Fortune (Page 4)

On Wednesday, Uber CEO Dara Khosrowshahi announced the company will be cutting 3,300 jobs, a tenth of its workforce worldwide.  It’s the biggest cuts since Covid, when the ride-share app cut 14% of its workforce, amounting to 3,700 jobs. 

“This wasn’t a decision we made lightly, because it will have a real impact on our teammates and friends who have worked hard for Uber,” Khosrowshahi wrote in a letter to employees. “It’s important to say that these changes are about how we’re organized and what we’re prioritizing, not about anyone’s contributions to Uber, which we will always value.”

Khosrowshahi said that the cuts were not because of a downturn in the economy—in fact he acknowledged Uber “has grown by orders of magnitude” in the past five years. Instead, it was simply because Uber had gotten too big to manage, and the cuts would help the company run it “faster and smarter.”

Uber made a 20% reduction in the number of employees who were seven layers or more removed from the CEO. The company also cut the number of “micro-teams” (those with only one or two direct reports) by half, and combined each delivery operations team in the restaurant, retail and direct sectors into one big team. 

“The outcome is a simpler org chart geared toward building versus managing,” he wrote. “Running these three businesses separately made sense in their early days, but that structure is no longer serving us at scale.”

Khosrowshahi said the decision was informed by conversations with employees who indicated that the company spent too much time coordinating instead of “building, shipping or serving customers.” 

Uber shares were up 1.61% at $76.45 on closing time. 

Khosrowshahi said the expected savings from making Uber leaner will be reinvested “in growth, innovation, and the capabilities that will matter most over the coming years,” one of which is putting more self-driving cars on the road in Uber’s network. 

“Our opportunity from here is enormous: we have the chance to bring Uber to hundreds of millions more people; to invest even more in drivers, couriers and merchants; and to innovate across our core businesses and build the autonomous future,” Khosrowshahi wrote. 

Uber’s president and chief operations officer Andrew MacDonald  said that nobody will own a car in the next 15 to 20 years because it’s a depreciating asset that sits idle most of the time, and that other forms of transport like self-driving cars and bikes will fill the void. 

“I think autonomous vehicles will be part of that,” MacDonald said last month. 

Uber’s ongoing $10 billion bet on self-driving cars 

Uber originally launched its autonomous vehicle unit Advanced Technologies Group in 2020—an over $7 billion valuation that accounted for more than 10% of its $61 billion market cap at the time. But Uber ended up selling the research unit of it to a startup by the end of the year, recognizing that it didn’t have the money to develop its own robotaxis. 

But the company didn’t give up on self-driving cars, pivoting to deploying them with partners like Rivian, Baidu, and Pony.ai in the Uber network. The ride-share app now expects to commit $10 billion to help bring them to market at scale, with 120,000 cars committed by partners. 

Khosrowshahi described the goal on the Q2 2026 earnings call as making Uber the “world’s leading commercialization platform for autonomous vehicles.”  

“Those results give us the ability to continue investing from a position of strength,” Khosrowshahi said. “We also continue to invest behind one of the largest opportunities in Uber’s history: autonomous vehicles.”

This story was originally featured on Fortune.com

When a company that builds fire trucks, rail-maintenance machinery, or electric vehicle chargers is invited to bid on a big contract, the request that lands in its inbox can run to thousands of pages of technical specifications. Developing the bid is traditionally a slow and stubbornly manual process: sales engineers, technical specialists, lawyers, and commercial staff pass documents back and forth for weeks, sometimes months, before a price can be quoted.

Atira, a Munich-based startup, wants to use AI to automate that work—and it has just raised $17.5 million in venture capital funding to make good on that vision.

The company said it has closed a $15 million seed round led by Accel, along with a previously undisclosed $2.5 million pre-seed. UVC Partners, Fortino, and BOOOM also invested, alongside individual investors that included, among others, Whirlpool chairman and CEO Marc Bitzer and Bastian Nominacher, co-founder and co-CEO of process-mining company Celonis. Atira declined to disclose its valuation following the funding.

Atira estimates that industrial “sales engineering”—the work of turning a customer’s request into a technically viable, priced bid—accounts for more than $128 billion in annual labor spending worldwide. “Sales engineering is one of the largest workflows in manufacturing that software has never truly automated,” Atira cofounder and CEO Florian Diegruber said.

Diegruber previously worked as a commercial lead at Palantir, where he spent much of his time serving one of Germany’s large automotive manufacturers. He and August DuMont Schütte, a former machine learning software engineer at Google, began working together in mid-2024 on a broad thesis about reindustrializing Europe and the U.S. Half a year later they settled on quoting software, a category Diegruber notes “sounded very unsexy back then, which is always a good place to start a venture.” They incorporated Atira in Munich in November 2024.

His argument is that industry has focused on automating factory processes and ignored the “front of house” work needed to create industrial orders in the first place. “When you need six months for your sales cycle to go from request received to final proposal out, it’s great that you’ve automated a week or a month of your factory floor,” he told Fortune. “But the big chunk is on the front door.”

Atira’s platform plugs into a customer’s existing customer relationship management (CRM). enterprise resource planning (ERP), and configure-price-quote systems and deploys AI agents that read incoming requests, highlight key requirements, flag specifications the company cannot meet, then generates the technical documentation, configuration proposals, and pricing options that make up a bid.

Since launching commercially in November 2025, the startup has signed about 15 customers, Diegruber told Fortune. He said that all these customers are now using the system in full production, rather than simply experimenting with pilots, and that five of them have over 100 users each on Atira’s platform.

Among Atira’s customers are ABB E-mobility, the EV charging business that is majority-owned by European engineering giant ABB. Another customer is Chiron Group, the German industrial parts manufacturing firm, has more than 70 people using Atira’s software and reports that it’s able to process inbound quote requests 80% faster than before. (Markus Flik, a former Chiron Group CEO, is among Atira’s investors.) Meanwhile, railway equipment maker Robel says that Atira’s software is saving 95 hours of sales and engineering time per quotation request.

Several other companies are working on similar AI systems that speed up complex quotation processes, including Salesforce and ServiceNow. The major consulting firms are also helping their clients implement similar AI tools and the enterprise sales teams of AI companies, including OpenAI, Anthropic, Google DeepMind, and Mistral are also looking to snag industrial customers. There are other startups in the hunt too, including Roadrunner, which recently raised $27 million from Silicon Valley venture firms Kleiner Perkins and Founders Fund.

But Diegruber said that he is not afraid of this competition. For one thing, he said, the business model of the consulting firms means that they often offer expensive software solutions that require a lot of time to implement. Other software, he said, can’t capture the tacit knowledge that exists within most industrial firms that is essential to getting quotes right. Atira’s AI agents ping human experts over Microsoft Teams to resolve questions no document can answer.

“I always like to joke, if you give Claude to 20 sellers, you still get 20 different answers,” he said, referring to Anthropic’s chatbot. “So it’s really that aspect of scalability across the org versus just solving it for a single person.”

On accuracy—a persistent worry for enterprises deploying large language models—Diegruber said Atira builds dedicated front ends for different types of agents, and that when users review the agents’ answers, the software shows its sources and requires a human to confirm each one. “You can’t just deploy an agent and say the agent now does the work that you’ve previously done manually,” he said.

Harry Nelis, a partner at VC firm Accel, said in a statement that developing quotes for complex manufacturing “involves huge amounts of unstructured information, technical judgement and knowledge spread across different people and systems,” which makes it hard to automate with conventional software. “AI changes that equation because it can understand that context and orchestrate work across the entire process,” Nelis said.

Atira’s traction is notable given the received wisdom that European industrial giants are slower than American ones to buy from startups. Diegruber thinks attitudes among European industrial executives are shifting and that they are increasingly willing to take a risk on a startup’s software. He also said that it helps that Atira uses a deployment model that gets customers up and running within a day or two and lets them cancel after a month. “Luckily, so far this has never happened,” he said, referring to the later scenario.

