as long as they can, ‘harvest’ their data, and then ‘hide’ the truth from the public when making public statements.” She added, “It was especially bad for kids.”

As part of the deal, Meta agreed to nationwide safeguards for teen users of Facebook and Instagram, including daily usage limits and nighttime blocks.

Eclipsing Meta’s own record

The settlement more than triples Meta’s previous high-water mark: the $5 billion penalty the Federal Trade Commission imposed in 2019 over Cambridge Analytica-era privacy violations, which regulators at the time called “almost 20 times greater than the largest privacy or data security penalty ever imposed worldwide.”

Across the tech sector more broadly, the new settlement exceeds the EU’s four separate antitrust fines against Google—on search, Android, ad-tech and shopping—which together total roughly $12 billion over nearly a decade. It’s more than 10x Anthropic’s $1.5 billion payout to authors, the largest copyright settlement in U.S. history, and more than 10x Google’s $1.375 billion privacy settlement with Texas last year (Meta had a $1.4 billion settlement of its own with the state).

Wednesday’s deal caps a brutal year for Meta in the courts. A New Mexico jury found in March that the company had willfully violated state consumer-protection law by concealing what it knew about child sexual exploitation on its platforms, awarding $375 million in penalties.

As the case lingered in between phases, New Mexico Attorney General Raul Torrez, who has pursued Meta aggressively, criticized the company in April 2026 for threatening to shut down in the state rather than install safeguards: “Meta is showing the world how little it cares about child safety,” Torrez said. “Meta’s refusal to follow the laws that protect our kids tells you everything you need to know about this company and the character of its leaders. We know Meta has the ability to make these changes. For years the company has rewritten its own rules, redesigned its products, and even bent to the demands of dictators to preserve market access. This is not about technological capability. Meta simply refuses to place the safety of children ahead of engagement, advertising revenue, and profit.”

In March, a Los Angeles jury delivered the first verdict of its kind, finding Meta and Google’s YouTube negligent for designing their platforms to be addictive to children, in a case brought by a then-20-year-old plaintiff identified only as Kaley, or K.G.M., who said she became compulsively hooked on Instagram and YouTube as a child and suffered resulting depression, anxiety and suicidal thoughts. Jurors awarded her $6 million total—$3 million in compensatory damages and $3 million in punitive damages—after concluding both companies knew their products could harm minors and failed to warn users, with Meta shouldering 70% of the liability and Google 30%. The verdict was significant on principle: it marked the first time a jury had treated social media apps as defective products engineered to exploit developing brains, validating a legal strategy that targets platform design rather than content.

Kaley’s case was also the first of nearly 2,500 plaintiffs in a consolidated Southern California proceeding against Meta, Google, TikTok and Snap, meaning the verdict served as an early bellwether for the wave of similar suits still working through the courts. Meta said in a statement it “respectfully disagree[d] with the verdict” and would appeal, arguing that “teen mental health is profoundly complex and cannot be linked to a single app.”

In August, a New Mexico judge added another $567 million, ruling that Meta had created a “public nuisance” similar to air pollution. Two more bellwether cases remain scheduled for trial in October, and Meta faces thousands more similar lawsuits that are currently pending.

Whether Wednesday’s settlement slows that pipeline of litigation, or simply removes the most immediate and costly case from Meta’s docket, remains to be seen when the next bellwether trials begin this fall.

What Meta is actually changing

Pending judicial approval, Meta says the agreement will bring a specific set of default protections to under-18 users of Instagram and Facebook, most of which must remain in place for 10 years:

• A default two-hour daily time limit, cumulative across both apps, that teens can only disable with a parent’s permission.
• A default night mode blocking Facebook and Instagram use between midnight and 6 a.m.
• Muted notifications between 8 a.m. and 3 p.m. on school days, aside from direct messages and safety alerts.
• Usage prompts after every 15 minutes of continuous use, and again at the 60- and 90-minute daily marks.
• An option for a non-algorithmic, non-personalized default feed, which parents can require.
• Hidden like counts, disabled autoplay by default, and a ban on extreme makeup filters in addition to Meta’s existing cosmetic-surgery filter ban.
• Expanded age-verification technology and new parental alerts when a teen links a secondary account or interacts with a flagged adult account.

Notably, Meta structured its own commitment on a sliding scale tied to industry adoption. The Time Limit and Night Mode provisions start on a five-year commitment at the levels above; if TikTok and YouTube sign onto the same framework, Meta will extend those commitments to 10 years and tighten them further, to a one-hour daily limit and a 10 p.m.–7 a.m. night block. Meta also published an open letter Wednesday explicitly calling on TikTok and YouTube to adopt the same standard, arguing that “when teens are restricted on one app, they simply move to another.”

The agreement additionally creates an independent research foundation, to which Meta will contribute consented user data for studies on teen well-being, and calls for an independent auditor to assess Meta’s compliance annually for five years.

C.J. Mahoney, Meta’s chief legal officer, framed the deal as an industry challenge as much as a settlement: “Our new Time Limit commitments, Night Mode features and usage limits during school hours set the right path forward for our whole industry, but this framework will only work if all our peers join us,” he said. “We therefore call on our industry peers, TikTok and YouTube, to implement this new framework, right away.”

For this story, Fortune journalists used generative AI as a research tool. An editor verified the accuracy of the information before publishing.

This story was originally featured on Fortune.com

In January, Jeff La Marca got a prescription for the popular weight loss drug Zepbound. But he couldn’t afford the $750 monthly price tag.

Then Medicare launched an 18-month pilot program that offers GLP-1 medications to some enrollees for only $50 a month. La Marca thought he might finally be able to afford the drug.

“I thought, ‘Thank God, there’s a path,’” said La Marca, who lives in Basking Ridge, New Jersey, and has tried numerous diets and exercise regimes.

But the 68-year-old’s celebration was short-lived.

His application to the pilot program was denied.

La Marca has severe obstructive sleep apnea, one of several diagnoses that exclude patients from the Bridge program’s $50 monthly price. The notification didn’t say why he was rejected. He thinks that if he didn’t have that diagnosis, he would qualify due to his weight.

“I’m obese, morbidly obese, BMI 42. I had quadruple heart bypass surgery. I’m at risk for stroke. I’m prediabetic. And yet I can’t get it. I’m livid,” he said.

A Temporary Patch for a Long-Standing Gap

About 1 in 5 American adults have taken a GLP-1 medication, and most of them, including those with health insurance, say the drugs are difficult to afford. Federal law has long barred Medicare from covering drugs prescribed solely for weight loss, which is why the Medicare GLP-1 Bridge program made a big splash when it launched in July.

It’s a short-term pilot program in which Medicare is offering coverage of three GLP-1s for weight loss and management, to see if that would save Medicare money later. Eligible patients must be enrolled in Medicare Part D, a prescription drug coverage add-on to Medicare. Even though people must have Part D insurance to qualify, the preauthorization request doesn’t go through the insurer; it’s instead submitted to a separate system run by a contractor for the Centers for Medicare & Medicaid Services.

The pilot includes Wegovy, the KwikPen formulation of Zepbound, and the oral medication Foundayo.

Under the pilot, many Medicare beneficiaries with a body mass index of 35 or higher — the upper range of obesity — qualify for coverage of one of those drugs, if prescribed. Those otherwise eligible who have a BMI of 27 to 34 can qualify if they also have certain health conditions, such as prediabetes or cardiovascular disease.

But buried in the fine print is a distinction that’s tripping up patients like La Marca: The $50 price under Bridge applies only to people using the drug solely for weight loss. Anyone who has a qualifying medical condition that the Food and Drug Administration has approved GLP-1s to treat, such as Type 2 diabetes or moderate to severe obstructive sleep apnea, is instead routed back to their Medicare Part D prescription drug plan, which can require copays of hundreds of dollars a month for GLP-1s.

“The Bridge program was designed to target those people who can’t get GLP-1 coverage through Part D but would benefit from taking one for weight loss,” said Juliette Cubanski, who directs the Program on Medicare Policy at KFF, a health information nonprofit that includes KFF Health News.

The cost to Medicare of subsidizing the drugs will depend largely on how many people use the program, and the federal government hasn’t released an estimate.

Cubanski has estimated that 3.8 million people qualify and that, if a quarter of them enroll in Bridge and remain on treatment for the program’s full 18 months, it will cost Medicare about $3.3 billion. If three-quarters enroll, costs could rise to $10 billion.

If the government expanded the program to include the additional 5.9 million people who are overweight and already eligible for GLP-1 coverage through Medicare Part D, it would add billions more to the program’s cost.

The demonstration’s initial weeks have been positive, and most prior authorization requests have been completed in under 12 hours, CMS spokesperson Timothy Foster said.

“This has allowed thousands of eligible beneficiaries to access GLP-1 medications for weight loss at pharmacies nationwide,” Foster said.

GLP-1s Aren’t Covered

Patients like La Marca are left in a tough spot, qualifying for Part D coverage of a GLP-1 but facing much higher cost sharing.

“‘Coverage’ doesn’t always mean ‘affordable,’” said primary care physician Taylor Lacy, who describes herself as a “big proponent” of GLP-1s and practices at Sunflower Medical Group in Roeland Park, Kansas.

The Bridge program is leaving behind patients with the greatest medical need, she said. She noted that many Medicare patients already must navigate prior authorization and spend months trying alternate, often cheaper treatments, a process known as step therapy, before finally getting approval — only to arrive at the pharmacy counter and discover that their GLP-1 copays will run them $200 to $600 a month, if not more.

Researchers studying how Medicare insurers cover GLP-1s have found that recipients have faced increases in out-of-pocket costs and that almost all plans now require prior authorization, which can make getting the drugs more difficult.

Chris Bond, a spokesperson for insurance industry trade group AHIP, blamed drugmakers’ prices, “which they alone set and they alone can lower.”

La Marca’s insurer declined to answer specific questions about La Marca’s case.

Left Waiting

For now, La Marca’s GLP-1 prescription remains unfilled. The severe sleep apnea diagnosis that helps establish his medical need is also what excludes him from the discount program that would bring the cost within his reach.

As he reflected on his appeals and the dead ends, La Marca paused, his eyes filling with tears of frustration.

“This is now my quest, because it’s my only chance to improve my health,” he said. “It’s the only thing left. I’ve tried everything.”

KFF Health News is a national newsroom that produces in-depth journalism about health issues and is one of the core operating programs at KFF—an independent source of health policy research, polling, and journalism. Learn more about KFF.

This article first appeared on KFF Health News and is republished here under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.

This story was originally featured on Fortune.com

Cambodian and U.S. anti-drug agencies have uncovered a network that authorities say uses cryptocurrency to launder money for Mexico’s notorious Sinaloa Cartel, officials from both countries said Friday.

Meas Vyrith, secretary-general of Cambodia’s National Authority for Combating Drugs, said Cambodian authorities are further investigating money laundering operations by the Sinaloa Cartel, a powerful criminal conglomerate designated a “foreign terrorist organization” by the Trump administration.

Cambodian anti-drug police working with the U.S. Drug Enforcement Administration had made several arrests and seized assets, drug laboratories and storage facilities them at four locations in the Cambodian capital, Phnom Penh, and neighboring Kandal province from Aug. 1 to 5, He said.

They confiscated over 200 kilograms (440 pounds) of illegal drugs and more than one metric ton (2,200 pounds) of precursor chemicals used to produce them.

Meas Vyrith also said that U.S. officials had seized about $7 million in cryptocurrency linked to the cartel in an earlier law enforcement operation in New Jersey and suspects in that case are now being sought in Cambodia.

The U.S. Embassy had said in a statement posted Thursday on its Facebook page that collaborative efforts involving Drug Enforcement Administration offices in Phnom Penh and New Jersey “successfully uncovered a local network laundering cryptocurrency on behalf of the Sinaloa Cartel.”

The Sinaloa Cartel has become a major producer of the synthetic opioid fentanyl, blamed for tens of thousands of overdose deaths each year in the U.S. The group has obtained precursor chemicals, including from China, to produce fentanyl in Mexico and smuggle it into the U.S.

Cambodia serves largely as a transit area for drugs such as methamphetamine that originate in the Golden Triangle, the region where the borders of Thailand, Myanmar and Laos meet.

But in recent years, it has also been a stronghold for Chinese-led organized crime groups, best known for running scam centers that have bilked people around the world out of billions of dollars while employing many thousands of foreigners in slavelike conditions.

Over the past year, Cambodia has intensified a crackdown on such operations, shutting down scam centers and arresting several alleged kingpins, including some who had established legal financial institutions.

A report issued in July by the United Nations Office on Drugs and Crime said several transnational organized crime groups from outside Southeast Asia are important criminal actors in the region, primarily involved in trafficking large quantities of methamphetamine and cocaine.

It said Latin American cartels’ engagement with Southeast Asia “appears to extend beyond drug and precursor chemical supply, and the region now functions as a supply, financial management, and logistics hub for multiple Latin American criminal organizations.”

In 2024, U.S. federal prosecutors said the Sinaloa Cartel worked with Chinese underground banking groups in the United States to launder more than $50 million from the sale of fentanyl, cocaine and other drugs.

___

Associated Press writer Grant Peck in Bangkok contributed to this report.

This story was originally featured on Fortune.com

Workers are concerned that they’ll lose their jobs to AI—and Bill Gates thinks they are right to worry. He has identified a loophole which may be incentivizing businesses to shift away from human capital to robots.

Last week, Pew Research released a study finding that 71% of adults think AI will lead to fewer jobs in the United States over the next two decades, up from 64% in 2024—only 5% think it will lead to more jobs. And young people, those whose job prospects are most likely to be impacted over the long run, are equally as concerned as their older counterparts: 73% believe they’ll get fewer career opportunities because of the transformative technology over the next 20 years.

Despite the concerns of the public, and the watchful eye of policymakers like former Fed chairman Jerome Powell, Microsoft co-founder Bill Gates has suggested that employers’ bottom lines, under current tax frameworks, may actually benefit from using AI-empowered machines rather than human workers.

In a new essay posted to his blog, the entrepreneur and philanthropist wrote: “Right now, if you’re an employer and you hire someone, you pay payroll taxes on their earnings. But if you buy a robot, you can usually write it off right away as a business expense. The tax system nudges you toward replacing people with machines.”

As a result, Gates suggests taxing AI tokens and robots.

This revenue generation is also necessary for governments, he suggests, which will be subject to lower revenues from income tax if fewer people are working, combined with greater demand for retraining and social security benefits.