The money Atira has raised in its latest venture capital round will go mostly into engineering and the company’s first non-founder commercial hires, Diegruber said. Atira had just four employees at the start of the year and has about 15 now, with nine more due to start soon. The company is also eyeing the U.S., where it has a first customer in a pilot, he said. 

This story was originally featured on Fortune.com

Good morning. After nearly a decade helping steer Charter Communications’ cable and broadband business, CFO Jessica Fischer is leaving for a Blackstone- and Google-backed venture investing billions to meet soaring demand for AI computing capacity.

The move puts an experienced Fortune 500 finance chief at the center of one of corporate America’s biggest investment booms. It also reflects how the AI infrastructure buildout is pulling capital and senior talent from established industries as companies race to add computing capacity.

“I’m embarking on an exciting new opportunity in the AI infrastructure space, joining the newly formed joint venture between Blackstone and Google in October,” Fischer wrote on LinkedIn Monday. A Blackstone spokesperson confirmed to CFO Daily that Fischer will become CFO of the venture.

Blackstone and Google announced the venture in May, with plans to build a U.S.-based company providing computing infrastructure for AI models and applications. Blackstone is committing an initial $5 billion in equity. The company is targeting 500 megawatts of capacity by 2027 and plans to expand substantially beyond that. Google will provide its AI chips and related technology, while longtime Google infrastructure executive Benjamin Treynor Sloss will serve as CEO.

Fischer will step down from Charter on Oct. 15. Kevin Howard, a Charter veteran and its EVP, chief accounting officer, and controller, was appointed interim CFO on Monday. Charter will search for a permanent finance chief.

Fischer, age 41, joined Charter as corporate treasurer in 2017, later becoming EVP of finance and CFO in 2021. Before Charter, she was a partner in EY’s national tax department. Charter CEO Chris Winfrey called Fischer a key member of the executive team and credited her with helping build Spectrum into the country’s largest broadband and video provider.

Her departure comes as Charter, No. 85 on the Fortune 500, grapples with steeper-than-expected broadband subscriber losses amid intensifying competition from fixed-wireless providers. Charter said in an SEC filing that Fischer’s resignation did not stem from a disagreement with the company and was unrelated to its operations or financial reporting.

Fischer is moving between two capital-intensive industries at very different points in their growth. Charter is competing for subscribers in a mature connectivity market. Her new employer is preparing to spend billions building computing capacity as demand for AI infrastructure accelerates.

That environment also raises the stakes for the CFO. Building AI infrastructure requires huge upfront investments and decisions about how quickly to add capacity when future demand and returns remain uncertain. Fischer brings experience overseeing finance at a large public company accustomed to making long-term infrastructure investments.

Her move reflects a broader shift in where corporate capital and executive talent are flowing. As billions pour into the infrastructure behind AI, finance chiefs with experience managing large investment programs are becoming central to how quickly the industry can grow and whether those investments ultimately pay off.

Sheryl Estrada
Sheryl.Estrada@fortune.com

This story was originally featured on Fortune.com

“Europe has been worrying about slowing growth since the start of this century,” Mario Draghi said in 2024. “Various strategies to raise rates have come and gone, but the trend has remained unchanged.” 

There is a tendency for gloom to descend when thinking about Europe’s economic and business prospects. In comparison with America, the Gulf and Asia, the mature markets of the EU and the rest of the continent have languished. Since the financial crisis, GDP growth in the euro-area has averaged 0.9% a year. In the U.S., it is above 2%. 

Being European, overdoing the downsides comes naturally. We are a broadly skeptical and conservative bunch, not overly impressed by flamboyant displays of confidence. 

Admittedly the continent has labored as the AI hyper-scalers of America and China have produced products (and valuations) that make the eyes pop. Progress towards a European capital markets union is lumpy. The effects of the U.K. leaving the EU are still being felt. The EU’s Digital Markets Act has been criticized for being both anti-consumer and anti-growth. 

There are, though, plenty of bright spots. On September 16, we will reveal our annual Fortune 500 Europe, the list of the 500 largest companies across the continent by revenue (here’s a link to last year’s list). These are the powerhouses of the European economy, led last year by Volkswagen, Shell and Glencore. The 2026 index will be a treasure trove of statistics on profits, revenues and growth—with many lessons from the successes of those listed. 

On the same day, C-suite leaders from across Europe and the Middle East will be gathering in London for Fortune CEO Forum to talk about growth and share successes and best practices. Leaders from Anthropic and OpenAI will be in the room with the CEOs of Ferrari and Volvo Cars U.K. The U.K. chairman of energy giant EDF will sit alongside board members from NatWest and the in-country CEO of Société Générale. Defense sector policymakers will discuss infrastructure investment with the likes of Honeywell and Tech Mahindra. Entrepreneurs from banking, AI delivery and telecoms will talk about future opportunities. From Microsoft to Shell, C-suite executives representing nearly $2trn of wealth will be in the room. 

Alongside the data from the Fortune 500 index, there are other reasons for optimism. Europe’s Innovation Scorecard, a test of research and investment trends compiled by the European Commission, revealed that innovation performance has increased by 11.6 percentage points since 2019. The U.K., Europe’s second-largest economy, sits happily above the EU average by more than 30 percentage points. “Europe continues to perform well,” the most recent scorecard said. 

The continent boasts some of the greatest universities in the world, is an AI-intellectual powerhouse, has booming financial centers of which many are rightly envious, best-in-class manufacturing from cars to windmills and leads the way on energy sustainability research and non-fossil fuel production. Global leaders flock to Europe for its unique position, geographically and politically, between China, the rest of Asia, the Gulf, and America. Education and healthcare systems are in the top tier. The U.K. wants to see closer co-operation with the rest of the EU. 

In a research note at the end of July, Goldman Sachs said that Europe’s economic growth had been “more resilient than expected’ given the energy price shock which followed the U.S. and Israeli attacks on Iran and the closure of the Strait of Hormuz.  

“We see several reasons for this resilience,” the note said, “The economy’s energy dependence has declined. Fiscal policy supports growth [with] rising defense spending across Europe. Real household income growth remains robust, and labor markets have remained resilient despite sub-potential growth, with the unemployment rate at an all-time low.” 

As a continent keen on saving, consumer confidence remains positive despite stubbornly high inflation. Most families are comfortably liquid and spending is continuing to rise. Incomes are up without the deleterious effects on wealth equality seen in the U.S. 

“We estimate that broad financial conditions—including bank lending conditions and the European Central Bank’s policy stance—point to a positive impulse to growth,” the bank said. 

Business leaders want to turn that impulse into a trend and know that collaboration across the continent and globally is key. Policy makers will also need to play their part. 

“We must take a new stance towards cooperation,” Draghi said. “In removing obstacles, harmonizing rules and coordinating policies, our confidence that we will succeed in moving forward should be strong.” The plan is clear. Now it is time for the execution phase. 

For the latest coverage and updates from Fortune CEO Forum, as well as insights into the companies on our list, visit this page.

This story was originally featured on Fortune.com

In a world of infinite information, trust is the scarcest resource.

In 1999, I co-founded Epinions, one of the first user-generated review sites, and helped pioneer the reviews economy. The premise was that anyone could review anything, and the collective judgment of the masses would sum to something useful.

Epinions succeeded. We sold the company to eBay. But it, along with the review sites that followed, set an industry in motion that would eventually be gamed. Today’s online reviews are a mix of the genuine and the fabricated: real experiences sitting next to anonymous ratings that can be manufactured or bought, often indistinguishable from the real thing. It got bad enough that in 2024, the Federal Trade Commission finalized a landmark rule making deceptive review and testimonial practices illegal.

I had a stake in that fight in 1999, and I find myself with one again now. I co-founded Nextdoor, a neighborhood network built on real identity and local trust. A large part of how neighbors use Nextdoor is to gather local advice and recommendations — reviews, in effect — on local businesses and service providers. My co-founders and I didn’t set out to get back into the reviews business. But here we are, and the timing works in our favor: in today’s world, who’s behind the review matters more than ever.

The platforms at the center of this problem aren’t oblivious to it. They’ve built out enforcement teams, sued fake-review brokers, and removed hundreds of millions of fraudulent reviews over the past year alone. The fake and manipulated share of reviews kept growing anyway. That’s not evidence the effort isn’t real. It’s evidence that policing anonymous content after the fact can’t keep pace with how cheap it’s gotten to fake it in the first place.

Eighty-five percent of consumers now suspect online reviews are fake, and eighty-eight percent say they can’t tell which ones are real. None of that has killed the habit. People still want reviews and still act on them. What’s eroding is their ability to tell which ones deserve their trust. What they actually want to know is who’s telling them this, and why they should believe them.

What makes a recommendation trustworthy?

What the latest issues with the review industry have shown us is that any trust signal is only as durable as what it costs to fake and what you lose when you’re caught. Star ratings fail both tests. There’s no way to know who’s behind the rating, or what their bar for five stars even is. An anonymous rating costs nothing to manufacture, and no one pays a price when it’s exposed.

An accountable written recommendation passes both tests. When a neighbor recommends a contractor under their real name, you know exactly who’s vouching for them, and what that neighbor stands to lose, in reputation and relationship, if the contractor wrecks your kitchen. The reviewer has skin in the game. The star never did.

Early research fielded this summer by the consumer-insights platform Echo points in the same direction: nearly half of respondents said they’d suspected a fake or manipulated review in the past year, and 81% said they’d trust a recommendation more if the person making it were held socially accountable.

When a machine answers instead of a person

Ask a model today to name the best pizza place in town and you’ll get a confident, instant answer for free. But abundance always changes what’s scarce. Information used to be scarce. Now it’s free and instant. Provenance isn’t.

The machines giving those answers are trained on the same reviews we already can’t trust. It turns out they’re no better than we are at telling which ones are real. A 2025 study out of Nottingham University Business School found that AI-generated fake reviews are now indistinguishable from real ones, to human readers and to other AI models alike. This reality just creates more unreliable confusion for the end consumer.  

And AI presents problems for the businesses being reviewed. They have no way to influence the answers the LLMs produce, and even more problematic, many of them have been targets of “AI bombing”, the practice of using generative AI to rapidly create and post large volumes of fake, misleading, or low-quality reviews about a business, artificially manipulating its online reputation.

A competitor, a bad actor, anyone with a prompt, can now bury a business under thousands of reviews it never earned, and no model, human or otherwise, can reliably tell the fake from the real. When reputation can be manufactured at scale, provenance stops being a nice-to-have. It becomes the whole business.”

A simple test you can use

What I started with Epinions became an industry all its own, one that’s still serving, and potentially costing, the consumers who rely on it today. Before you act on a recommendation you find online, ask who’s telling you this and what they have to gain or lose. That question hasn’t changed since 1999. What’s changed is how much the industry profits when you don’t ask it.

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

• In today’s CEO Daily: How leaders are contending with AI slop.
• The big leadership story: Trump’s salary freeze prompts worries about worker retention.
• The markets: Mixed globally amid more fighting between the U.S. and Iran.
• Plus: All the news and watercooler chat from Fortune.

Good morning. What’s your take on AI slop? I now get multiple AI-generated PR pitches every day. The format is comically uniform, from the greeting to the bullet-point topics, and I think it’s a disservice to the person being pitched. We all crave that human touch. More than 1 million people have clicked LinkedIn’s “seems like AI slop” button, and Graphite reports that AI now generates more online articles than humans do.

The explosion of AI-written content is sparking debates over how to handle op-eds like the one that investor Stan Druckenmiller wrote for the Wall Street Journal using AI, which sparked a response from his editor, and much debate about what’s appropriate. (My take, with strong caveats, is that editors are to blame if a piece feels manufactured, predictable, and devoid of personality. I also debated the topic yesterday on BBC’s The Media Show.)

I’ve been asking people how AI is shaping what they consume and create. At a party for David Booth’s terrific book Stay Calm earlier this week, a prominent academic told me he’s now using AI up to 10 hours every day, admitting he’s both obsessed and worried about what it can do. 

But Booth, a pioneer of index investing and founder of Dimensional Fund Advisors, is a fine example of what doesn’t change. His book tells a story that only he can tell—how he went from a shoe salesman in Kansas to a $1 trillion asset manager—and it synthesizes a philosophy of investing that he’s discussed and shared for decades. 

Did he use a ghostwriter or AI? If he did, you wouldn’t know it.

What’s emerging from my conversations:

AI leaves leaders craving a good read. CEOs tell me they’re reading fewer books and reports because of AI, and that worries them. They’re not alone. About 40% of Americans didn’t read a book last year. One CEO in financial services told me that they’re now trying to read more the old-fashioned way. “I’m paid to think,” they told me, “and I realized I’m losing the nuance and originality of an argument when I read these summaries. A slower read, a deeper read, gives you more time to digest and come to your own conclusions.” 

We value human creativity. I spoke yesterday with Andy Hunter, CEO of Bookshop.org, which acts as an e-commerce platform for 90% of the country’s 3,400+ independent booksellers—a category that’s booming, by the way. He uses AI for code review but not customer recommendations. He worries less about slop than fraud, as he’s seeing a lot of AI-generated books created on the spot to capitalize on search trends. “There’s a lot of garbage content being created to trick people into buying it; I don’t think Amazon is doing a good enough job of protecting their customers from that,” he told me. “We want human beings to recommend books to other human beings. That’s what independent bookstores do. The solution to AI slop is human curation, human creation.”

When everyone’s better, it’s hard to be the best. The baseline for content and communication is now around B+. Anyone can tell a coherent story. But having the tools to create a Hollywood movie on an iPhone doesn’t make you a great director any more than the ability to spit out a column makes you a great thought leader. AI expands the mushy middle, where nobody wants to be. As David Meadvin, CEO of One Strategy Group, puts it: “I think the right question to ask here is whether AI is making you a little bit more ordinary.”

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

This story was originally featured on Fortune.com

Good morning. On Fortune’s radar today:

• The White House’s misleading stats on oil shipments from the Gulf.
• There was ‘near-zero employment growth in August,’ Vanguard says.
• Markets: We’re having an up day.
• Chart: How Elon Musk’s Starlink dominates space.
• Foldable iPhone Ultra will be Apple’s most expensive handset ever.
• Women and Millennials want their dates to earn way, way more than average.

This story was originally featured on Fortune.com

Pope Francis spent much of his papacy pushing the Catholic Church to treat climate change as a moral issue in addition to a scientific. Now, the Holy See’s plan is putting money behind that morality and looking to make the Vatican the world’s first carbon-neutral state.

Vatican City, which covers about 44 hectares, or 109 acres, will put roughly 100 million euros ($117 million) toward an agrivoltaic plant expected to power the microstate within two years, sources told Reuters.

The plan builds on what Pope Francis first ordered in 2024. Pope Leo has spent the past year moving it through the bureaucratic steps needed to build it by creating a foundation, a memorandum with an Italian utility, and a joint commission. What’s new is that this is the first time a dollar figure has been attached to the plan. And it also puts into context what we’re doing stateside.

The Vatican is roughly an eighth the size of Central Park. It’s smaller than a single data center campus like Meta’s Hyperion data center project in Louisiana, which is on track to draw as much as 5 gigawatts of power by 2030. Home to roughly 800 residents and nearly 3,000 commuting lay workers, it’s the smallest country on Earth by a wide margin.