“The funds will have to come from somewhere at a time when budgets are stretched,” Gates writes. His nearly 6,000-word essay came hot off the news that U.S. government borrowing has hit $40 trillion.

“A tax would slow the rush away from human labor a little and raise money for retraining and a stronger safety net,” Gates explained, “It would need to be targeted so it does not slow down the purely beneficial uses of AI, like making medicine and education cheaper.”

Mixed response

Gates made the same suggestion nearly a decade ago and was subject to robust criticism at the time. Writing before the current AI boom, former Treasury Secretary Larry Summers said in 2017 that Gates was “seriously astray” with the suggestion, adding: “Gates’s robot tax risks essentially being protectionism against progress.”

Robert Seamans, a professor at NYU Stern, similarly chimed in 2017 that “there’s no question that the potential increase in robots and automation requires policymakers to rethink fiscal policy for the 21st century (and other policy as well, such as education and retraining policy).”

But he added that “based on the data we currently have, a tax on robots would be bad policy. Robot taxes would dissuade firms from investing in robots, which would lower economic growth, and, to the extent that robots complement labor in some cases, would lead to less hiring and lower wage growth.”

Gates, writing in the piece shared with Fortune ahead of publication this week, is aware of such criticism. But he insists that critics are “not considering the broader value of work for individuals and society. And with all the accelerated innovation we will have, we’ll be able to afford a little inefficiency as the price for keeping people employed.”

He added that while tax is not the whole solution to the threat of AI (the three major risks of which he outlines in the rest of the essay), it is a suggestion that forms part of a “wise response.”

Gates adds: “However we raise money for more assistance, it needs to reach the people who need it most, including workers who lose their jobs to AI and robots, people whose hours or wages decline, and communities where the losses are concentrated.”

This story was originally featured on Fortune.com

Two Saudi brothers have amassed a fortune of about $1.4 billion as the technology company they control rides the kingdom’s multibillion-dollar push to build artificial intelligence infrastructure.

Al Moammar Information Systems Co. hit a record high Wednesday after it said a data center project with Humain, backed by Saudi Arabia’s sovereign wealth fund, was being expanded. The contract is now valued at more than 8.76 billion riyals ($2.34 billion), a shade under MIS’s market capitalization that hit 9.75 billion-riyal after an 87% surge this year.

Founded in 1979, MIS began as an information technology services provider before pivoting to data centers. Brothers Ibrahim Al Moammar and Khalid Al Moammar own just over 50% of the company, giving them a combined fortune of about $1.4 billion, according to the Bloomberg Billionaires Index, underscoring the fortunes being minted as Saudi Arabia makes AI central to its economic diversification.

MIS said the AI data center project with Humain will have capacity of 250 megawatts, up from 50 megawatts earlier. The expanded contract is worth about seven times the company’s 2025 revenue. It also signed a separate deal with Humain this week to provide data center co-location services.

Besides Humain, the company’s Saudi clients include Aramco, STC and Al Rajhi, along with various ministries and other public-sector entities, according to its latest annual report.

MIS is benefiting from a broader Gulf spending boom on AI infrastructure. Sovereign wealth funds in Saudi Arabia, Qatar, Kuwait and the United Arab Emirates oversee assets worth more than $4 trillion, with the oil-rich states seeking to make AI a core part of their diversification efforts. That has turned the Gulf into a key player in an industry poised to reshape many aspects of everyday life.

Gulf states are pressing ahead with large-scale construction projects even as data centers have been targeted in the US war with Iran.

State-backed Khazna Data Center Ltd. in the UAE continues to develop a sprawling AI campus in Abu Dhabi. Meanwhile, Saudi Arabia’s Center3, a subsidiary of the kingdom’s largest telecommunications company, has built nearly half of the country’s data centers.

Qatar’s wealth fund is also partnering with Brookfield Asset Management Ltd. on a $20 billion venture to invest in AI infrastructure.

This story was originally featured on Fortune.com

Dubai International Airport, one of the world’s busiest hubs, said Wednesday it saw a sharp drop in passengers during the second quarter as the Iran war disrupted air travel and prompted many travelers to choose routes that avoided the Middle East.

The airport recorded 13 million passengers during the second quarter, following 31.5 million passengers in the first half. That was a 31.3% drop over passenger traffic in the first half of 2025, the airport said.

Air traffic in the United Arab Emirates and elsewhere in the Gulf region was disrupted by shutdowns and restrictions after the United States and Israel attacked Iran on Feb. 28. The Dubai airport itself suffered minor damage in attacks from Iran.

Aircraft movements totaled 150,600 in the first half, down 32.1% year on year, with only 62,500 in the second quarter, the airport said.

CEO Paul Griffiths noted, however, there has been a steady increase in passenger volume each month in the second quarter, and said the momentum is expected to continue.

“The events of the disruption, which lasted two months, that had an impact on traffic,” Griffiths told The Associated Press. “However, the trajectory now is universally positive.”

He said the airport was waiting for overseas airlines, particularly from Europe, to restart operations.

“I’m very confident we’ll be back to full operation in a very short space of time,” he said.

This story was originally featured on Fortune.com

The emergence of new technology has long inspired humans to shape our environments.  The invention of steel permitted the rise of skyscrapers, the automobile inspired highway systems and the internet gave rise to a whole new set of digital spaces.

Now, a new force for change is upon us. The pace of improvement in AI models has led many observers to predict that human intelligence may soon be overtaken by the cognitive capacity of machines. Anthropic CEO Dario Amodei imagines a world with “a country of geniuses in a data center.” Elon Musk posited this summer that “AI may exceed the sum of human intelligence in around five years. There really won’t be anything that AI can’t do better than humans, apart from being human, perhaps.”

In such a world, it seems likely that the digital and physical affordances we have constructed for the benefit of humans will be superseded by or at least complemented by infrastructure that is specifically designed for ready and effective use by “bots,” both digital and physical.

Today, our world is designed for human cognition, cadence and trust. Now imagine a world built for machine intelligence, wire speeds and code-based contracts. Humans may no longer be the primary actors in this new theater of our own making.

Bending and Breaking Physical Spaces

AI agents may transform the physical world. It seems superfluous to say that our homes, offices, stores, and public spaces are designed for humans. Think of the centuries of architectural design dedicated to making those spaces useful, accommodating, safe, and efficient. Never mind the time and creativity devoted to making them aesthetically pleasing. But as robots begin to populate our society and workplaces, these locations will need to be accessible and effective for humans and robots alike. Both will need architecture and design for human-machine cooperation.

Building new physical structures is often a slow and expensive process. But humans have rebuilt our physical world to accommodate new technologies in the past, with changes in transportation offering perhaps the clearest examples. Ancient Greek roads had wheel ruts to guide carts over steep or slippery terrain. Railroads necessitated new logistical hubs and services around rail terminals. This had downstream effects, changing the value of land, expanding the potential to commute long distances to cities, and creating physical dividing lines between neighborhoods.

In the 20th century, the rise of mass-produced automobiles led to multilane highways, the construction of new parking facilities, and the rise of the suburbs. In our own time, how might autonomous vehicles reshape the physical environment? Autonomous vehicles do not have the same requirements as traditional cars with human drivers. A human driver requires parking near the driver’s end destination. An autonomous vehicle could drop off a passenger and move on to its next task. Alternatively, it could queue elsewhere, waiting for pickup time. Today, roughly 22% of land in cities with over 1 million inhabitants is used for parking. The rise of autonomous vehicles could reduce that figure, opening land to other uses. Merge lanes could be smaller, given that AVs often operate with tighter tolerances for risk than human drivers. Roads could also be narrower, with expanded room for sidewalks or bike lanes. Furthermore, without the need for steering wheels or pedals, the physical shape of cars may change, opening the possibility for new forms for the cars of the future.

Such changes won’t happen overnight, but our physical world is already being rebuilt to accommodate AI-enabled technologies. In Dallas, Texas, efforts are underway to prepare for the advent of Zipline’s drone-based delivery service. Born from a successful effort to speed the delivery of blood plasma to remote hospitals in Rwanda, Zipline has evolved to serve consumers with fixed-wing drone deliveries of everything from Starbucks to DoorDash. One of Zipline’s primary partners in the region is Walmart, which is beginning to adapt its store design to drive more convenience and efficiency in staging deliveries by drone.

Workers at one such facility recently began cutting openings into the walls of a Walmart Supercenter — not for a renovation, but for delivery drones, letting employees load packages directly from the sales floor into the sky, bypassing the free-standing charging posts the companies used at first. It’s a small, literal crack in a retail architecture built for nearly a century around one assumption: that the customer walking through the front door is human.

The impact of such efforts could soon spread far beyond individual facilities. In August Zipline announced a partnership with Uber to accelerate the use of drones in deliveries. Noting the potential effect of such parentships beyond how quickly customers could now get their next order, Zipline co-founder Keller Cliffton stated, “Every great transportation revolution has changed where people live, how businesses operate, and how economies grow.” 

Ultimately, supply stations for Zipline may become purpose-built for these kinds of workflows. New apartment buildings may be designed with convenient landing spots for the drones to drop their payloads on rooftops or courtyards. 

While many efforts exist to build humanoid robots that share our form and therefore can operate reliably in our environments, it is likely that such machines will be outnumbered by robots that abandon the human form in favor of utilitarian shapes designed for efficiency. As former Uber CEO Travis Kalanick observed in the public unveiling of his new “Atoms” platform, a specialized robot that looks nothing like a human chef could optimally tackle the task of “making 1,000 pancakes an hour,” while a humanoid, not optimized for that function, would struggle to complete this enormous task.

Some transformations are even further along. Kalanick’s CloudKitchens business, a subsidiary of Atoms, is also shifting further in the direction of purpose-built infrastructure for robots. With the remarkable rise in online food delivery, many restaurants struggle to balance service and food prep for diners in their establishments while also preparing staging and queuing delivery for takeout orders. Kalanick initially addressed that challenge by offering restaurateurs new spaces that were solely for the preparation of meals for delivery and were located strategically around cities and suburbs. Supplies came in on one side via loading docks. Kitchens were structured for high throughput and delivery vehicles lined up on the other side of the building to take finished meals and speed them along critical thoroughfares for timely and low-cost delivery. Now, that model is shifting to retrofitting these facilities for robotic food prep and kitchen design optimized for robots rather than chefs. With these new technologies, as Kalanick says, “digitizing the physical world is my life’s work.”

As we reconfigure our physical spaces for robotic collaborators, expect new expectations and disruptions. Humans want physical infrastructure built for their comfort and efficiency, with an eye towards aesthetic beauty, or at least the familiar; robots function best with simplicity, easy transit and “beauty” expressed in utility not visual appearance.

The End of Software (As We Know It)?

The software ecosystem is already being re-shaped for the convenience of AI agents. AI agents, born from large language models (LLMs), are designed to pursue goals on behalf of users with some degrees of autonomy and reasoning capability. They also feature the ability to invoke and utilize computing resources such as browsers, websites, applications and data stores to help accomplish these goals. 

Progress in this area has been rapid. The state of the art has quickly evolved from early demonstrations of agents navigating web pages to perform online shopping, a mode that visually approximates your grandparents learning how to use Amazon.com in 1999, tentatively clicking around, back-spacing and often invoking the wrong commands. Now, agents are capable of performing sophisticated workflows and traversing multiple applications by leveraging a critical artifact of modern computing, the application programming interface (API). 

APIs have become a valuable and dominant way to connect applications to data sources and to each other. Think of them as on-ramps and off-ramps that connect highways to cities—and other highways. They were part of the arcane plumbing that lies beneath the foundation of our increasingly well-crafted software and workflows. Humans engaged user interfaces to command software and engage with outputs like dashboards, while APIs labored in the background. 

Now, AI agents can command individual applications and autonomously compose those API interactions into multi-step workflows. In this world, APIs are not only the plumbing, they are the interface. Agents don’t require a beautiful canvas to function. Rather, they need to access applications seamlessly, go right to the heart of the data store or logic layer to perform an operation and then move on to the next step. Agents favor applications that are set up to help them navigate – modern APIs, machine-readable content that functions as a user guide to the platform, service-level guarantees, transaction capability and telemetry to assess effectiveness.

These shifting requirements signal the rise of “headless software,” platforms that are optimized for agents and that favor utilitarian interfaces over elegant design. This evolution has been well described by leading software entrepreneurs such as Dharmesh Shah, co-founder and CTO of HubSpot, and Aaron Levie, founder and CEO of Box. Levie has observed that “enterprises need to be able to ensure all of their software works across any set of agents they choose.” 

With the rise of headless software platforms, SaaS applications act increasingly like data repositories that agents can traverse and stitch into complete workflows. To the extent this paradigm continues to emerge, it implies a kind of relegation of some classical software applications with their elegant interfaces designed to engage humans. So-called systems of record remain valuable as reliable, persistent stores of corporate data, but they become a watering hole along the agentic journey, not a destination. Consequently, the value proposition of classical (in other words, human-centric) software may shrink commensurate with this new role and its pricing power. Perhaps the entire pricing model shifts in favor of usage or outcome-based revenue models that better align software vendors with customers’ desire for value realization.

The Worldwide Agent Web

Websites are undergoing a similar transformation. Decades of work to perfect the human appeal of websites’ user interfaces and commerce sites allow us to browse and shop in a familiar manner, loading virtual shopping items into virtual carts, and seeking our own optimal combination of price, quality and availability. However, such designs may now be superseded by austere sites that allow agents to act efficiently on our behalf. 

Each form optimizes for different functions and users. Humans browse the internet looking for images, drop-down menus, buttons, slider bars, folders, dashboards, and web forms that allow us to interact with computing in ways that are familiar and intuitive. But to AI agents, these features are distractions, barriers to their direct access to data, capabilities and logic they need to complete tasks. Humans want intuitive interfaces and appealing visuals. AI agents want “clean” API surfaces, a descriptive markdown file, and JSON schemas.

The transformation of the internet to more agentic interfaces is already underway. As Mathew Prince, the co-founder and CEO of Cloudflare, a leading internet infrastructure company, Cloudflare, said in June of this year, “Agentic traffic [is] growing so fast that bots have now passed human traffic online for the first time in the Internet’s history.”

As agents traverse the web on our behalf, the practice of Search Engine Optimization (SEO), through which websites compete for human visitation by tuning their appeal and seeking referral of users from Google and other discovery platforms, is giving way to Artificial Engine Optimization (AEO), which is designed to induce agents to promote, visit and even transact on websites by increasing their visibility and appeal to our digital delegates.

As with software, the features that optimize for agents differ significantly from the intricate features of modern internet sites built for human use. Commerce and content purveyors are scrambling to contribute their data to train LLMs and place themselves squarely in the transactional path of agents. The implications for online commerce are particularly significant: The parallel processing capabilities of AI can allow agents to comparison shop at scales and speeds far beyond the capacities of humans. 

Millions of websites have advantages of legacy and incumbency. But over time, people’s loyalty to online brands may be supplanted by the efficacy of their interfaces for agents. Unlike humans, agents don’t shop habitually or “get used to” shopping on any given platform. Dynamic pricing may become more pervasive amid these accelerated shopping sprees. A logical outcome for this development could be the rise of auction pricing at scale, where agents put out “requests for proposals” for every sweater or light bulb purchase they make on our behalf and induce online stores to compete to win every piece of business in real time. Tokenized payments may rise in use for agent-based transactions, where immutability and speed are most desirable. By the same token, tolling infrastructure may emerge that permits agents to autonomously “pay for” access to content or other online resources as they pursue our goals. Both of these possibilities could meaningfully challenge the current economic structure of the web, where human-driven search lies at the heart of online monetization and existing payment methods and rails are necessary to consummate transactions. 

How this Might Play Out 

We are beginning to see evidence that our digital and physical spaces are evolving with the advances of AI and robotics. This raises a series of critical questions. How dominant will “machine spaces” become? What models for co-existence and control will emerge? Will “human spaces” become a rounding error too?

Many of these changes are nascent. As such, they will coexist with established architectures. In these early years, we should expect the development of “parallel universes.” Software and websites built for humans won’t vanish. Factories, warehouses and stores won’t suddenly close. But a new and different infrastructure will eventually emerge – one that is designed to serve our digital workers in contexts where humans may no longer be the primary actors. 

We may begin to consider the ergonomics of agents and robots as much or more than we consider optimization for humans. While some may view this wistfully, it seems likely that these changes and the tremendous efficiency gains they promise may free up capital and creativity to build new, entirely human-centric architectures that are not burdened by the compromises necessary to accommodate people and machines but rather are tuned exclusively to our highest tastes and aspirations.

The irony is that a world redesigned for machines may eventually allow us to recover something more purely human. If factories, kitchens, warehouses, websites, and workflows become increasingly machine-native, then human spaces may be relieved of some of their utilitarian burden. We may build more places for beauty, reflection, play, learning, and community precisely because the machine world has absorbed more of the work. The danger is that human spaces become incidental. The opportunity is that they become sublime.

The opinions expressed in Fortune.com commentary pieces are solely the views of their authors and do not necessarily reflect the opinions and beliefs of Fortune.

This story was originally featured on Fortune.com

Adam Mosseri, the Meta executive at the helm of Instagram, defended the platform’s safety record in testimony Tuesday and took issue with a plaintiff attorney’s emphasis on low usage numbers for a feature designed to encourage young users to take breaks.

Mosseri was appearing at a trial that pits Meta against the states of California, Colorado, Kentucky and New Jersey, which accuse the social media giant of contributing to the youth mental health crisis by knowingly and deliberately designing features that addict children to its platforms.

Jason Slothouber, a senior prosecutor for the Colorado Attorney General’s Office, pressed Mosseri on the low adoption rate among teens for the “take a break” feature, which prompts users to step away if they’ve been scrolling a while. Very few teens used the feature when Instagram first introduced it, but since launching separate teen accounts in 2024, Meta has made it the default setting for teenage users.

Mosseri appeared frustrated that Slothouber was focusing on a single Instagram safety feature when the Meta executive said, “There are many features we launch over many aspects of safety and well-being” that the company tries to improve over time.

The trial began last week in federal court in Oakland, California, and is expected to last about six weeks. The four states were among 29 that sued the tech giant in 2023 over child safety and privacy — the other 25 will go to trial later. The company also faces lawsuits in state courts, including one underway in Tennessee.

Mosseri, who has led Instagram since 2018, testified for about an hour and is due back on the stand Wednesday. He also testified earlier this year in the Los Angeles state court trial that pitted Meta and YouTube against a 20-year-old plaintiff identified only by the initials “KGM,” who said she became addicted to social media as a child. Meta and Google were found liable for harms caused by their platforms, and KGM was awarded a total of $6 million.

In earlier testimony at the Oakland trial, Francesco Fogu, Instagram’s director of product design, had contentious exchanges with the plaintiffs’ lawyer over internal chats and documents shown to the jury. In one instance, jurors were shown a passage from a document saying, “As a company, we make different choices when it comes to regulatory response. Sometimes we don’t comply and accept a fine. Sometimes we comply in the most minimal or literal way possible.”

Fogu said several times he did not remember whether he wrote those words. As Slothouber kept pressing, Judge Yvonne Gonzalez Rogers stepped in to say, “We get the point.”

Jurors also heard from former Meta employee George Volichenko, who worked in marketing analytics and as a data scientist from 2016 to 2018 and from 2022 to 2023. During his second stint at the company, when he worked on Instagram’s mental well-being team, Volichenko said he was told the team exists primarily to protect the company in upcoming lawsuits.

In his experience, Volichenko said, “Everything I saw was kind of supporting that” — from having limited agency to feedback and guidance from leadership.

Discussing Instagram’s “take a break” feature, Volichenko said his team had argued it should be automatically turned on for younger teens who use Instagram. That’s because people are far more likely to use a feature if it is “opt-out” — that is, they need to manually turn it off if they don’t want to use it — rather than “opt-in,” which means they have to manually turn it on.

They did not get the approval.

“The tradeoff to core metrics was not desirable,” Volichenko said, referring to Meta’s metrics for how long and how often people use its products.

He soon left Meta a second time.

“I didn’t feel like the company at large was aligned with my values,” he said.

This story was originally featured on Fortune.com

President Donald Trump’s trade war with Canada is escalating as the midterm elections approach, threatening Republican efforts to address voters’ economic concerns in a year when control of the U.S. Senate hinges on states along the border between the United States and its northern neighbor.

The dispute flared over the weekend after negotiations broke down, leading Trump to raise tariffs on $20 billion in Canadian imports. Canada plans to announce tariffs of its own on Tuesday, and the spiraling conflict could lead to higher prices and scrambled supply chains for Americans already aggravated at the president’s management of the economy.

Republican Sen. Susan Collins of Maine, one of Democrats’ top targets this year, warned that fallout from Trump’s approach would hurt U.S. businesses and consumers.

“Imposing new tariffs on Canada is a mistake,” Collins said while campaigning Monday, and she mentioned lobsters, blueberries, lumber and other Maine products that end up in Canadian markets.

The issue also puts pressure on Republicans in Michigan, Ohio and Alaska, states where Canada is an important trading partner. Many Democrats seem eager to capitalize on the matter as they try to regain the Senate majority, despite the party’s own history with protectionist sentiments.

“Trump is escalating a trade war with Canada for his own vanity,” Michigan’s Democratic nominee Abdul El-Sayed said on social media, adding that his Republican opponent, former Rep. Mike Rogers, is a “rubber stamp” for such policies. A third of the state’s exports go north of the border.

Marc Short, a top adviser to then-Vice President Mike Pence during the first Trump presidency, said the issue is a political trap for Republicans.

“It’s hard, obviously, because you don’t want to incur the wrath of the president,” he said. “But at the same time, I think if you’re representing agricultural states, especially, your voters are probably anxious to have somebody representing their interests in Washington right now.”

Trump charges forward on tariffs

It’s possible that Trump will change course or delay his plans. But for now, the president is making no apologies for the economic turmoil.

“Canada has been ripping off the United States for years,” Trump blasted on his Truth Social platform Monday, adding that he will raise tariffs on all Canadian automobiles and auto parts and steel to 50% in 2027. He added, “WE DON’T NEED CANADA, THEY NEED US!”

Trump’s top trade official more calmly downplayed the dispute. “This is something where we don’t actually expect a huge impact,” U.S. Trade Representative Jamieson Greer told reporters outside the West Wing.

Vice President JD Vance visited Maine on Monday, where he praised “our very independent friend Susan Collins” and assured voters “we’re very mindful of the fact that Maine is a border state with Canada.” He said the administration is trying to make sure Maine “gets a fair deal.”

Collins did not appear with Vance on Monday or during his last trip to Maine. She campaigned on her own as she tries to hold off a challenge from Democratic nominee Troy Jackson, a former state legislative leader.

Jackson, a logger before going into politics, said tariffs are another example of how Collins does not do enough to stand up to the president.

“Troy spent most of his life working along the Canadian border, so he knows how important this relationship is to Maine’s economy,” Jackson spokesman Dan Gottlieb said.

Republicans are trying to defend Senate control

Trump made no secret of his affection for tariffs during his comeback campaign, promising that higher taxes on imports would generate a windfall for the U.S. Treasury and boost domestic manufacturing. But concerns about inflation and affordability have not receded, including in states with key races this year.

Maine, Ohio, Michigan and Alaska boast industries including fisheries, auto parts, lumber and produce that export items across the northern border, while Canadian imports are sold by a range of U.S. retailers.

Iowa, which also has a competitive Senate race, does not border Canada or its waters, but also exports more goods to Canada than any other nation.

Majority Forward, a political action committee tied to Senate Democrats, already ran television advertisements against Republican Sen. Dan Sullivan of Alaska during last year’s partial government shutdown.

“The tariffs are hitting Alaska the hardest,” the ad said. “Call Dan Sullivan and tell him … stop raising our costs.”

Trade is a key issue in Ohio

Former Ohio Sen. Sherrod Brown is trying to return to Washington by unseating Republican Sen. Jon Husted. Brown has long been a union-friendly protectionist Democrat. But he’s argued against Trump’s approach, saying it’s one thing to get aggressive with an adversarial economic powerhouse like China but another to impose uneven, unpredictable tariffs on neighboring nations.

Husted signed a bipartisan letter earlier this year urging the administration to proceed carefully while renegotiating a trade agreement with Canada and Mexico. But he’s also embraced the White House’s economic policies, recently appearing with Vance at an Ohio steel plant to praise the administration’s economic agenda.

“Today is a new day, it truly is,” Husted said. “It’s a new day because of the ‘America First’ agenda.”

Brown has not yet criticized Husted on Canadian tariffs, concentrating instead on the senator’s support for data centers and Trump’s war with Iran. But Senate Majority PAC spokeswoman Lauren French said the Canada tariffs fight fits seamlessly into the broader case that Brown and other Democrats are making about Trump and his allies.

“It’s another proof point for the argument that this is a guy who continues to raise your costs for no reason at all,” she said.

Vance says Trump wants ‘fairness’

Short said Trump’s first-term protectionism was easier to defend because it was more focused on China. In the second Trump presidency, Short said, “we’ve so alienated our normal trading partners that part of their retaliation has been not to buy agricultural products,” thus cutting off replacement markets for any lost trade with Beijing.

In Maine, Vance insisted Trump only wants to level the playing field with Canada.

“They don’t expect anybody to fight back,” Vance said. “We’re sick of that.”

He also criticized Canada as treating China more fairly than the U.S. in trade negotiations.

“It’s over,” Vance said. “We expect fairness in our trade policy.”

Collins shared a different goal.

“I really want us to go back to the very friendly, economically beneficial relationship that we have with our Canadian neighbors,” she said.

___

Barrow reported from Atlanta. Associated Press writers Julie Carr Smyth in Columbus, Ohio, and Seung Min Kim in Washington contributed to this report.

This story was originally featured on Fortune.com

Retailer Target pulled a children’s Halloween costume from its shelves Monday after widespread social media backlash, apologizing for selling a garment critics said evoked racist minstrel caricatures.

Critics say the costume, along with its promotional image of a Black child modeling the get-up, is reminiscent of caricatures from Jim Crow-era minstrel shows that embedded stereotypes into popular culture and helped justify discrimination, segregation and racial violence.

“As a company, we know we got this wrong, and we are deeply sorry. The costume is offensive and should never have been part of our assortment. It is no longer available for sale,” Target said in a statement.

“We know this is especially hurtful for our Black guests, team members and partners. Removing the costume is an important first step, and the company is looking closely at how this happened and what needs to change to ensure this won’t happen again.”

The costume misstep happened as Target’s sales have started rebounding from earlier company moves that alienated some customers and from a perception the quality of its stores and merchandise had declined.

It is the latest controversy for the retailer, which joined companies like Meta, Walmart, and McDonald’s in rolling back their diversity, equity, and inclusion efforts after President Donald Trump won the 2024 election.

The “Kids’ Glows Under Blacklight Circus Clown Halloween costume,” an orange-and-black circus clown costume, now deleted from the Target website, featured gloves, a top hat and a grinning mask with large teeth.

‘Not simply getting it wrong’

“A Jim Crow-era minstrel costume is not simply getting it wrong. It is a profoundly harmful symbol of racism that should never have been designed, approved, or sold by Target,” said the Rev. Jamal Bryant, a Georgia pastor who led a 40-day “Target Fast” boycott last year. “It shows that Target still lacks corporate diversity among decision-makers and needs concrete internal change.”

In January 2025, the Minneapolis-based corporation ended its three-year DEI program, stopping reports to external groups like the Human Rights Campaign and ended a program focused on carrying more products from Black- or minority-owned businesses, replacing it with a new “Belonging” strategy. The company previously stated that the murder of George Floyd was a catalyst for its DEI initiatives.

Activist-organizers called for a boycott of Target, which lost roughly $12.4 billion in market value by the end of February from January.

Minnesota civil rights activists, separate from Bryant, organized a national Target boycott, resulting in a 33% fall in the corporation’s stock prices and wiped out over $20 billion in market value.

Target’s recent recovery

Target has managed a recovery in recent months. The stock closed at $165.44 on Aug.21, a fresh 52-week high, after a strong second quarter. On Aug. 24, Target’s shares closed at $169.89.

Target’s stock is up roughly 63% this year, a sharp rebound from the depths of the boycott. The company’s turnaround strategy, launched by new CEO Michael Fiddelke after former CEO Brian Cornell stepped down, includes store refreshes, increased staffing, and more aggressive pricing. Target announced a plan in March to invest $2 billion in 2026, including $1 billion in additional operating investments, with plans to open more than 30 new stores this year as part of its path to 300 new stores by 2035, while investing in over 130 planned full-store remodels.