It’s so small that the plant won’t fit within the country’s 109 acres: It’s being built on the roughly 1,050-to-1,100-acre Santa Maria di Galeria estate the Vatican owns outside Rome, land close in size to the campus for OpenAI’s Stargate data center in Abilene, Texas. But where Stargate’s footprint is built to support 1.2 gigawatts of AI computing, the Vatican’s similarly sized site is expected to generate just 80 to 90 megawatts of capacity, per Reuters, or roughly a dozen times less power on similar acreage. But that’s because it’s built to run a few hundred residents’ worth of lights and a radio transmitter, not racks of GPUs.

At risk of power shortages

A single large AI data center now consumes as much electricity as a city of 80,000 people, Khosla Ventures founder Vinod Khosla wrote in Fortune this year, and companies are building hundreds of them at once. Data centers accounted for roughly half of all new U.S. electricity demand in 2025, and globally, data center electricity use is projected to more than double by 2030 to around 945 terawatt-hours, equivalent to Japan’s entire current electricity consumption.

The buildout is reshaping the U.S. utility business, too. NextEra Energy’s $66.8 billion deal to acquire Dominion Energy in May was driven largely by AI data center demand clustered in Virginia, rather than residential growth. American Electric Power plans to spend $78 billion through 2030 to keep pace. Goldman Sachs projected the AI buildout will push US electricity costs up 6% between 2026 and 2027, with a further 3% increase by 2028, increasing neighboring homeowners’ utility bills. By contrast, the Vatican’s entire planned plant would generate less power than a single mid-sized AI campus, let alone the industry racing to build hundreds more.

The new plant would also dwarf what the Vatican currently produces. Its only existing solar installation—2,394 panels on the roof of the Paul VI Audience Hall, installed in 2008—generates about 300 megawatt-hours a year. The new plant’s 80- to 90-megawatt capacity alone is hundreds of times larger than that. The Vatican’s 2023 climate pledge to the UN sets a 20% emissions cut from 2011 levels by 2030 and a “zero emissions” goal by 2050.

Trump at odds with the Vatican

The Church’s position on climate traces to Francis’s 2015 encyclical “Laudato Si’,” which argued that wealthy nations bear disproportionate responsibility for emissions and called for a shift away from fossil fuels. Pope Leo XIV has continued holding that stance since Francis’s death, calling for “urgent” climate action at a UN conference this year and saying that God’s creation is “crying out.” He’s also labeled new technology, including AI, a moral concern: Leo has said he chose his papal name partly in reference to Pope Leo XIII, who wrote about protecting workers during the first Industrial Revolution, and is applying that same lens to AI’s effect on labor today.

That stance puts the Vatican at odds with the current U.S. administration: President Donald Trump has called climate change “the greatest con job ever perpetrated on the world,” and the U.S. skipped this year’s UN climate conference entirely. Trump and Leo have also clashed more publicly over Trump’s military actions in Venezuela and Iran and his administration’s immigration policies, with Trump at one point suggesting Leo’s election as pope was illegitimate. Fiscally, the U.S. ended a 2025 federal credit for residential solar as the EPA canceled a $7 billion low-income solar program in August. A BlueGreen Alliance tally found 223 clean-energy projects, worth $82.8 billion, canceled or stalled since the policy shift.

After Francis formalized the Vatican’s plan in a June 2024 letter called “Fratello Sole,” Leo kept it moving, visiting the site in June 2025 and calling it “a wonderful opportunity.” Italy agreed to the plan in July 2025, and the Vatican formalized ties with the Italian utility ACEA in June 2026 to build it jointly. Sources told Reuters the full build is expected to take 18 to 24 months.

This story was originally featured on Fortune.com

Tim Cook may be giving up Apple’s top job, but the company isn’t letting its longtime CEO fall far from the tree just yet.

Cook, who handed the CEO title to longtime hardware chief John Ternus on Tuesday, will stay on as executive chair for the tech company, earning an annual salary of $2 million effective Sept. 26. As chair, Cook will continue assisting with certain aspects of Apple, including “engaging with policymakers around the world,” the company said in April when it announced his new role.

Part of the reason Cook may stay on at Apple is that during his 15 year tenure, he became a sort of “Trump whisperer” in navigating the company’s relationship with the president. Cook earned the nickname “Trump whisperer” for his ability to strike deals and keep things smooth with the president, as seen through both his tariff negotiations and this year’s CEO transition.

Cook even said he’d continued to manage the company’s relationship with not only Trump but also the Chinese government, according to Bloomberg. Those relationships have become increasingly important to Apple as its sprawling Asian supply chain puts the company in the middle of Trump’s push to use tariffs to bring more manufacturing to the U.S.—and leaves it navigating tensions between Washington and Beijing.

His package won’t be limited to the $2 million annual comp either—he was approved for an equity award with a target value of $45 million, according to a regulatory filing.

Cook leaves the CEO job after 15 years at Apple’s helm, having taken over shortly before cofounder Steve Jobs’ death in 2011. During his tenure, Apple’s market capitalization grew from roughly $347 billion to $4.6 trillion, all while increasingly navigating government alongside products and operations.

How Tim Cook became the ‘Trump whisperer’

Cook figured out relatively early how to get Trump’s ear.

“He’s the one that calls me,” Trump told reporters in 2019 when he was asked why he had a good relationship with Cook compared with other tech executives. At the time, tariffs were a problem. Trump said Cook had argued that levies on Apple’s China-made products could put the company at a disadvantage against South Korean rival Samsung.

That relationship would be tested again after Trump returned to the White House. In February 2025, Apple announced it would spend $500 billion in the U.S. over four years, including plans to hire 20,000 workers. 

But Cook’s access to Trump did not mean Apple always got what it wanted. By May 2025, Trump was publicly criticizing Apple as the company expanded iPhone production in India, as part of its efforts to reduce its reliance on China.

“I had a little problem with Tim Cook yesterday,” Trump said at the time, recounting a conversation with the Apple CEO.

Trump said he told Cook that he didn’t want Apple expanding production in India and wanted the company to manufacture more in the U.S. Days later, the president threatened Apple with a tariff of at least 25% on iPhones sold in the U.S. but manufactured elsewhere.

Three months later, Cook was back in the Oval Office standing beside Trump, and he arrived with a customized “Made in USA” glass plaque mounted on a 24-karat gold stand for the president. 

Last August, Apple announced another $100 billion commitment to U.S. manufacturing, bringing its planned domestic spending to $600 billion over four years. 

The timing mattered. Trump was preparing to impose a roughly 100% tariff on imported semiconductor chips, but said during the Oval Office event that companies building in the U.S. would not have to pay the levy. 

That work extends well beyond the White House. Apple remains heavily dependent on Asia for manufacturing even as it has shifted some production away from China, leaving the company exposed to the relationship between the world’s two largest economies. Cook has spent years navigating Apple’s relationship with Beijing, and earlier this year, he was among a group of U.S. executives expected to accompany Trump on a trip to Beijing as the president prepared to meet Chinese President Xi Jinping.

Trump has offered his own explanation for why the relationship endured. When Cook announced in April that he would step down as CEO, the president recalled receiving his first call from the Apple executive during his first administration.

“When I got the call I said, wow, it’s Tim Apple (Cook!) calling,” Trump wrote on his social media app, Truth Social, a nod to when he years earlier referred to Cook as “Tim Apple.” Trump called it the beginning of “a long and very nice relationship.” He said Cook continued calling him over the years—but “never too much”—and that he would help the Apple CEO where he could.

“What I do is I interact on policy, not politics,” Cook told ABC’s “Good Morning America” earlier this year in March. “I focus on policy.”