Bruce Winder, a retail analyst and former retail buyer, says that inventory errors, such as the recent Prada and Gucci controversies, “can happen to the best of them.”

“These things happen over time. I think the most important thing for Target, and they did it, was to quickly take accountability, quickly say sorry, acknowledge that this was a big mistake, and pull the product from the shelves,” Winder said.

Boycott organizers undeterred

Target boycott organizers in Minneapolis, including Nekima Levy Armstrong and community leaders Monique Cullars-Doty and Jaylani Hussein, remain steadfast in their divestment calls.

They have continued to press the company to reinstate its DEI programs and continue its commitments to Black communities, and they say the costume controversy is a refresher on their mission.

“It’s important for people to understand that the Target boycott never ended; that it was indefinite from the beginning, unless and until Target reversed course on its decision to roll back diversity, equity, and inclusion,” said Armstrong, an organizer of the national Target boycott.

“What we’ve seen is Target engaging in PR campaigns: recruiting high-visibility Black ministers to meet with them, partnering with (rapper and businessman) Jay-Z, all types of different shenanigans to deflect from the fact that they abandon their commitment to diversity, equity, and inclusion, and capitulation to the Trump administration.”

Hussein, another boycott organizer, said the costume incident was “not surprising.”

“It was almost predictable because of the behavior of Target, and I think this is going to build on the momentum of the boycott. More people are going to be upset,” he said.

“It’s not just an incident with a costume. There are a lot of people right now in this country across many parts who’ve changed their culture of shopping, who have found other places to put their money behind.”

Target Corporation’s stock closed at $163.47 on Aug. 25, dropping 3.78% from its previous close the day prior.

This story was originally featured on Fortune.com

The “ economic onslaught ” that Treasury Secretary Scott Bessent has declared on Iran’s financial connections around the world may have one major caveat: China.

Beijing is Iran’s biggest trading partner and its leading oil buyer. While the U.S. is trying to isolate the Islamic Republic from its remaining economic partners, President Donald Trump also is preparing to host Chinese leader Xi Jinping next month to maintain a fragile trade truce.

Absent from Bessent’s remarks this week were specifics about how the Trump administration would target China, casting doubts on how effective the new campaign would be when the U.S. must balance putting maximum pressure on Iran and avoiding higher tensions with China that could be costly to the American economy.

“The announcement yesterday was very careful in my view to avoid specifics (against China), which could have led to a disruption in the summit,” Edgard Kagan, senior adviser and Freeman Chair in China studies at the Center for Strategic and International Studies, said Tuesday. “For Xi, a state visit to Washington is a big deal; and for Trump, hosting it is a big deal.”

That means both Washington and Beijing will have a dance to do, said Kagan, who served as U.S. ambassador to Malaysia from December 2023 until this February.

“I think the real question is, is there room to push (the Chinese) to reduce what they’re doing with Iran, to put more pressure on the Iranian regime in a way that doesn’t lead them to say, ‘This is unreasonable, and we’re not going to comply,’” he said.

China is likely to wait and see and do the minimum

In response to “Operation Economic Outcast,” which Bessent announced Monday, Beijing said its cooperation with Iran has always been “within the framework of international law.” China receives more than 80% of Iranian oil shipments but usually through indirect channels.

Lin Jian, a spokesperson for the Chinese foreign ministry, said China-Iran cooperation “should not be disrupted or undermined.”

“China is closely monitoring relevant developments and will take all necessary measures to resolutely safeguard its own rights and interests,” Lin said. He repeated China’s opposition to “illegal unilateral sanctions.”

Kagan described Beijing’s remarks as “a holding response” and said Beijing will try to do the least to comply without openly confronting the U.S.

“They’ve tended to be careful not crossing explicit red lines, but they haven’t addressed the spirit of what the U.S. has sought,” Kagan said, pointing to practices such as ship-to-ship oil transfers that obscure the origin of Iranian crude oil to evade American sanctions.

Sun Yun, director of the China program at the Stimson Center, a Washington think tank, said China will not go along with the new campaign if the U.S. goal is to destroy the Iranian economy and seek the government’s collapse.

But “if the goal is to exert enough pressure for Iran to make concessions on the Strait of Hormuz and potentially ending the conflict, I think China can and will demonstrate its cooperation without having to completely sever ties with Iran,” Sun said. “China only needs to do enough to demonstrate it is cooperating, such as cut back on its oil import from Iran.”

With a planned summit between Trump and Xi weeks away, “neither side wishes to have a major escalation bilaterally at this point,” she said. “China needs to give U.S. something, and U.S. needs to understand and accept that it is not going to be everything U.S. asks for.”

US is not likely to act tough on China, analysts say

So far, the Trump administration has refrained from imposing sanctions on major Chinese businesses or banks connected to the U.S. financial system and thus vulnerable to American penalties.

The Treasury Department said Monday that it was penalizing nearly 60 Iran-linked entities, accusing them of involvement in Iran’s nuclear and missile programs, cyber activities and oil shipments.

It targeted a number of entities and individuals based in mainland China and Hong Kong for supporting Iran’s missile and nuclear programs. It sanctioned a China-owned crude oil tanker for transporting millions of barrels of Iranian oil to China this year as well as a Hong Kong-based business for its role in the shadow fleet that ships out Iranian oil.

With Xi’s visit coming up, Trump may not act tough on China now, said Ali Wyne, senior research and advocacy adviser on U.S.-China relations at the International Crisis Group.

“Given how keen Trump has been to maintain both a trade truce between the United States and China and his personal rapport with Xi, he seems unlikely to do a volte-face just a month before Xi’s state visit and adopt a highly confrontational posture,” he said.

Plus, Xi’s visit to the U.S. could pave the way for Trump to return to China in November for the leaders summit of the Asia-Pacific Economic Cooperation grouping.

In his second term, Trump has been less hawkish on China than in his first presidency and frequently touts his good relationship with Xi following a major trade war last year that featured back-and-forth escalating tariffs. The U.S. business community also welcomes Xi’s visit, saying it’s a good sign if the two leaders meet in person, even when any substantial deals may be elusive.

“Beijing is betting that Washington will be reluctant to jeopardize the current leader-level dynamic by targeting major Chinese entities before the summit,” said Craig Singleton, senior director for China at the Foundation for Defense of Democracies, a hawkish Washington think tank.

___

Amiri reported from New York. Associated Press writer Aamer Madhani contributed to this report.

This story was originally featured on Fortune.com

In rural eastern Nebraska, a technology revolution could rise from fields of corn and soybean in the form of new data centers to power the artificial intelligence boom.

But just like around many parts of the country, opposition has been building in ways that defy typical partisan boundaries. Conservative farmers and the state Sierra Club chapter recently found common cause against labor leaders as they filled the firehouse in Murdock — population around 275 — to share concerns about data center development moving too quickly.

The scene exemplified an unlikely coalition that worries about dwindling farmland, declining water supplies and rising electricity bills — not to mention how massive corporations could reshape small town America into nodes in a national network of computing warehouses.

The debate around data center projects is one of the rare modern issues to cut across party lines, demographics and geography — from Republican-dominated Nebraska, Texas and Wyoming to swing-state Pennsylvania to Democratic-leaning New Mexico.

Supporters of data centers, including Republican President Donald Trump and some union leaders who usually back Democrats, praise a potential jobs and economic bonanza, while helping the U.S. thwart China in the geopolitical race for technological supremacy.

But hostility toward developments coming at breakneck speed has surged ahead of November’s midterm elections, and forced national, state and local elected officials to adapt to a shifting political landscape.

Judy Stroy, a fifth-generation Nebraska farmer whose late husband grew corn and soybeans and whose son still does, worried that with “the amount of ground that’s getting gobbled up every year, our food source is in trouble.”

“That should scare everyone,” she said.

Some areas have paused data center expansion

Stroy’s fields are just outside Murdock, a village named for a railroad official at the company that first laid tracks here in the 1890s. There’s a school, a post office, a bar and a Corn Grower’s State Bank branch, plus the Midwest Farmer’s Coop where growers sell their produce and buy supplies like feed and fertilizer.

Murdock is in Cass County, where officials recently approved a 12-month moratorium on allowing data center development.

That might freeze energy company Tenaska’s plans to potentially option 1,300-plus acres of land along Highway 75, south of the nearby villages of Murray and Beaver Lake, in a project that activists fear would pave the way for a data center and its own natural gas power plant.

Tenaska spokesperson Leighton Elise Eusebio said the company has been exploring “the feasibility of developing power generation to serve large loads, whether that is a data center or other form of economic development.”

“We believe there are opportunities to work with public power, communities and other stakeholders to develop power generation that will benefit Nebraskans,” Eusebio said in a statement.

Meanwhile, data center operator CyrusOne has nearly 400 acres under its control outside another Cass County community, Plattsmouth, also along Highway 75, but no development has occurred nor has a project been announced, according to a CyrusOne spokesperson.

The area may be seen as particularly favorable because the highway has been widened, offering easy access to Omaha to the northeast, and to the college town of Lincoln to the southwest. The Missouri River provides a reasonably close, major water source.

Some landowners accepted large checks to sell. Local land typically can go for $10,000 an acre, residents say, and developers sometimes offer four times that, or even more.

Such paydays can be extra enticing as farming has gotten more expensive amid rising diesel prices and soybean exports that have declined in the midst of Trump’s tariff fight with a major market for the crop, China.

“What if somebody offered me $40,000 an acre for my ground? It would be hard to turn down,” Stroy conceded.

Nebraska Sierra Club attorney Ken Winston noted that tech giants have made donations to municipal programs in Lincoln to help mitigate the impact of data centers.

“We need to be aware of the fact that they’re going to happen somewhere,” Winston said. “But communities need to be able to protect themselves and if they don’t want them, they should be able to say no.”

Major project sparks New Mexico divisions

There is no development moratorium in New Mexico that could stop Project Jupiter, a $165 billion data center complex planned near the Mexican border, but concerns over dwindling water supplies are mounting.

“It is not about blue vs. red, or whether we’re a purple state or county,” said Samantha Barncastle Salopek, a longtime water attorney, who is running as a Republican for a seat on the Doña Ana County Commission partly because of the project.

“This is about the local community coming together and saying, ‘Hold on, hold on.’ We didn’t get enough information, and now we’re very concerned because also we have to cut back our water use anyway,” she said.

Barncastle Salopek, whose family grows pecans, said she frequently fields worried calls from farmers, ranchers and local business owners across the political spectrum and in other states.

“When we have to fight over the last piece of the pie, those big companies are going to win,” she said of pressure on remaining water supplies. That’s because New Mexico has been hammered by persistent drought, and large swaths of the Rio Grande have dried up in the central part of the state.

The largest reservoir stands at 1.3% capacity, lows unseen since the early 1970s, even though New Mexico is on the hook to deliver more water to Texas under a recent settlement approved by the U.S. Supreme Court. Activists worry that New Mexico’s pecan, chile and other farms could disappear, taking schools, hospitals and supporting businesses and incomes with them.

Oracle, which announced Project Jupiter, has offered funding for youth programs and pledged to invest $50 million in water and wastewater systems in Doña Ana County.

It also plans to partner with an agriculture technology company to install sensors providing real-time data to farmers to help them use water more efficiently — estimating an eventual savings of 21 million gallons annually.

An Oracle spokesperson called the project a generational New Mexico investment, saying it would generate $4.7 billion in taxes and create more than 7,000 construction jobs and 1,500 long-term jobs while being designed to use little water over 15 years, given its closed-loop cooling and fuel-cell systems.

Still, the Center for Biological Diversity last week filed an emergency petition with the New Mexico Supreme Court, challenging the state’s approval of well drilling to supply millions of gallons of water for Project Jupiter’s construction.

The court on Sunday granted that request, halting authorization to use water from a new well until the legal fight is concluded.

During a recent protest in the state capital of Santa Fe, the demonstrators included Jesus Rodriguez, who traveled four-plus hours from his micro-farm in southern New Mexico.

“The fact is that now the money has taken over,” said Rodriguez, who described himself as disillusioned with American politics. “Now it shows that the Democrats, Republicans, or whatever other party is involved, they don’t care about the land. They care about money. And that’s because they want power.”

Some labor leaders see new job opportunities

Data centers have found an ally in organized labor, including Mike Gage, president and secretary-treasurer of the Nebraska State AFL-CIO. Gage said he personally was worried that some workers are seeing their jobs replaced by AI, but that the centers can also create construction jobs and longer-term positions for people running the finished product.

“This problem is also a very big opportunity for some small communities,” Gage said. He noted that a data center built decades ago in Council Bluffs, Iowa, near Omaha, still employs about 60 people today.

Heather McKenzie, who works in labor relations in Lincoln, said putting more data centers in Nebraska can slow an exodus of skilled young people leaving the state.

“I don’t think that enough people understand that this is going to create really good jobs,” she said. “I think a lot of people view it in a negative sense because it’s being sprung on them.”

Chris Backemeyer, a Democratic congressional candidate in the district that includes Murdock, is a former State Department employee. He said he heard about Saudi Arabia and the United Arab Emirates offering to host data centers for the U.S., which might concentrate technology outside the country.

“That scares the bejesus out of me,” Backemeyer said.

Trump champions data centers, but some Republicans are wary

The president has warned about China’s technology gains and Republican House Speaker Mike Johnson said that a federal pause on data center construction might be a “dangerous prospect.”

Still, most registered voters oppose building a data center in their area, according to a July Fox News poll, including 60% of Republicans and 53% of MAGA voters.

Democratic New York Gov. Kathy Hochul ordered a one-year ban on large data centers. Pennsylvania Gov. Josh Shapiro, a fellow Democrat, said his administration would no longer prioritize data center projects when it comes to issuing construction permits or granting developers lucrative tax exemptions if they don’t meet certain standards.

Republican Texas Gov. Greg Abbott announced a halt on approval of new data centers until audits determine their impact on the state’s electrical grid.

Abbott told ABC News that data centers hadn’t been working in collaboration with state or local officials and “basically dug their own grave for the problem that’s been caused for them, and that’s why they got the backlash they deserve.”

The Republican Party’s Senate campaign arm is worried enough about the issue potentially deciding races in Ohio, that it drafted a memo outlining how to respond. But politicians on both sides of the aisle have turned data centers into a punching bag.

Wyoming Secretary of State Chuck Gray, a Republican running for an open House seat, has pledged to halt their construction.

“If Silicon Valley wants to build their liberal empire, they can do it somewhere else,” he said in an ad.