This story was originally featured on Fortune.com

It seems like so long ago, but in the not-so-distant past, all of us were roaming the supermarket aisles for the last roll of TP, cursing at those who began stockpiling the rolls, and blaming the relative who brought a bidet back from Japan for not bringing one back for you as well. While the days of reaching for the last TP under a “limit one roll per customer” sign shoddily taped to the rack may be over, the thought of a shortage possibly returning to haunt the same president six years later isn’t far-fetched.

After trade talks between the U.S. and Canada broke down in late August, Washington imposed a 50% tariff on roughly $20 billion of Canadian goods under a section of the Tariff Act of 1930—a Depression-era law that lets the president retaliate against countries seen as discriminating against U.S. commerce. In response, Canadian Prime Minister Mark Carney vowed to match the US “dollar for dollar,” and unveiled retaliatory tariffs on close to 900 American products, set to take effect next week, on Sept. 8. Chief among them? Toilet paper and paper products.

“Canada will match those tariffs dollar for dollar,” Carney said two weeks ago. “Because we were attacked. Like, you’re at war when you get attacked.”

Paper goods are among the hardest hit. Canada’s retaliation puts a 25% duty on U.S. toilet paper and facial tissue, and a 50% duty on paper towels, napkins, and raw wood pulp. Canada is also charging 50% on American dairy products (except cheese, which gets 25%) and 25% on American fish and seafood, including frozen lobster. It’s in response to U.S. tariffs, including a 25% tariff on Canadian cars and auto parts—and Trump is now threatening to double that to 50% if no deal is reached by Jan. 1, 2027.

A Covid-era toilet paper shortage

On March 12, 2020, U.S. toilet paper sales surged 734% compared with the same day the year before, making it the top-selling grocery item that day. By the time the panic subsided, roughly 70% of the world’s grocery stores had run out of toilet paper at some point. Shelves that normally held Charmin and Cottonelle sat bare for weeks as retailers imposed purchase limits, and bamboo-paper startups saw Amazon sales jump more than 5,000% month over month as shoppers hunted for anything they could find. Researchers who studied the episode afterward found no actual break in production. Mills kept running. Hoarding, not a supply failure, created nearly all of it.

Eight months later, due to a confluence of the COVID-19 pandemic and his administration’s handling of the virus, President Donald Trump lost the presidency to Joe Biden. Toilet paper didn’t decide that election, but it became a small symbol of an economy that felt like it was slipping out of a president’s control, just as voters decided whether to give him another term.

Americans make up just 4% of the world’s population but use more than 20% of the world’s tissue supply, burning through an average of 141 rolls a year. Six years after the pandemic, Trump is again facing affordability and kitchen table prices, just as the midterms campaign season kicks into high gear.

The toilet paper supplier and tariffs

Trade talks between the US and Canada broke down in late August 2026. In response, Washington imposed a 50% tariff on roughly $20 billion (C$27.6 billion) of Canadian goods under Section 338 of the Tariff Act of 1930—a Depression-era law that lets the president retaliate against countries seen as discriminating against US commerce. Canadian Prime Minister Mark Carney vowed to match the US “dollar for dollar,” and Ottawa unveiled retaliatory tariffs on close to 900 American products, set to take effect September 8.

Paper goods are among the hardest hit. Canada’s retaliation puts a 25% duty on US toilet paper and facial tissue, and a 50% duty on paper towels, napkins, and raw wood pulp. The dispute doesn’t stop at paper: Canada is also charging 50% on American dairy products (except cheese, which gets 25%) and 25% on American fish and seafood, including frozen lobster, a move that drew criticism from Maine Senator Susan Collins. The US, for its part, has also placed a 25% tariff on Canadian cars and auto parts, with Trump threatening to double that to 50% if no deal is reached by January 1, 2027. To soften the blow at home, Ottawa announced a C$7.5 billion relief package for small and medium Canadian businesses hurt by the standoff.

Americans make up just 4% of the world’s population but use more than 20% of the world’s tissue supply, burning through an average of 141 rolls a year—the highest rate globally, just ahead of Germany’s 134.

In 2024, Canada supplied $328 million worth of toilet paper to the country, more than any other country, just as retailers like Costco source much of their private-label paper stock from Canadian mills. The supply chain complicates this as American factories rely on northern bleached softwood kraft pulp, or NBSK, harvested from Canadian forests, which makes up roughly 30% of a standard American toilet paper roll and close to half of U.S. paper towel production. U.S. mills buy about 2 million tons of Canadian NBSK a year.

Procter & Gamble, which makes Charmin, said tariffs are creating a roughly $0.25-per-share earnings headwind and is raising prices on a chunk of its lineup. Americans are paying 96% of the cost of Trump’s tariffs, and Federal Reserve research showed tariffs are now seeing “full pass-through” to consumer prices.

The American Forest & Paper Association, which did not respond to Fortune’s requests for comment, argues that any shortages would be localized rather than become national. It’s still urging consumers not to stock up ahead of the September 8 deadline.

All of this is posing a problem for Trump: Ahead of the midterms, his economic standing has slipped further than the White House would like. Just 39% of Americans approve of his job performance, and only 30% approve of his handling of inflation, with a 31% plurality naming inflation and prices their top concern — 17 points ahead of the next issue, jobs and the economy. Even Marc Short, who served in Trump’s first administration, said it’s hard to argue his tariffs aren’t contributing to the affordability problem. Goldman Sachs research found that the cost of living is voters’ top issue at a higher share than heading into the 2024 election.

This story was originally featured on Fortune.com

Anthropic has become the second leading AI lab to reveal it temporarily paused some advanced AI training amid concerns over rogue agent attacks.

The company said this week it paused training of unreleased models for several weeks following two incidents reported in late July, including one in which Claude Mythos 5 took unauthorized actions during a U.K. AI Security Institute cybersecurity test. OpenAI, the company’s bitter rival in the AI race, took a similar step last month when it paused some AI training for two weeks after several of its models breached AI company Hugging Face’s infrastructure during an internal test.

The training pauses, which come as both companies reportedly prepare for trillion-dollar initial public offerings, demonstrate how much the industry has been disturbed by the recent rogue AI agent hacks. It marks a shift for an industry that for the last few years has been locked in a fast-paced race, with rival labs competing to bring ever more capable models to market as fast as possible. Now, two of the leading companies appear to be competing on which can show it is the most attuned to AI safety concerns—while also not slowing model development so much that it risks customers defecting to a competitor’s more capable offering.

Notably, the wave of rogue AI incidents prompted an open letter called “Pacing the Frontier,” in which more than 1,100 employees across OpenAI, Anthropic, Google DeepMind and Meta asked the U.S. government to help build a governance mechanism that could slow frontier AI development if needed. Signatories included Anthropic chief executive Dario Amodei and co-founders Jared Kaplan and Jack Clark, alongside OpenAI chief scientist Jakub Pachocki. Both companies endorsed the letter at the corporate level within hours of its publication.

The recent training pauses from Anthropic and OpenAI were seen by some in the industry to be a direct result of the letter.

“Pacing the frontier success story?” Roon, a popular AI commentator widely believed to be a pseudonym for OpenAI researcher Tarun Gogineni, wrote of the announcements on X. “Next time let’s do it proactively before there’s any absurd loss of control events.”

Anthropic, like OpenAI, announced it would be working with independent AI safety evaluation group METR to conduct an outside review of the incidents, saying it wanted to ensure the resulting studies were thorough and promising more detail in the coming weeks.

The two companies’ accounts of what went wrong when their respective agents took real world actions against instructions are also similar. Anthropic said its Mythos 5 model showed a pattern of “motivated reasoning,” maintaining an initial belief that it was operating in a simulated environment even after encountering evidence it was connected to the live internet, alongside a willingness to take harmful actions in pursuit of completing an assigned task.

Redwood Research, one of the outside groups OpenAI brought in after the Hugging Face breach, also described the behavior it observed with OpenAI’s agents as score-seeking misalignment rather than a longer-term scheme by the model.