That overlaps with an ad for the mayor of Scranton, Pennsylvania, Democrat Paige Cognetti, who said “we are not for sale” when it comes to data centers. Cognetti is trying to unseat first-term Republican Rep. Rob Bresnahan in a battleground district.

Trump is undeterred, saying that what data centers mean “for a community is jobs and lots of money and lower taxes.” The president has also conceded, however, that tech companies and developers maybe “can use a little public relations help.”

Seeking to soften fears about rising utility prices, Trump says his administration has urged data centers to build their own power plants.

Yet, during a recent meeting in the conservative East Texas town of Tyler, residents complained of being forced to live in a “gas cloud” so data centers can get enough electricity.

The complaint echoed a similar objection from Kardal Coleman, who leads the Democratic Party in big-city Dallas County. He wrote a recent email decrying data centers for “replacing our jobs, polluting our air, and exacerbating the climate crisis, not to mention the noise disturbing our neighborhoods.”

___

Montoya Bryan reported from Santa Fe, New Mexico.

This story was originally featured on Fortune.com

Iran will not allow military vessels to transit through the Strait of Hormuz under an agreement the Islamic Republic is currently negotiating with Oman, Iran’s Deputy Foreign Minister told state TV late on Tuesday night.

The top diplomats of the two countries met to discuss managing commercial shipping traffic in the vital waterway, which remains largely shut down nearly six months after the Iran war began.

Officials in Thailand said the USS Lincoln was scheduled for a brief stopover in the country as the aircraft carrier prepares to reposition after a lengthy deployment, including support for the Iran war.

Meanwhile, Israel’s strikes continued in Gaza, testing the already fragile ceasefire with Hamas.

Here’s a look at the latest developments in the Iran war and the wider Middle East. Full coverage can be found here.

No military vessels will transit through Hormuz, Iran says

Iran and Oman have reached an understanding on a new framework for managing ship traffic in the Strait of Hormuz that would exclude military vessels, Iranian Deputy Foreign Minister Kazem Gharibabadi told state TV late on Tuesday night.

“If the understanding becomes binding, no military vessel will be permitted to pass through the Strait of Hormuz,” he said. “No military vessel at all.”

Gharibabadi didn’t provide further details about military traffic. He described the proposed commercial route through the Strait of Hormuz, the vital shipping waterway.

Traffic into the Persian Gulf from the Gulf of Oman would pass entirely through Iranian waters, while the outbound route would pass partly through Iranian waters and partly through Oman’s territorial waters, he said.

The Trump administration told Oman it opposed parts of the evolving deal, including the joint Iranian and Omani management of the exit route out of the passage that’s critical to global supplies of oil and natural gas.

U.S. President Donald Trump last week threatened to bomb Oman if it “gets in the way.”

Israel launches deadly strike in Gaza, health officials say

An Israeli strike on a tent for displaced people Wednesday in southern Gaza killed one person and wounded at least four others, according to officials at the territory’s Health Ministry. The Israeli military said it struck a “military operative,” without elaborating.

Even as U.S. efforts to advance the fragile ceasefire in the nearly three-year Israel-Hamas war continue, Palestinians in the strip have reported a surge in deadly Israeli strikes.

Israel has said it targets Hamas and other militants who pose a threat, as well as those who participated in the Oct. 7, 2023, attack that triggered the war. About 1,200 people, mostly civilians, were killed that day in Israel and 251 abducted by militants. Israel’s retaliatory military campaign in the enclave killed more than 73,400 people, according to the Gaza Health Ministry.

The health ministry, part of the Hamas-controlled government, is staffed by medical professionals and maintains detailed records viewed as generally reliable by United Nations agencies and independent experts. It does not distinguish between civilians and militants.

Traffic at Dubai’s airport dropped sharply during Iran war

Dubai International Airport, one of the world’s busiest travel hubs, said Wednesday it saw a sharp drop in passengers during the second quarter as the war with Iran disrupted air travel and prompted many travelers to choose routes that avoided the Middle East.

The airport recorded 13 million passengers in the second quarter, down 31.3% in passenger traffic in the first-half of 2026 compared to the same period last year, with 31.5 million passengers

Air traffic in the United Arab Emirates and elsewhere in the Gulf region was disrupted by shutdowns and restrictions since Feb. 28. The airport itself suffered minor damage in attacks from Iran.

CEO Paul Griffiths noted, however, that passenger volume has steadily increased each month in the second quarter, and said the momentum is expected to continue in the second half.

“The trajectory now is universally positive,” he told The Associated Press.

US aircraft carrier will stop in Thailand as its Mideast deployment draws to end

The U.S. aircraft carrier USS Abraham Lincoln will make a stop in Thailand after its grueling deployment in the Middle East, Thai officials said, without specifying a date or port for security reasons.

The Lincoln has been deployed during the Iran war. Its very lengthy deployment, including a record-setting uninterrupted time at sea of more than 250 days, raised concerns about deteriorating mental health among the crew.

A statement from the Royal Thai Navy on Tuesday said a short stopover by U.S. Navy ships was for recuperation and that no military exercises would be carried out during the visit.

Germany draws down its mission in Iraq’s Kurdish region

Germany has ended the military mission of its advisers in Iraq’s semiautonomous northern Kurdish region after years of supporting the U.S.-led coalition’s efforts against Islamic State militants, the Kurdish region’s Ministry of Peshmerga Affairs said Wednesday.

The ministry said it held an official ceremony at Camp Stephan at Irbil International Airport to mark the mission’s end. Germany had already reduced its footprint in Iraq but had maintained a small contingent at the base.

The end of the German mission comes as U.S. forces have also begun withdrawing from northern Iraq ahead of a Sept. 30 deadline to end their military presence in the country.

U.S. bases in the Kurdish region have regularly come under attack since the U.S. and Israel launched the Iran war on Feb. 28.

This story was originally featured on Fortune.com

Ukrainian President Volodymyr Zelenskyy on Wednesday awarded Elon Musk the Order of Freedom, one of Ukraine’s highest state honors, for his “outstanding personal merits” in protecting human life and freedom and strengthening ties between Ukraine and the United States.

Musk, the founder of SpaceX, has become one of the most influential, and contentious, private sector figures in Ukraine’s war effort. His decision to provide Ukraine with access to Starlink, SpaceX’s satellite internet network, was critical. It remains a communications lifeline for Ukrainian forces and civilians after Russia’s full-scale invasion in 2022.

Starlink has become deeply embedded in Ukraine’s military operations, with commanders and soldiers relying on satellite connectivity to communicate and coordinate operations.

Musk’s decision-making could shape Ukraine’s future military capabilities. Starlink doesn’t currently operate over Russian territory, but Zelenskyy wants Musk to reconsider the policy. That would allow Ukrainian forces to maintain satellite connectivity for longer-range drone attacks targeting Russia’s military and economic infrastructure.

Musk made no immediate comment on the award, which was announced by presidential decree.

Musk has restricted Starlink use in Russia

The technology has been integrated into Ukraine’s drone operations, helping operators maintain links with drones and navigate them over longer distances. Ukraine’s access to Starlink is the main reason for the early successes of its middle-strike drone campaign, which caught Russian forces off guard and undermined their logistics lines earlier this year.

Musk has allowed the Ukrainians to use Starlink inside their own territory, including in Ukrainian regions occupied by the Russians, but has restricted use of the satellite system inside Russia.

SpaceX’s decision earlier this year to accept Kyiv’s pleas and block unauthorized Starlink terminals being used by Russian forces in occupied territory also strengthened Ukraine’s position.

The campaign has put pressure on Russian logistics in Crimea, giving Ukrainian forces a degree of battlefield momentum that Kyiv has not enjoyed since its successful counteroffensives around Kherson and Kharkiv in late 2022.

Musk’s oversized influence over Ukraine

Musk is both a crucial partner for Kyiv and a potential source of vulnerability and dependence. As Starlink has become embedded in Ukraine’s wartime communications, Musk and SpaceX have significant influence over how the technology can be used.

Musk’s decisions can have direct consequences for Ukraine’s military capabilities, making it important for Kyiv to preserve a positive relationship.

Zelenskyy has sought U.S. help in persuading Musk to permit Ukraine to expand its ability to use Starlink beyond Ukraine. Musk has said he believes that would be a significant and dangerous escalation, Zelenskyy told reporters last week.

Musk’s influence has spilled into politics

A reported mix-up over whether SpaceX had approved Ukraine’s use of Starlink for operations inside Russia contributed to tensions between Zelenskyy and then-Defense Minister Mykhailo Fedorov, who was the government’s main interlocutor with Musk.

Fedorov was dismissed in July as part of a broader government shake-up.

Zelenskyy said inaccurate information from the Defense Ministry that SpaceX had approved the bid had contributed to an awkward exchange with U.S. President Donald Trump, who accused the Ukrainian president of deceiving him. He did not say when that occurred.

That contributed to “some of the steps” taken against Fedorov, Zelenskyy told reporters last week, without explicitly accusing him of deliberately providing false information.

Federov’s dismissal also highlights the political element of Musk’s role in Ukraine’s war effort.

This story was originally featured on Fortune.com

Alex Karp isn’t crazy, and he isn’t just talking his book. On July 1, the Palantir CEO went on CNBC and asked the right questions about model providers: who owns the data, where is it cached, is anything transferred back to the provider. Too many people focused on his style and missed his point – they are stealing your alpha.

OpenAI and Anthropic tell commercial customers they will not train on their data. After I recently moved my company off Anthropic, following a Supply-Chain Risk designation, I read the actual agreements and found a hole big enough to drive the entire AI industry through.

The OpenAI Services Agreement says: “OpenAI will not use Customer Content to develop or improve the Services, unless Customer explicitly agrees to such use.” Reasonable, until you check the definitions. “Customer Content” means the Input and the Output. “Input” is what the customer sends the model. “Output” is what comes back based on the Input. Here’s why that’s vaguer than it sounds.

The hidden tokens

Early large language models generated answers token by token without scaling their effort to the difficulty of the question. Humans don’t work that way. Ask someone what 2 + 2 is and they’ll answer instantly. Ask them to plan a family reunion, and they’ll think it over first.

Newer reasoning models do the same thing, spending computation on intermediate steps before returning a final answer, breaking a hard question into parts and working through each one. That intermediate reasoning produces real data on the way to the final output, which raises the question: is it actually capital-O Output, legally? Nothing in these agreements says so.

Who owns what

The definitions matter because another clause ties ownership directly to them. OpenAI’s agreement states that the customer “retains all ownership rights in Input” and “owns all Output,” with OpenAI assigning its interest in Output to the customer.

Picture handing a consultant your confidential financial forecast and asking for next year’s headcount budget. The consultant reasons by filling a notebook with calculations, then hands you a one-sentence answer, and keeps the notebook. Your contract with the consultant covers the question and the answer. It says nothing about the notebook.

That’s what happens every time you use a reasoning model. It generates intermediate reasoning tokens before returning a final answer: a digital scratchpad of facts pulled from your prompt, intermediate conclusions, and newly derived insights about your business. The labs know it’s valuable. OpenAI has said it hides raw chains of thought for reasons that include safety, user experience, and competitive advantage. Anthropic bills customers for full internal thinking even when none of it is ever shown to them.

So you pay for the production of this data. You never see it. And no one will say whether it’s legally yours. OpenAI’s no-training promise covers Input and Output, but the notebook fits neither definition. If reasoning tokens are Output, say so, and extend the protections to cover them. If they’re not, OpenAI has created a third category of data its agreement never defines and never protects.

Anthropic has the same problem. Its API bills for reasoning tokens that never appear in the visible response, and returns the raw reasoning encrypted, so only Anthropic can decode it. So do you own it? This isn’t a drafting oversight. It’s a convenient ambiguity, given how much this data is worth.

Distillation proves the notebook is valuable

The labs can’t dismiss reasoning tokens as meaningless computational exhaust. Their own conduct proves otherwise. Anthropic says DeepSeek, Moonshot, and MiniMax generated more than 16 million Claude exchanges to help train competing models, calling it industrial-scale distillation, a shortcut to capabilities that would otherwise take enormous time and money to build. Labs protect outputs aggressively because outputs transfer intelligence, and raw reasoning tokens are an even richer record of how a model reaches an answer. When OpenAI launched o1, it said this directly:

“After weighing multiple factors including… competitive advantage… we have decided not to show the raw chains of thought to users.”

They want it both ways: bill you for the reasoning, hide it because it’s strategically valuable, and decline to say whether it’s legally your Output.

The fair use playbook

The double standard is hard to miss. This industry was built on the argument that a company can ingest someone else’s protected work, transform it through training, and own the resulting asset. Apply that logic to your enterprise data. You provide confidential material as Input. The model transforms it into reasoning tokens that aren’t identical to your Input but are valuable, newly generated material derived from it. Why would anyone expect the labs to resolve that gray area against their own interests?

As Karp put it, “the jig is up.” You don’t own your alpha. The guy everyone called erratic was trying to tell you. Even Zero Data Retention doesn’t close the gap: both OpenAI and Anthropic offer it, but it requires separate approval and doesn’t guarantee reasoning data gets discarded rather than retained. If not keeping your data takes a special request, retention is the default for a reason.

Which is it?

I expect intermediate reasoning tokens to be assigned to me, the same way I’d want my own notebook back. I may need to reason from it again. The labs haven’t clearly assigned those rights, because doing so means giving up the value. And if reasoning tokens aren’t Customer Content, it’s entirely possible they’re being used to train the next model. Nothing in the labs’ conduct gives me confidence otherwise.

Sam Altman and Dario Amodei can resolve this in a sentence each. Until they do, Karp was right. You own the prompt. You own the answer. They keep the notebook.

Adam Fish is the CEO and co-founder of Ditto, an edge data platform built for unstoppable operations. Its peer-to-peer sync technology keeps applications running with no cloud dependency, in U.S. defense programs including Special Operations Command and for commercial brands like Chick-fil-A.

The opinions expressed in Fortune.com commentary pieces are solely the views of their authors and do not necessarily reflect the opinions and beliefs of Fortune.

This story was originally featured on Fortune.com

Good morning. On Fortune’s radar today:

• Bill Gates says the world is unprepared for AI’s risks: “There is no plan.”
• There is a deal to reopen the Strait of Hormuz, but Trump is silent so far.
• Markets: Nvidia, Nvidia, Nvidia.
• No, the U.S. can’t grow its way out of the national debt, experts say.
• Increased productivity from AI is refusing to show up.
• The insane price of decorating a college dorm room.

➡️ Did someone forward you this email? If you would like to receive this information directly, every morning before the markets open in New York, sign up here.

This story was originally featured on Fortune.com

OpenAI’s CRO position has been in flux for, well, years.

The most recent change came as Denise Dresser, former CEO of Slack, unexpectedly stepped away after less than a year on the job—or so she said on LinkedIn. My colleague Emily Forlini’s sources suggest OpenAI decided to let her go, its fifth C-suite shakeup in the past year, and take the position in a different direction as it races to the public markets in 2027 (apparently lagging behind Anthropic).

We didn’t expect OpenAI to hire Dali Rajic, previously president and COO at Wiz. Though he keeps a low public profile, Rajic, 53, has earned a reputation as one of the most disciplined, successful enterprise sales leaders in tech. He will play a key part in the IPO. This week, Fortune published a profile on Rajic. Here’s what he’s like, as my colleague Emily Forlini writes: 

“Rajic is disciplined, blunt, and intellectual, according to four people who have worked with him previously, two of whom requested anonymity to speak freely about the increasingly influential Silicon Valley figure. Rajic declined to comment for this piece, but confirmed the accuracy of the details included here.

Rajic was born in Yugoslavia, in an area which is now Croatia, and moved to Germany when he was one year old. Growing up, he always dreamed of coming to America, and moved to the U.S. at age 16 for his last years of high school, he said on a 2022 episode of the Grit podcast. He then attended California State Polytechnic University, Pomona for undergrad, followed by an MBA from Northwestern University’s Kellogg School of Management.

“He is one of the most intense people you will ever meet,” one former friend and colleague of Rajic’s tells Fortune. “We would call him the Croatian Sensation. I remember we would get breakfast, and his version of breakfast was on our way to a meeting would grab two hard-boiled eggs at Starbucks, stuff them in his mouth, and be like, ‘That’s breakfast. Let’s go.’”

That good-to-go instinct will likely be vital for Rajic (and OpenAI) in the coming months, as he tries his hand at a seminal, revolving-door job. As Emily writes: 

“It’ll be interesting to see if Dali is there in 12 or 18 months, or if he will burn out like so many other people do at OpenAI,” one source tells Fortune. “That’s the open question.”

And the answer, with the public markets near, matters more than ever. Read more about Rajic here. 

See you tomorrow,

Allie Garfinkle
X: @agarfinks
Email: alexandra.garfinkle@fortune.com

Submit a deal for the Term Sheet newsletter here.

Joey Abrams curated the deals section of today’s newsletter. Subscribe here.

This story was originally featured on Fortune.com

• In today’s CEO Daily: The idea of giving workers more access to the tools of wealth-building is getting traction
• The big leadership story: How Asian healthcare companies are grappling with a fast-aging region
• The markets: Global markets are inching higher ahead of Nvidia’s earnings today
• Plus: All the news and watercooler chat from Fortune.

Good morning. In a week where we’ve been flooded with dubious long-shot ideas—President Trump now wants to rename Lake Ontario as Lake America amid his escalating trade war against Canada—let’s pause to consider an intriguing one. As my colleague Eleanor Pringle reported this week, entrepreneur Mark Cuban wants to address America’s growing wealth inequality by making employers choose between paying higher taxes or giving every member of staff company stock. As the Shark Tank star wrote on X: “It’s exactly what I have done for employees in companies I have started. Most wealthy people get that way from selling their companies or taking them public.”

As a mandate, Cuban’s idea is unlikely to fly: Congress would have to pass a law to raise corporate taxes on companies that don’t grant equity to every employee. That’s not likely in any regime, never mind one in which the president made $2.2 billion last year. But the concept of giving workers more access to the tools of wealth-building are intriguing and getting traction in different ways:

Employee Stock Ownership Plans (ESOPS) are growing in popularity for private companies, in part fueled by retiring baby boomers who want to keep their companies independent without selling to private equity. While politicians may not agree on taxes, they all love employee ownership. The Senate passed two bills last year to encourage ESOPS, of which there were around 6,600 ESOPS, covering around 15 million people in 2023. The federal government first created tax incentives for companies to implement employee ownership in the 1970s when stagflation was rampant and Washington wanted to generate more retirement assets for working Americans. Ronald Reagan loved ESOPS, as does Bernie Sanders.  While they can be expensive, complex and a headache to maintain, ESOPS boast voluntary quit rates that are roughly one-third the national average and workers retire with more than double the savings on non-ESOP counterparts.

Employee Benefits. Companies already offer access to equity grants, restricted stock units, profit-sharing, and stock options.  The problem, as Cuban identifies, is that those tools are often deployed to enrich the best paid people at the company, further widening the CEO-to-worker wage gap. One antidote may be Trump Accounts, which are designed to democratize access to the markets and compounding returns. These tax-deferred accounts, seeded with $1,000 in federal money for every child born during Trump’s second term, have been opened for more than 7 million children since being launched last month. CEOs have been lining up to provide incentives for employees to open these accounts with philanthropists like Michael Dell and Ray Dalio donating funds to help lower-income families fund the accounts for older children. As Dell told me when announcing a $6.25 billion donation with his wife Susan: “When children have accounts like this, their outlook on life just changes.”

But the problem that Cuban identifies is not going away. While pay-transparency laws and talent shortages can create more equitable gains for employees, the reality is that wealth gains remain modest at the bottom and substantial at the top. Women make about 82 cents for every dollar that a man makes, a figure that’s gone down. And affordability has dropped. The most useful tool for some leaders in this environment may be a mirror.

Contact CEO Daily via Diane Brady at diane.brady@fortune.com

This story was originally featured on Fortune.com

Much of corporate America has spent the past 18 months second-guessing and editing itself. 

Many diversity programs were renamed or eliminated entirely. Sustainability language quickly disappeared from corporate communications. Executives became considerably more careful about advancing or even discussing different social impact issues. Under intense political, legal, and regulatory pressure from the Trump administration and conservative activists, companies reconsidered once-celebrated social impact policies and practices.

I have watched this upheaval from inside corporate conference rooms as an attorney and adviser who works in social impact. My clients—many of whom count on me to keep our work private—range from global corporations, founders, and CEOs to their charitable and social impact initiatives, often including partnerships with their peers and other industry stakeholders. 

The specifics of my conversations with clients are likewise bound by confidentiality. But I can tell you that I’ve fielded questions—from executives, attorneys, and impact leaders everywhere from Fortune 500 companies to America’s largest law firms—that would have sounded ridiculous back in 2020: Which words create the most risk? Which commitments still belong in public communications? Which programs remain central to the business? How can we build social impact strategies that withstand a rapidly changing political environment? What will our talent, customers, and other stakeholders think?

Several of the organizations I work with are in genuinely precarious positions at the moment. One of them came close to being legislated out of existence by Congress. Others have been bracing for years about growing scrutiny over inclusion initiatives. (Naming them in a national business publication could put a target on their backs, so I won’t.)

At the same time, even as good corporate citizens navigate these turbulent waters, it feels like the tide is beginning to turn. By that I mean, corporate America’s anti-woke retreat appears to be approaching its limits. And a new model of corporate purpose is emerging in its wake: more legally disciplined, more closely connected to business strategy, and more measurable and designed to survive political change. At its core, it is still deeply rooted in two realities: that caring about all people is inherently part of doing good; and that doing good is good for business. 

Consider what shareholders are actually saying, as assessed by Harvard Law School Forum on Corporate Governance. Through May, conservative activists had filed 43 anti-DEI shareholder proposals, dwarfing the number of proposals supporting DEI. Yet the 22 anti-DEI proposals that reached a vote received an average of roughly 1 percent support. Looking across environmental, social and governance issues more broadly, anti-ESG proposals averaged about 1.7 percent support, compared with about 13.3 percent for proposals supporting ESG-related actions or disclosures.

The same data show declining enthusiasm for many prescriptive pro-DEI proposals. Investors appear increasingly selective about how companies address these issues. They also appear to have remarkably little appetite for the anti-DEI agenda being offered in their name. Additionally, new research out of University of California at Berkeley’s Goldman School of Public Policy shows that firms that either kept their DEI policies or voted down anti-DEI shareholder resolutions have performed just as well financially as those that didn’t.

An even stronger signal arrived this summer. Benevity, which provides corporate giving and volunteering technology, surveyed 420 corporate impact professionals for its latest State of Corporate Purpose report. Seventy-eight percent reported that their organizations had continued their purpose work as before. Among respondents from large companies, the figure was 57 percent. Meanwhile, 69 percent said their organizations had changed how they described their programs publicly.

That distinction matters. While the public language of corporate purpose changed dramatically, much of the underlying infrastructure survived. Put another way, you can argue, as Jones Day’s Robert Profusek did in Fortune earlier this year, that “social purpose stakeholder capitalism” may have gone too far too fast; and at the same time, you can acknowledge that there is important business value in corporate impact initiatives. “Most companies support key ESG objectives already, recognizing that they are essential to the operation of any company positioned to succeed in the 21st century,” Profusek wrote. “ESG considerations are important means to an end, not an end of themselves no matter what the loudest voices on electronic and social media might say.”

Corporate leaders (and their lawyers) are scrutinizing language, eligibility rules, legal exposure and public communications far more carefully. They are also asking harder questions about which initiatives serve employees, customers, communities, and the business itself. These conversations increasingly sound like strategy discussions. That evolution will likely make corporate purpose more durable for the years, headlines, and headwinds ahead.

Some of the companies illustrating this point hardly fit the stereotype of progressive corporate activism. Chick-fil-A, long associated with conservative Christian culture, still maintains a webpage explaining how it “values diversity, equity and inclusion,” using that exact phrase, which it either admirably or mistakenly never scrubbed from its website. And at a time when talking about environment or conservation invites increased scrutiny, Bass Pro Shops still proudly describes itself as “United for Nature” and continues to declare that it is leading North America’s largest conservation movement.

This makes sense. Corporate purpose has always belonged—and still belongs—across the political divide. Humans care about their fellow humans, as well as our shared planet. Companies have employees to attract, communities to operate in, customers to earn, stakeholders to serve, reputations to protect. Human goodness is a constant, and humans show up more fully in workplaces and marketplaces that serve the innate demand for good, even if we reasonably disagree on the edges of what constitutes appropriate corporate citizenship.

At the other end of the corporate spectrum, Anthropic recently demonstrated just how consequential a company’s impact commitments can become. The artificial-intelligence titan resisted Pentagon demands concerning uses of its technology that included mass surveillance and autonomous lethal weapons. The dispute eventually led President Trump to order federal agencies to stop using Anthropic technology and produced an extraordinary confrontation between one of America’s fastest-growing AI and technology companies and the federal government.

Whatever one thinks of Anthropic’s particular limits, the episode illustrates a larger point: Corporate principles can remain operational even when adhering to them becomes expensive.

The commercial incentives remain powerful, too. Edelman’s 2026 Trust Barometer, based on nearly 34,000 respondents across 28 countries, confirms that employers are particularly well positioned to build trust among people with differing values and perspectives. Its 2025 consumer research found that 64 percent of respondents choose brands based in part on their beliefs and 68 percent consider it highly important for brands to make them feel positive emotions such as confidence, inspiration or safety.

None of this requires a return to the controversies of corporate activism of the early 2020s. There is ample opportunity for refining and growing a more settled, durable approach to impact for the rest of the decade to come. 

The next generation of corporate purpose can be more disciplined. Companies can choose issues connected to their businesses and stakeholders. They can comply rigorously with civil-rights laws. They can measure results. They can explain why a particular investment belongs in their strategy. They can approach employees and customers as politically diverse human beings whose trust has to be earned, who may disagree on matters of style and degree, but not on the substance of caring for our fellow human, fellow colleagues, and fellow consumers.

This is also where the anti-woke backlash may ultimately prove surprisingly useful. It subjected corporate purpose to a stress test. Some initiatives proved legally vulnerable. Others lacked a clear connection to business strategy. Some corporate pronouncements outran the work behind them. Stronger programs survived because leaders could clearly explain why they existed and what they accomplished.

America’s political winds will keep changing. A company that rebuilds its values every four years will eventually exhaust the trust of employees, consumers, investors and communities alike.

Corporate leaders now have an opportunity to design their social impact strategies for durability: grounded in law, connected to business, supported by evidence and broad enough to serve stakeholders who see the world differently.

The next era of corporate purpose will be built to survive the next election and thrive for decades.

Scott M. Curran is a social impact attorney, professor, strategic adviser, and the author of Better Good: A Simple System for Creating Lasting Impact, published this month by Simon & Schuster.

The opinions expressed in Fortune.com commentary pieces are solely the views of their authors and do not necessarily reflect the opinions and beliefs of Fortune.

This story was originally featured on Fortune.com

SpaceX CEO Elon Musk described plans to build factories on the moon as he presided over the now public rocket maker’s first earnings call this month. Even Musk admitted it all sounded “totally nuts.” But while Musk’s rhetoric and his timing estimates often don’t line up with reality, he’s not alone in seeing opportunity on the moon. Now, a new report is offering a look at the financial underpinnings of what could become a vast $566 billion lunar economy, and the ripple effects that could double it to more than $1.1 trillion.

The new report by advisory firm Deloitte and shared with Fortune in advance of its Wednesday release, estimates the budding moon-based economy could generate between $343 billion and $566 billion in upside through 2050. The range represents a conservative- to accelerated-growth scenario based on how quickly infrastructure, energy, and transportation services can get up and running on the moon, and then how quickly commercial businesses follow—which is no small feat. The most bullish scenarios also hinge on whether several bleeding-edge technologies advance over the next two decades, including rocket fuel made from water ice at the lunar poles, extraction of helium-3 to cool quantum computers, and AI data centers built to orbit the moon’s surface. 

And despite the high barrier to entry for non-aerospace businesses, new companies are getting in on the action. Luxe fashion brand Prada used its textiles expertise to help design spacesuits that can withstand extreme temperatures. Sunglasses brand and optical manufacturer Oakley developed a gold-plated visor for astronauts for use in both darkness and under direct exposure to the sun.  

“What was once the domain of governments and a handful of aerospace contractors now includes venture-backed startups, investors, defense firms, and some of the world’s largest companies all seeking a role in the emerging lunar economy,” the report states. “Although still in its earliest stages, the upside potential could be massive.”

The report, “Building the Lunar Economy,” includes insights from interviews with founders, engineers, investors, and government officials, and more than 400 model inputs. It maps out the infrastructure required for working on the moon, including the transportation, energy, communications, surface mobility, and life support needed—and it considers the unpredictable innovations and value that could be unlocked when humanity enters this new realm.