Both companies have also pointed to their reinforcement learning environments as a source of the problem. Reinforcement learning is a training method where AI models learn by trial-and-error to maximize some reward. One problem with the technique though is that it can lead to “reward hacking,” where AI models learn that they can achieve the reward in a way that those training the model didn’t intend, such as cheating on an assessment or taking harmful actions in pursuit of the reward.

To resolve some of the issues, OpenAI has brought in new monitoring tools designed to alert internal safety and security teams within 30 minutes of detecting concerning model activity, and to trigger an automatic pause if that alert cannot be resolved in that window. 

Anthropic said it has built a similar tool, a system that scans a model’s actions as it works and automatically blocks anything that looks like an attempt to escape or exploit its test environment, ending the task and alerting a human before it can go further. The company also said it moved about 150 product engineers onto security work starting in April, and tightened access to its systems, including cutting off most outbound internet traffic from its computing clusters by default.

While safety experts say the new controls and pauses are a welcome change, some say there’s still more needed.

“The temporary pace changes are a good first step, but there’s still a way to go,” Steven Adler, a former OpenAI employee and co-founder of the non-profit Guidelight AI Standards, told Fortune. “We need predictable, verifiable pacing across the frontier, not just ad-hoc decisions to slow down. And we need companies to use the additional time to implement serious preventative controls, which still seem to be missing.”

Anthropic, at least in the blog post, has indicated that it may be willing to go further in the future to help pace AI development.

“Some of our senior leadership and many of our employees recently signed a letter calling for greater coordination on pacing, and we will say more in the coming weeks about how we intend to contribute to that effort,” Anthropic wrote in the post. “We believe the world would benefit if the industry adopted a lawful, verifiable, effective mechanism for coordinated pacing as soon as possible,” the company wrote.

This story was originally featured on Fortune.com

Iran fired on U.S. allies in the Gulf early Wednesday following a night of American bombardment that Iranian state media says included a deadly strike on a wedding party.

The latest violence came after a monthlong lull in fighting. The six-month war has led to a spike in oil prices, roiled the global economy and posed increasing political problems for U.S. President Donald Trump’s Republican Party ahead of November midterm elections.

Fighting has escalated since the American military hit Iranian rocket launchers on an island in the Strait of Hormuz over the weekend, saying Iran was planning to use them to lay mines. Iran retaliated by firing missiles at American bases in Jordan that were intercepted.

American military resources are stretched and political pressure in Washington is mounting, while Iran’s economy has been battered by sanctions and an American naval blockade. Iran’s currency hit another record low on Wednesday, with traders in Tehran exchanging 2.20 million rials for one U.S. dollar, a 10% hike from last week’s record.

Neither side can continue indefinitely between full-scale war and a truce, said Hamidreza Azizi, senior Iran analyst at the Crisis Group, an international think tank.

“At some point, one of the sides may see a need to escalate in order to break this cycle,” he cautioned. “So that could bring the two sides to a point that they have been trying to avoid until now, which is a full-scale war.”

The struggle is focused on the Strait of Hormuz and oil prices

The latest flare-up in violence came as the standoff seemed to be shifting slightly more in the U.S. favor, with the price of oil hitting around $80 a barrel last month and Iranian leaders talking about growing economic hardships, said Sascha Bruchmann, an analyst with the International Institute for Strategic Studies’ Middle East office in Bahrain.

“The U.S. is trying to reinstall some kind of confidence into the global oil markets … and the Iranians had to respond to that by reinserting insecurity and doubt,” Bruchmann said.

Since the fighting resumed over the weekend, the price of Brent crude oil, the international standard, has climbed to around $95, up more than 30% from the start of the war.

U.S. Secretary of State Marco Rubio said Iran will “continue to feel the squeeze” from the United States until it drops any nuclear weapons ambitions and ends support for armed groups in the Middle East. Iran has always insisted its nuclear program is peaceful.

U.S. allies in the region fend off Iranian attacks

Following the strikes on Jordan, U.S. Gulf allies Kuwait and Bahrain said they had also come under attack, without reporting any casualties.

“You will soon see that Iran’s new strategy on the battlefield, in diplomacy, and in confronting the economic blockade will shatter your foundations,” the recently appointed head of Iran’s powerful Supreme National Security Council, Mohsen Rezaei, said on X.

In Iraq, where Iranian-backed forces are active, Kurdish authorities said they had intercepted 10 drones laden with explosives in the Irbil area. Iranian television reported that American bases in Irbil had been targeted.

Israel’s military also said that Iran-backed Hezbollah fired two explosive drones at its soldiers in southern Lebanon, without causing casualties.

The U.S. Central Command said Tuesday’s strikes focused on Iranian military targets, including air defense sites, radar systems and maritime assets.

Strike on wedding killed 4 and wounded dozens, state media says

One U.S. strike hit a home hosting a wedding in Kuhestak, a small coastal town overlooking the Strait of Hormuz, according to Iranian state media.

Four people were killed — two women and two children, aged 4 and 16 — and at least 68 were wounded, the state news agency IRNA and state TV said Wednesday. Just after the strike, Ahmad Nafisi, the province’s deputy governor, told Iranian state television that five people were killed. There was no explanation for the change.

Several heavy explosions shook Kuhestak, with fireballs that lit up the night sky, according to security camera footage carried by the semiofficial Fars news agency.

Images from Iran’s semiofficial Tasnim news agency showed several hospitalized children being treated for injuries at a hospital in the nearby town of Minab, where a U.S. missile strike on the opening day of the war hit a primary school, killing more than 100 children.

Footage from the Iranian Red Crescent showed rescue workers entering a damaged home in the wake of the strike. It showed a hole in the roof of one house in a neighborhood of densely packed homes.

Expert says fragments point to a U.S. weapon

A weapons expert told The Associated Press that images of fragments of ordnance that Iranian media said was used in the strike on Mallahi’s home, indicated it was an “American air-delivered guided weapon.”

N.R. Jenzen-Jones, director of Armament Research Services, initially identified the weapon as a SLAM-ER cruise missile, an air-launched missile that can be retargeted in flight by a human operator. But after examining further imagery, he said it was more likely to be a JSOW, a type of guided bomb, though he did not rule out a SLAM-ER and said multiple weapons may have hit the site.

U.S. Central Command declined to comment on the analyst’s findings. Spokesperson Capt. Tim Hawkins said earlier that the U.S. military was “aware of the reports, which originated from Iranian state media, and we are looking into them.” He would not say if CENTCOM had launched a formal investigation.

“Make no mistake: These crimes will not go unpunished,” Ebrahim Azizi, chairperson of the Iranian parliament’s National Security and Foreign Policy Committee, said on X.

Separately, Iran has said four members of its paramilitary Revolutionary Guard and 10 members of the Guard’s volunteer Basij forces died in the U.S. attacks. A total of 108 people were injured in the latest round of attacks, state broadcaster the Islamic Republic of Iran Broadcasting quoted the country’s health minister as saying.

The Strait of Hormuz remains largely shut down

During last month’s lull in fighting, the U.S. ramped up economic pressure on Iran as Washington sought to fully open the Strait of Hormuz.

About a fifth of the world’s traded oil was shipped through the key waterway before Israel and the U.S. attacked Iran on Feb. 28 and Iran largely shut it down in response. Despite many pronouncements from Washington that the strait is open, only a handful of ships are making the passage each day and Iran has regularly attacked vessels.

On Wednesday, Saudi shipping company Bahri said two Filipino sailors were killed in an attack late Monday as their vessel transited the strait.

A ceasefire deal reached in June quickly fell apart amid exchanges of strikes that same month. Mediator Pakistan has been working to restart talks.

Late Tuesday, Trump said he would not force Iran to the bargaining table.

“I couldn’t care less if they sign a worthless, to them, agreement. I like our position now much better, with almost total control of the Hormuz Strait, and their economy totally collapsing. They are just playing out the inevitable. When are the Iranian people going to rise up and fight?” Trump wrote on social media.