Why now?

Brett Loubert, who leads Deloitte’s space practice and co-authored the report, said momentum in the sector took on even greater intensity last year, which was also a record for venture capital investment in space technology. Seraphim’s space tracker report saw $7.5 billion investment in Q2 of 2026, with trailing 12-month investment at an all-time-high of $23 billion. 

“What you’re seeing generally is excitement in and outside the industry for what is an explosion of data sources and services that are being delivered from orbit and beyond,” said Deloitte’s Loubert. 

Another major catalyst for the commercial momentum is inarguably SpaceX itself, which has Musk behind it as founder, CEO, and hype man for both the company and for space as a mainstream concept. SpaceX went public in a record-setting IPO on June 12, which saw its valuation soar to $2 trillion. Its market cap has since slipped to $1.8 trillion but Musk, who controls the majority of SpaceX through his ownership, has a knack for drumming up interest from retail and institutional investors in his companies and SpaceX has not, thus far, shown it will be any different. 

In the weeks before it went public, SpaceX headlines focused on the company’s interplanetary Mars mission and Musk’s goal of establishing a human colony on the planet with 1 million inhabitants. But weeks after the largest IPO in history, Musk talked more about the moon during SpaceX’s first earnings call, which investors say is a strategic threshold that will likely be crossed before SpaceX can fully set its sights on Mars. 

SpaceX has already invested more than $15 billion in its massive Starship rocket, designed to carry up to 100 metric tons to orbit and, eventually, the company hopes, transport passengers and equipment to Mars. On the most recent earnings call SpaceX President Gwynne Shotwell laid out near-term milestones that included an Artemis III ship docking in 2027 and “boots on the moon in 2028.”

Ex-SpaceX employees have also struck out and forged their own space mobility and infrastructure businesses with 141 companies cropping up worth $10.6 billion, according to Forbes. An analysis of publicly disclosed equity rounds by space companies between August 2025 and July 2026 found 47 deals with a median round size of $40 million, and an average round size of $116.2 million, according to New Market Pitch. Spacecraft manufacturers dominated, raising $2.3 billion in 26 deals and representing 43% of the capital raised. 

Musk meanwhile, continues to add his unparalleled brand of boosterism to the space sector, moving SpaceX’s internal projection for hitting $1 trillion in revenue forward from 2031 to 2030 during the SpaceX earnings call (much of the lofty revenue target is based on the company’s AI business accelerating).

The Deloitte Numbers

The Deloitte report’s $566 billion high-growth scenario breaks down into two value pools. The first, “core lunar activity,” refers to the foundational lunar infrastructure that has to be built before anything else can really happen. Getting to and from the moon, transportation and energy once you’re there, and dealing with the surface regolith—the moon’s jagged specks of dust—make up $206 billion of that total, more than a third. Power tacks on another $44 billion. Communications, surface mobility, construction, and life support make up the rest of the first pool, which Deloitte estimates could generate an estimated $282 billion through 2050.

The second value pool refers to “enabled activity,” and describes potential downstream markets that could unlock to the tune of $284 billion because of the infrastructure, power, and transportation foundation from the first pool. Under the accelerated-growth scenario, the market for new resources and materials could grow to $114.5 billion, primarily driven by rocket propellant made from lunar water ice and helium-3. The latter is a rare isotope embedded in lunar soil that could become a major fuel source and has a use for cooling quantum computing systems. The report values in-space production in the high-growth scenario at $105.9 billion.

As for fuel, the further out in space you want to go, the more fuel you need, and figuring out how to extract lunar propellant in space is one of the major challenges companies are trying to solve. A single kilogram of rocket fuel costs $1 on Earth, $4,000 in low Earth orbit, and $36,000 on the lunar surface if it goes up from home, the report states. On the other hand, water ice in lunar soil can be processed into liquid oxygen and liquid hydrogen, which can be used as rocket fuel. 

“Water ice is the oil of the moon,” economist Jim Zukin told Deloitte, per the report. It goes on to state: “If unlocked at scale, it could do for space what gas stations did for the road: reduce the cost of existing trips, and make entirely new ones possible, creating a network linking every station, city, and state with fuel and a delivery infrastructure.”

While lunar propellant is nowhere close to scaling, the case for compute is moving much faster. Google has announced its Project Suncatcher, which would put satellite clusters in orbit for compute, and Nvidia-backed Starcloud launched a satellite last year with an H100 Nvidia chip built for space which it used to train an AI model in orbit, the report notes. SpaceX has asked the FCC for permission to launch 1 million satellites to support data centers in space. The thinking is that orbital data centers can leapfrog past some of the ground-level issues such as where to build them, and upset communities that don’t want to host data centers, despite their municipalities making deals for construction. 

“Data-processing satellites benefit from space’s unique conditions, offering near-continuous access to solar energy and radiative cooling,” the report states. “However, the challenge is scale.”

Things We Haven’t Imagined Yet

Beyond core lunar activity and enabled activity, there is also innovation spillover, human inspiration, and the value of research and potential opportunities that haven’t been imagined yet, said co-author Raquel Buscaino, who leads Deloitte’s Novel & Exponential Technologies team. 

Deloitte pins the unimagined cascade effect at an additional $541 billion and did not add it to the final estimate to avoid overstating the size of the lunar economy. However, taken with the high-growth scenario, this pushes the lunar economy above $1.1 trillion, although there is a healthy dose of uncertainty to go along with that figure. Buscaino said uncertainty is a feature and not a flaw when it comes to space.

“There is extraordinary possibility and it’s also extraordinarily hard to do those things,” she said, referring to the opportunities the report lays out. People are excited because there are early demand signals on infrastructure and on the early markets, Buscaino added. 

“Some of these distant opportunities could have extremely large upsides, and we don’t know which ones will pan out, or if what will ultimately matter the most could be a market that we haven’t imagined yet.”

This story was originally featured on Fortune.com

During the past couple of months as you’ve scrolled through social media, you’ve likely come across posts with links to teacher wishlists for items for their classroom. That’s because most teachers have to pay for supplies, decor, and other classroom necessities out of their own pocket.

But one nonprofit foundation stepped in to help ease that burden. 

Exponential Scholars announced Monday it funded every open North Carolina project on the DonorsChoose website, which connects teachers in high-need communities to donors for the supplies kids need. The group’s $2.25 million donation will cover more than 3,200 classroom projects across more than 1,000 schools in North Carolina. 

Exponential Scholars was founded in 2025 by self-proclaimed “quiet philanthropist” Holly May, who serves as its president and CEO. She grew up in rural North Carolina in the 1990s, where her hometown had a population of just 2,000 and only one elementary school. 

A full financial-aid scholarship to an independent school, followed by a full ride to Harvard College, and a business career in San Francisco, shaped May’s belief that “good fortune comes with a duty to pay it forward.” She and her husband, Travis, returned to North Carolina to raise their family and launched the foundation to support gifted students in overlooked communities across her home state.

This gift comes at a time when the amount teachers spend continues to grow. According to a 2025 study by consumer platform CouponBirds, teachers in the state spend an average of $1,632 of their own money each year to stock their classrooms. That’s a 22% year-over-year jump, and the second-highest figure in the country. Pennsylvania beats that total by $5, and the national average was $1,021. In that same survey, 95% of teachers reported dipping into their own wallets for classroom materials, and 82% said their classrooms would suffer without it.

Meanwhile, teacher pay hasn’t kept up. North Carolina’s average teacher salary for the 2023–24 school year was $58,292, ranking the state 43rd nationally and landing nearly $14,000 below the national average, according to the National Education Association. The Economic Policy Institute estimates teachers earn just 73 cents for every dollar made by similarly educated professionals.

This practice of clearing teacher supply wishlists is called “flash-funding,” and another notable example came in 2015, when Stephen Colbert teamed up with two organizations to cover roughly 1,000 South Carolina classroom projects with an $800,000 donation. Colbert has served on DonorsChoose’s board of directors.

Three years later, cryptocurrency company Ripple funded every open project on the site nationwide, some 35,647 campaigns, with a gift valued at $29 million.

DonorsChoose was founded in 2000 by former teacher Charles Best, and has become one of the biggest pipelines for classroom needs across the U.S. More than 6 million supporters have contributed nearly $2 billion to more than 3 million projects, supporting more than 92,000 schools. In North Carolina specifically, more than 143,000 projects have been funded, according to the DonorsChoose site.

For the North Carolina teachers whose requests were sitting open this week—for books, pencils, lab equipment, and more—the gift means those supplies are on their way. Now they don’t have to tap into their own savings to cover necessary supplies for their job.

“I still believe in the American Dream,” May wrote. “Mine came true: work hard, use your gifts, and it pays off…and you leave your kids more room for ambition and optimism than you had.”

This story was originally featured on Fortune.com

Learning to read and write cursive was once a rite of passage for practically every American schoolchild.

Until the early 20th century, children in the U.S. were generally taught to write in cursive or script, rather than the print writing typically taught to young children today. Some people at the time thought a person’s character could be evaluated based on the quality of their cursive.

In the first half of the 20th century, penmanship was often taught as a distinct subject in American elementary schools, where children could spend as much as 45 minutes a day on handwriting instruction and practice.

Students during these years generally learned print in kindergarten, first and second grade. Cursive was typically introduced in third or a later grade, in part because some child development researchers at the time believed that writing in print was easier for young children to learn and aligned with the fonts they found when reading.

I am a product of this system, learning cursive in third grade and winning awards for my excellent cursive penmanship from the retired nuns who once worked in my Catholic school. Now, I am a clinical professor of literacy studies and a teacher-educator, who researches reading and writing.

My experience is no longer typical, though, as cursive instruction has become less popular over the past few decades. Some states and schools, though, are reconsidering the value of cursive for today’s students.

A shift away from cursive

By the 1980s, the instructional time dedicated to cursive instruction in some schools had been reduced from 45 minutes a day to 30-60 minutes per week.

And the time allotted kept falling, in part because of the 1990s standards-based movement, which created school curricula overcrowded with content.

The new math standards called for data analysis and probability to be incorporated into mathematics education beginning in the early grades, including kindergarten. The social studies standards, meanwhile, reflected a broader approach to history that brought greater attention to figures such as Mansa Musa, the leader of the Mali empire in the early 1300s.

At the same time, computers became more widely available in schools and in homes, making the ability to type on keyboards seem like a necessary skill.

Cursive began to seem antiquated.

The Common Core state standards, a nationwide set of academic learning standards for K–12 students, were adopted in some fashion by 45 states between 2009 and 2014. These standards called for kids to learn how to type on keyboards, ensuring that whatever valuable teaching time was left for transcription would be spent on learning how to type.

A return to cursive?

It did not take long for the pendulum to begin to swing back in favor of cursive in the early 2010s.

Framed as a back-to-basics issue, North Carolina passed the first cursive law in 2013. It required students to read and write in cursive by the end of fifth grade.

The law also required children to memorize their multiplication tables.

By 2015, 14 other states had some kind of cursive instruction requirement on the books.

As of 2026, more than 25 states, including Pennsylvania, Florida, California and Michigan, require cursive instruction.

Lawmakers cite several reasons for mandating cursive in schools.

Pennsylvania and Florida legislators argue, for example, that an informed citizenry needs to be able to read historical documents, which are often written in cursive script.

In 2025, a Michigan representative suggested that reading cursive was also essential to preserving family history. Her granddaughter was able to read a letter that her deceased father, the lawmaker’s son, had written in cursive only because her school still chose to teach the skill.

A California assembly member had more modern realities in mind, suggesting that cursive could be used to ensure students aren’t using artificial intelligence to complete their schoolwork.

Benefits of cursive

Being able to write in cursive can have benefits for learners.

A 2020 study conducted in Norway hooked up 12 adults and 12 preteens to brain-imaging devices while they wrote in cursive, drew and typed.

They found that writing in cursive activates the brain waves and neural pathways used for learning and for memory. Learning to write in cursive is also associated with better outcomes for learners with dyslexia and dysgraphia, a neurological learning disability that makes writing difficult.

Other studies have shown cursive’s impact on writing fluency and quality. Some studies show that cursive can help people write faster, while other studies have found that children writing in cursive may not have a speed advantage.

First graders in Portugal who were taught cursive wrote in longer “bursts,” which the researchers attributed to the unbroken flow of cursive writing.

There’s even research suggesting that cursive gives an early boost to students’ reading skills, not just their composition skills. A 2019 study of 141 first graders found that children taught cursive read better by the end of the school year than classmates who weren’t.

One theory is that learning cursive trains the brain to process letters more effectively.

Cursive in today’s world

Not everyone is convinced cursive deserves a comeback tour.

For starters, the science is shakier than cursive’s champions let on. There’s solid evidence that handwriting beats typing for learning and memory. What’s much less clear is whether cursive specifically has an edge over ordinary print.

It is possible that much of the recent brain research getting cited in statehouses from Pennsylvania to California was done on kids printing their letters, not looping them together.

Still, teaching cursive in today’s schools is not necessarily simple.

Many of today’s teachers, and many of tomorrow’s, never learned cursive themselves, which means states are now asking educators to teach a skill they don’t have. Then there’s the question of time. Teaching cursive properly can eat up roughly 70 minutes a week, as children learn to master the skill. How does this fit into busy school days? Also, not every kid is going to love learning cursive.

Just ask any teacher who’s tried to get a room full of 7-year-olds excited about learning how to properly form a capital G in cursive.

Mary Jean Tecce DeCarlo, Clinical Professor of Literacy Studies, Drexel University

This article is republished from The Conversation under a Creative Commons license. Read the original article.

This story was originally featured on Fortune.com

Luna, 26, no longer speaks to her father. There was no single betrayal, no dramatic rupture. What ended the relationship was his insistence on obedience and dominance.

“In his head it’s like, ‘I’m the father, you’re the daughters, we do things my way,’” she told me. Luna explained to him that she was “not willing to have a relationship like that,” and her terms for contact were simple: “We could only have a relationship where you understand that I’m a person, and I’m not just gonna do whatever you want me to do whenever you want me to.”

He refused to accept those terms. So after numerous attempts to make it work, she walked away.

Family estrangement has become a fixture of talk shows, advice columns and social media. It’s also the subject of my new book, “Families We Lose.”

Much of the current media coverage on estrangement tends to frame it as an epidemic that is the product of therapy culture.