___

This story has been corrected to show that Iranian state media are reporting that four people were killed in the strike on the wedding, not five. They gave no explanation for the change. It has also been corrected to show that Trevor Ball is an analyst with Armament Research Services, not Bellingcat.

___

Magdy reported from Cairo. Lee Keath and Fatma Khaled in Cairo, Sarah El Deeb in Beirut, Konstantin Toropin in Nuremberg, Germany, Stella Martany in Irbil, Iraq, Christopher Weber in Los Angeles, and Matthew Lee in Washington contributed to this story.

This story was originally featured on Fortune.com

When Simon’s wife of 43 years passed away, it felt a bit like he’d died too.

In his overwhelming grief, Simon sought companionship online, like so many other lonely widows and widowers. Almost immediately he met Emily. But over the next few months, Emily, who Simon realized too late was a fake, stole $800,000 from him.

And that was just the beginning.

Simon was stuck paying back $185,000 he had borrowed, and tens of thousands of dollars in additional taxes on money he had withdrawn and lost. He tried to report the crime to his local police and the FBI, he said, but nothing came of it. Instead, he was contacted by yet another scammer — offering to connect him with the Secret Service to recoup his losses in exchange for even more money.

Like Simon, virtually every American has been targeted by scammers in some way, exclusive new polling data shows. Scams in the U.S. have surged to a record high, causing billions in losses for Americans.

But the U.S. still lags behind some other countries in tackling the problem, holding financial and social media companies responsible and regulating the cryptocurrency scammers rely on, an investigation by The Associated Press and FRONTLINE found. And although both the Trump administration and Congress are pursuing new options, victims still have little recourse and sometimes end up losing even more.

In interviews with almost five dozen victims, AP and FRONTLINE found that many faced ridicule and stigma from friends and family, mounting pressure from banks and lenders, additional taxes, dismissive law enforcement and neglect from a government they believed would protect them.

“What happens after the scam might even be worse than the scam itself,” said Erin West, a former prosecutor and founder of the nonprofit Operation Shamrock that provides support for victims of online scams. “It’s a travesty.”

While victims say they seldom get any answers from authorities, AP/FRONTLINE found these crimes can be traceable. With the help of leaked documents and cryptowallet addresses, AP/FRONTLINE tracked Emily and Simon’s stolen funds to a notorious scam compound halfway around the world in Myanmar with thousands of victims.

Simon was stunned to see Emily’s fake profile and hundreds of pages of conversations he thought were encrypted.

“What a fool I was,” he said.

Today, much of the Myanmar compound where his scam originated lies in rubble, destroyed by local authorities. Yet many of the illicit operations simply moved, new sites opened immediately, and Simon is no closer to getting his money back.

He asked not to be identified by his full name because he is too ashamed to tell most of his family about the crime. Local police and the FBI declined to comment on the case.

“(The police) told me right away that, you know, ‘You have to kiss that money goodbye,’” he said, fighting back tears. “Already I experienced something very bad and now I have to pay for the consequences on top of it, and see my money evaporate all over again. You lose two ways.”

As scams surge, fees and taxes pile on

The size of the global scam industry is staggering.

Americans reported a record $15.9 billion in losses last year to the Federal Trade Commission, a 25% increase from 2024. And such estimates are likely a massive undercount because most scam victims are too embarrassed to report their crimes.

The FTC estimates that real losses in 2024 were close to $200 billion; that’s $550 million every day just in the U.S.

The ballooning costs are fueled by huge advances in artificial intelligence and cryptocurrency. AI allows scammers to work at a previously unimaginable scale and level of sophistication, and cryptocurrency, often used in investment scams, is a form of digital cash that can be hard to trace to its real owners.

No one is immune, the AP/FRONTLINE investigation found. Ninety-eight percent of Americans suspect they have been targeted by scammers, many every day, according to a poll conducted by The Associated Press-NORC Center for Public Affairs Research. Three in 10 said they have personally lost money or information to scams.

AP/FRONTLINE interviewed 58 people in the United States who lost money to cyberscams, aged 32 to 90, across all races and income levels. They said they lost several thousand dollars to $4 million each and included doctors, IT professionals, academics with advanced degrees and people just trying to make ends meet.

All said they felt alone and confused afterward. When they did report the crimes to law enforcement, many felt abandoned by a lack of support. Several victims said they contemplated suicide. Two attempted it.

The only one to get any money back won a rare settlement from her bank, not from her scammer.

Instead, scam victims sometimes owe additional money through bank fees and tax bills. The IRS often demands that retirees, including scam victims, pay taxes on any funds they withdrew from tax-deferred accounts like their retirement savings.

Before 2018, victims of theft or fraud could sometimes deduct losses incurred from their taxable income. But under a provision of the Trump administration’s Tax Cuts and Jobs Act, made permanent in 2025, personal losses from many common scams are not eligible for tax breaks. That means victims can owe taxes after money was stolen from them.

Retired nurse Susan Bivins said she was tricked into draining her retirement accounts and sending more than $200,000 to a scammer pretending to be a federal agent. The FBI and local police did not help, she said.

Then her tax bill arrived. She owed the Internal Revenue Service $80,000.

“I wanted to drive off a cliff,” she said. “I didn’t know how I was going to live.”

Bivins sold her home and moved into a small one-bedroom apartment. She is still paying off her tax bill, selling handmade quilts.

The FBI said in a statement that it takes scams and fraud seriously but cannot comment on individual cases to protect privacy. Local police did not respond to repeated requests for comment.

A law before Congress, the Tax Relief for Fraud Victims Act, would reinstate an allowance for victims to deduct their losses. AARP fraud prevention lead Kathy Stokes urged Congress earlier this year to help scam victims who are “stuck with a bill they don’t deserve and have no way to pay.”

“Essentially they’re being revictimized by the U.S. government,” Stokes said.

Banks also sometimes blame people who are scammed instead of seeing them as crime victims, or even accuse them of being complicit. Victims described having their accounts abruptly frozen or canceled, along with demands for repayment on loans and legal fees.

Debra Fox in Colorado lost $58,000 in a romance scam, she said. In a meeting with her bank branch, she said, a representative told her that she would be held responsible for anything fraudulent linked to her own accounts and forced to pay any legal fees.

She was in shock for 48 hours, terrified that the little she had left would vanish.

“I thought, I have no control over this process,” she said. “The crime was horrific enough… but it’s what happened next that was really unbelievable to me.”

Her local bank branch declined to comment on her case.

Earlier this year, American Bankers Association Chair Kenneth Kelly said banks spend “time, money and significant resources” trying to stop fraud, which usually refers to unauthorized transactions. But under current U.S. law, financial institutions are rarely liable for transactions their customers authorize.

There are rare workarounds. For example, under new legal protections, a theft from a senior citizen in California might fall under the definition of financial elder abuse.

Alice Lin, now 83, was the only victim AP/FRONTLINE interviewed to get some money back, after she sued her bank, citing this law.

Over their lifetimes, Lin and her husband had put aside a nest egg of about $720,000. After he died, she struck up what she thought was an online friendship with a man who’d also lost a spouse.

Her new friend Justin convinced her to take out money from her bank and invest it in cryptocurrency platforms.

At first, her financial statements showed big wins. But when Justin encouraged her to ask her daughter for a loan to invest more, her family realized it was a scam.

Lin sued JPMorgan Chase for failing to protect an elderly citizen and flag the extraordinarily large and unusual transfers. Earlier this year, they settled for an undisclosed amount, according to her lawyers.

JPMorgan Chase said it has multiple layers of scam protection, from bankers who asked Lin about her transactions to written warnings. Settling the lawsuit “was a business decision weighed with the cost of litigation on this matter,” said the bank in a written statement.