My research suggests that this framing misses what is actually happening. I conducted 68 in-depth interviews with adults who went no-contact with a parent, sibling, grandparent or other relative. (All participant names, including Luna’s, are pseudonyms used to protect their privacy.)

I found that estrangement is less a story of trauma or selfishness, and more a clash between competing expectations: an insistence on loyalty, duty and forgiveness on one hand, and the longing for respect, emotional safety and personal growth on the other.

Falling birth rates raise the stakes

Estrangement is fairly common.

In a 2022 study, my colleagues and I used national survey data that followed parent-child relationships over several decades. In that period, 26% of adult children in the U.S. experienced a period of estrangement from their fathers, and 6% reported a period of estrangement with their mothers. Our findings were consistent with what sociologist Karl Pillemer found in his 2020 national survey, which showed that 27% of Americans were estranged from a relative, or roughly 67 million people.

Headlines often treat numbers like these as evidence of a sudden crisis. But no study has tracked estrangement rates over time, and nothing in the existing data shows that it is skyrocketing.

Yet estrangement can have different consequences today. People have far fewer children than in most of human history, so the loss of one relationship may carry greater weight. Two generations ago, a parent might have had several other children with whom to maintain close ties; today, many parents have only one or two.

Smartphones, meanwhile, have created expectations of constant contact and connection within families. Even adult children who have quietly distanced themselves by moving away or limiting contact may still find themselves fielding regular requests for connection through texts, phone calls and social media. This can be stressful, emotionally taxing and make it difficult to establish boundaries.

Those I interviewed often had difficult childhoods, and some had experienced abuse. But they rarely pointed to childhood as the proximate cause of the estrangement, instead citing the poor quality of their relationships with parents and other relatives in adulthood.

‘Kinship culture clash’

For them, the path to estrangement was usually gradual, marked by mounting tensions and unresolved conflicts. It emerged out of the recognition that a parent or sibling would not, or could not, treat them the way they desired. For most, this looked like accountability and growth, intentional connection, mutual respect and safety.

Molly, 36, described a lack of accountability as motivation for her estrangement from her mom, Cindy, a decade ago.

Molly told Cindy: “You’ve done all these horrible things. I just need you to acknowledge it. I feel like I’m owed an apology.”

Molly felt that her desire for accountability was a basic need of any relationship, noting, “I didn’t feel like that was asking too much, but apparently it was. I put the ball back in her court, and she’s just sitting on it.”

Most of the people I interviewed recognized that they had different definitions of what family should look like and feel like. In my book, I call this a “kinship culture clash.”

Divorce offers a familiar example of this kind of shift.

For most of American history, marriage was an institution shaped by economics, law, religion and social pressure. You stayed married because that was what one did, and there were few alternatives that didn’t leave one spouse – usually the woman – socially ostracized and financially destitute.

Then, as sociologists Andrew Cherlin, Anthony Giddens and Stephanie Coontz have argued, marriage transformed into a relationship judged by its ability to provide love and support, promote ideals like fairness and lead to personal growth. Once a marriage started being thought of as an arrangement that needed to work for each spouse, divorce became thinkable, then ordinary.

Should family ties always bind?

The family you are born into can carry the same tension.

Ties to parents and siblings are often expected to be permanent and unconditional, reinforced by cultural norms and laws that bind parents and children to one another, even when children become adults. For example, in adulthood, 27 states have filial piety laws that require adult children to bear financial and, in some cases, direct caregiving responsibility for aging parents. This is part of what I term “compulsory kinship.”

Under this framework, you show up at Thanksgiving no matter what was said last year, because your family is your family. You take your dad into your home to care for him when he gets sick, because he’s your father.

This model does help many Americans. It can ensure aging parents have adult children to care for them, while adult children may have a place to call “home,” no questions asked.

In contrast, the estranged people I talked to for my book adhere to what I call “democratized kinship,” and it generally reflects views of contemporary marriage.

Family bonds require mutual respect, reciprocity and support in order to continue. Under this model, genetics take a back seat: Anyone can become family, just like anyone can become a spouse, but no one is automatically family.

Neither family framework is new, and neither belongs to a single generation. Some people have always held family of origin relatives to distinct standards of care. Queer people have been building chosen families for generations.

But what I chart in this book is the conflict between these two ways of thinking.

Estrangement often happens when these two ideas of family collide and cannot be reconciled – when one member insists that biological or origin ties and hierarchies reinforce the family unit, and another insists that behavior does. That is Luna’s story. Her father drew on the logic of age and hierarchy – “Because I’m your father and I said so” – and she demanded to be treated as a person first and a daughter second.

A family that works for you

People who leave their families of origin are not rejecting family altogether.

In a recent study drawing on these same interviews, I show that estrangement operates as a catalyst for what I call “redoing family.”

After cutting off contact with a family member or multiple relatives, people commonly went on to rebuild kinship networks, drawing friends, partners, in-laws and neighbors into roles once reserved for blood relatives, and constructing new family units organized around support and reciprocity.

Luna is not on her own because she no longer talks to her father. Her family today includes her sister, whom she talks to multiple times a day, and a large group of friends.

“I think you can have your chosen family, and whoever in your biological family you connect to,” she told me. “Everybody just gets to choose who they feel they have a real, intimate, deep connection with. And that’s your family.”

She didn’t want less family. She simply wanted a family that works.

Rin Reczek, Professor of Sociology, The Ohio State University

This article is republished from The Conversation under a Creative Commons license. Read the original article.

This story was originally featured on Fortune.com

If given the chance to gain thousands of admirers for your work, would you take it?

The artists who my colleagues and I interviewed as part of our research on content creators did – and many went on to question whether it was worth it.

Mark, an illustrator with millions of followers, revealed a deep sense of dread about what was once his dream: to make a living creating art and sharing it online. (We used pseudonyms to protect the privacy of the people we interviewed.)

“The thing that I got right, that I loved for the longest time and now has almost become the bane of my existence, is creating relatable comics,” he told us. “Now, I’m stuck … I would rather die than make relatable comics, but I’ve built an empire of relatable comics. So what the (heck) am I supposed to do with my life?”

Ironically, the very thing that Mark felt he needed to fulfill his dream – a large, admiring audience for his illustrations – began to feel like a prison. No longer a distant or abstract entity, he now felt inextricably linked to his followers and their reactions to his work.

As an organizational psychologist, I’m convinced that Mark’s existential angst is not unique, nor is it relegated to content creators. Rather, it has profound implications for anyone who cares about the future of work.

Although the strange oppressiveness of collective adulation isn’t new, it is becoming far more ubiquitous – and insidious – in the attention economy.

A ‘chamber of despair’

The pursuit of fame for fame’s sake has become an almost banal pursuit in the attention economy. Content creation is, according to some estimates, a US$300 billion industry. Goldman Sachs projects it to grow to half a trillion dollars within the next few years.

Becoming a content creator has officially elbowed out old standbys like doctor and astronaut as the job that Gen Z and Gen Alpha most aspire to have. Countless online courses now promise to teach the ropes of content creation, while Arizona State University will be offering a new content creation major.

And yet, the realization of this goal often comes with unanticipated psychological costs.

The 54 creators we interviewed as part of our multiyear research project – visual artists and musicians with massive online followings – repeatedly described the unrelenting sense they had of being oppressed or restricted by their audiences, using phrases that involved contorting themselves into two-dimensional versions of themselves.

One artist bemoaned that she had to “flatten and suppress her real self” to engage her audience. An illustrator revealed that she had felt like a “crumpled-up ball of paper” for years. An otherwise chipper cartoonist likened her audience to a “chamber of despair” that nobody else understands.

Over time, we came to think of this condition as “audience entanglement.” It’s a state in which creators become so psychologically intertwined with their followers that the audience begins shaping not just what they make – a phenomenon that has been referred to as “audience capture” – but also the meaning they derive from their work and how they understand themselves.

Audience entanglement plagued creators from all walks of life, regardless of age, gender or nationality. It struck musicians and illustrators, YouTubers and Instagrammers, self-proclaimed anxious wrecks and the freest of spirits. No single personality type or demographic was immune to the experience of a deep enmeshment with one’s audience, and to its profound consequences for their creativity and well-being.

Suffocated by a pressure to perform

Becoming entangled with one’s audience meant that followers’ reactions – whether positive or negative – often started to feel like a burden.

Creators described feeling hypersensitive to how their posts performed, and spending more and more time each day monitoring them.

Then there were the comments. As Selena, an illustrator, put it, a single mean message “would ruin” their week. While they recognized that they needed an audience to do the work they loved, managing it began to feel like a job in and of itself – and not one they signed up for.

At their lowest points, the creators we spoke to were not disrupted merely by the errant hostile comment. They described being unable to take in even the most effusive messages from their audience – the types of affirmations that had initially pushed them to begin creating in the first place.

One wildly successful creator, Maya, recalled receiving direct messages from fans that her art was the only thing getting them through death, divorce and other catastrophic life events. Maya experienced this profoundly positive feedback as being harder to handle than critique:

“I didn’t know how to really hold people’s pain in a meaningful way, and I was also getting really deep criticism – and, in a way, I could almost deal with that better than praise. It’s just a lot for a person to take on.”

As a result, many creators began to view this creative path they had once wanted more than anything as totally unsustainable. There was a tragic irony to this pain. The very thing that made their creative work possible – a large, mostly admiring audience – was also its biggest threat. The audience that allowed them to monetize their creations in the first place had now led to creative paralysis.

Healthy boundaries, intentional engagement

Not everyone we spoke to remained forever imprisoned by their so-called fans. For some, the intense enmeshment they experienced eventually morphed into something healthier.

The creators who shifted toward this dynamic did not stop caring about their audiences. Instead, they learned to develop clear boundaries around audience interactions and clarity on who they wanted to be online – what we called “entanglement management strategies.” By limiting time and energy engaging with the platform, they were able to escape the prison of their fans and once again find meaning and a sense of authenticity online.

As Charlie, an illustrator, explained:

“There is something about hitting this true internal place where, it’s kind of a mystical thing, but some balance between being yourself internally and being yourself externally. There’s some sweet spot there where you know what your audience is and what they expect, and you know where you as a person or creative person overlaps.”

Having spent the better part of two decades researching the psychology of work, I believe that audience entanglement doesn’t just happen to full-time influencers.

The ability to cultivate and manage an online audience has become an essential skill in more traditional professions, like journalism, politics and higher education, with large followings sometimes becoming a prerequisite for, versus a consequence of, success.

Most people will never have a thousand – let alone millions – of followers. But they’ve probably felt some hints of audience entanglement in their relatively banal online lives: the addictive siren’s call of rechecking a post as the “likes” roll in, or an idyllic afternoon with your kid ruined by a snarky comment from an internet stranger.

While countless influencers offer tips for growing online audiences, little training exists for how to actually deal with managing the stress of having large followings.

Previously, only the truly famous – actors, rock stars, pro athletes – were at risk of experiencing the anxiety of public scrutiny.

Now, many of us are just one viral post away from its grasp.

Julianna Pillemer, Associate Professor of Management and Organizations, New York University

This article is republished from The Conversation under a Creative Commons license. Read the original article.

This story was originally featured on Fortune.com

Pope Leo XIV warned that artificial intelligence risks becoming a new form of “economic colonialism,” deepening the gap between wealthy and poor nations, and said algorithms are already creating “a subtle form of domination” over who gets seen and heard.

“We must remain vigilant in this regard,” Leo told the network of officeholders who make up the International Catholic Legislators Network (ICLN) on Friday. “Lest innovation become another vehicle for ideological or economic colonialism.” He warned that AI’s rapid development risks leaving poorer countries increasingly dependent on wealthier ones for the technology.

He went further, describing what he called “a subtle form of domination when algorithms decide who is seen, and who remains invisible, when digital platforms shape public discourse without accountability, and when the dignity of workers is subordinated to the optimization of systems.” Such developments, he said, “reveal a new face of the ancient temptation to domination and mastery without service.”

To guard against that, Leo called for “robust legal frameworks, independent oversight, informed users and a political system that does not abdicate its responsibility,” so that “no single ideology or interest dictates the values embedded in artificial intelligence systems.” The pope’s words to call for a responsible political system echoes language he has used before in tension with the Trump administration’s deregulatory approach to AI. President Donald Trump has pushed to loosen federal AI rules and repealed the Biden administration’s AI executive order in January 2025. When Leo released “Magnifica Humanitas” in May, dubbed the pope’s “AI encyclical,” the Trump administration was split in response when Vice President JD Vance praised it and others dismissed the warning.

Since that time, the pope said AI and technology at large decreases the interactions and relationships people have with one another. This in turn is causing marriage and birth rates to decrease as the ages people reach these milestones increase, if at all. AI, the pope warned, “must never be allowed to erode” the family, and it does so by “reducing persons and relationships to data and simulations,” by “flooding young minds with content that distorts desire,” and through “economic models that make family life economically precarious.”

A redelivery of the church’s stance

The address built on his first encyclical, which made AI’s effect on human dignity, labor and family life the centerpiece of his papacy’s early teaching. Christopher Hale, a political consultant and founder of the newsletter Letters from Leo, said Friday’s remarks were less a new position than a redelivery of ideas already laid out in the encyclical.

“No one reads a 200-page encyclical,” Hale told Fortune. “Oftentimes what will happen is over weeks and months the pope will reveal different parts of that encyclical.” The pope’s Friday’s remarks were in gist a reiteration of his first encyclical—but the point is who it is redelivering the point.

Hale said Leo’s religious authority gives him standing that other AI critics lack in confronting the technology industry.

“Silicon Valley has an opponent that operates on a terrain that they’re not used to,” Hale said. “They’re used to dealing in transactional relationships, but Leo XIV represents something of a quagmire for them because he cannot be bought off, he cannot be terrorized, he can’t be indicted, he can’t be deported.”

Hale added that pairing AI criticism with religious language broadens its reach beyond activists already skeptical of the technology.

“When this language is combined with moral language, with religious language, what it does is it takes a leftist critique that might have marginal support in the United States and makes it mainstream,” he said.

Hale pointed to the backlash against AI data centers as evidence that opposition to the industry already cuts across party lines, even without a shared political language to unite it.

“If you look at the criticism of AI data centers, particularly over the summer, they’re really coming from all factions, from the left and the right,” Hale said. “What’s been hard about it, though, is that there has yet to be a language that can combine the two.”

“He strangely represents the fusion of the populist left and the populist right,” Hale said. “That’s what makes him so powerful.”

This story was originally featured on Fortune.com