Lin had been targeted by a sophisticated criminal network that ran dozens of different scams and took in at least $800 million from January 2022 to September 2024, blockchain analysis TRM Labs did for AP/FRONTLINE showed. This industrialization of scams makes it even harder for victims to get their money back, said Ari Redford, global head of policy at TRM Labs.

“We have not built out systems in the U.S. in many respects to not only alert victims but to really do restitution in a meaningful way,” he said.

Lin said the grief and embarrassment made her think about taking her own life. But she was infuriated by a question her scammer asked her long after she’d stopped responding.

“Are you still alive?” said the message. Lin shuddered as she read it again.

“I told myself, ‘I’m alive and I’m going to fight back.’”

Global models offer potential answers

The U.S. is taking steps to improve cooperation with the private sector on fighting scams but — unlike a growing number of other countries — has stopped short of requiring companies to do more to protect consumers or face financial penalties.

Eight in 10 Americans across political lines said the U.S. government isn’t doing enough to protect them against scams, according to a new Gallup survey. And more than half said financial institutions, social media and tech companies and the federal government should also share responsibility, new AP-NORC polling shows.

Since late 2024, financial services companies in the United Kingdom have generally had to reimburse clients tricked into sending money to scammers, giving them an incentive to invest in prevention measures. The downside has been slower bank transfers, the need for more employees and sometimes a conversation to explain why a customer might want to transfer money.

The European Union is also rolling out rules that make financial institutions potentially liable for scammed funds if they don’t put in adequate fraud protections. And the EU’s Digital Services Act, which was passed in 2022, requires platforms to quickly act on reported scam content and take broader steps to reduce online fraud.

In the UK, trained social workers are sometimes sent to the homes of scam victims to address the emotional fallout.

“You wouldn’t ever say to a victim, ‘Why did you fall for a mugging? Why did you fall for a burglary?’” said Louise Baxter, who serves on the Home Office Joint Fraud Task Force. “It’s secondary victimization, from a law enforcement perspective and a societal perspective.”

New regulations in Australia also spread the responsibility for preventing scams. Financial institutions, telecommunications companies and digital platforms in Australia can be fined or forced to compensate victims if they don’t do enough to prevent and respond to scam activity.

Singapore’s protections are among the strictest. Under the government’s Shared Responsibility Framework, banks and telecom companies may have to repay victims of certain phishing scams if they fail to implement required safeguards. The country’s Protection From Scams Act passed last year allows police to temporarily restrict bank transfers of someone who they believe to be a victim. Further, staff from banks and e-commerce platforms physically sit in the same office as police in their national anti-scam center.

The U.S. makes piecemeal attempts to teach law enforcement agents how to respond to scams, but requirements vary across agencies. And in the U.S., victims are generally only reimbursed if the money is taken from their accounts without their authorization, not if they were tricked into approving the withdrawals. U.S. law also offers social media companies broad protections from liability for the content they carry, even scams.

There are global differences when it comes to cryptocurrency as well.

China bans crypto-related businesses, while the European Union requires licensing, consumer protections and broad disclosures. But while the Trump administration has backed some cryptocurrency regulation, it has also promised to halt “aggressive enforcement actions and regulatory overreach” to promote innovation. And the GENIUS Act signed by President Donald Trump last year to regulate some cryptocurrencies did not require companies to return stolen funds to fraud victims – a gap that consumer advocates, prosecutors and some lawmakers have criticized.

Law enforcement officials say it’s harder to track, freeze or recover funds because institutional banking safeguards largely do not apply to cryptocurrency.

Unlike bank deposits, crypto assets are not backed by federal insurance, and so if an exchange fails, the money can be lost. Banks also require customers to provide identification, social security numbers and account details when moving cash, while cryptocurrency transfers can obscure the identities of those involved. Most significantly, many cryptocurrency exchanges operate across borders and through offshore entities where U.S. laws may not apply.

Locating stolen cryptocurrency takes time and money. Returning it is even more complicated. A cottage industry of attorneys specializing in tracing cryptocurrency is emerging, but they still need buy-in from overworked officials to obtain warrants.

This means that even if people can locate their stolen funds, it’s unlikely they’ll get them back.

After Brian Glick lost $575,000, he called the FBI, an elder abuse hotline, the Securities and Exchange Commission and the New York State Attorney General’s office, he said. He filed complaints at IC3.org and FightCyberCrime.org. He collected thousands of screenshots as evidence and turned them over to federal investigators. He even kept talking to his scammer to draw out more information.

But when the FBI went to cryptocurrency company Tether to try and freeze and claw back funds linked to Glick’s investment, Tether said it could not. The FBI agent asked Tether to reconsider, according to email exchanges shared with AP and FRONTLINE.

“I respectfully request that Tether take a second look at the below information in regards to freezing funds from bad actors …” said the email.

After being given more details, including Glick’s wallet addresses, Tether told AP/FRONTLINE that Glick’s funds were moved and commingled with other funds just hours after the money was stolen.

“By the time Tether was contacted, the two addresses contained a significant amount of unrelated, unattributed funds belonging to other users,” a statement from the company said. It said Tether could not “arbitrarily freeze funds of unrelated users without proper process.”

Tether CEO Paolo Ardoino told AP/FRONTLINE that there are countless examples of their work with law enforcement.

“I think that there is no company, even in the banking industry, in the traditional financial industry, that is so helpful as us,” he said.

Officials at the FBI and Secret Service said Tether works closely with law enforcement to combat crimes related to cryptocurrency. In a joint initiative, Tether, TRM Labs and TRON, a blockchain network, say they have frozen over $450 million in illicit funds worldwide since 2024.

Glick was furious.

“There are so many victims of these cybercrimes,” Glick said. “And we can’t get our money back.”

Steps taken in the US

The U.S. government is waking up to the threat.

Congress is now considering more than a dozen bills to prevent scams. One would establish a centralized website for complaints, ReportScams.gov. Another would require disclosures on deepfakes and other AI-generated audio and visual content.

The Justice Department in November unveiled a strike force to cut off access of Southeast Asian operations to U.S. infrastructure, pursue criminal networks behind the scams and seize stolen funds. The Treasury has levied sanctions in Southeast Asia, and Myanmar and Cambodia have staged high-profile crackdowns on scam compounds.

The Justice Department’s Scam Center Strike Force claimed on its website this summer that it had restrained $832 million in cryptocurrency from Chinese transnational criminals. But a spokesperson told AP/FRONTLINE it was unable to go into any specifics about each seizure, where the funds are held or what, if anything, has been returned to victims.

Trump also signed an executive order in March directing the attorney general to prioritize the prosecution of scammers and to submit a recommendation for a program to restore money to victims.

“President Trump is unleashing every available tool to stop criminal networks that exploit vulnerable Americans through cyber fraud and scams,” the White House said in a statement to AP/FRONTLINE.

However, advocates say funding hasn’t flowed to the federal agencies tasked with combating scams, investigating cybercrime, and developing a cohesive plan for attacking the problem. And the fledgling efforts to help victims get money back aren’t keeping up with the tsunami of those who have lost it.

At least 13 federal agencies touch on different aspects of these crimes, according to a report issued by the Government Accountability Office. Seto Bagdoyan, who runs the office’s Forensic Audits and Investigative Service, said this piecemeal approach results in a sluggish strategy that “falls short.”

“There is no government-wide estimate of the money lost to scams, no common definition of scams, and no national strategy for combating them,” he said.

FBI Financial Crimes Section assistant section chief Rebecca Keithley, who retired earlier this year, said tens of billions of dollars flow out of the U.S. economy through scams. The FBI’s Operation Level Up, an effort to identify victims being defrauded and call them to intervene, has stopped about 8,500 people from falling for scams over almost two years, she said.

But that’s a small fraction of the sheer volume of victims. The FBI said it receives nearly 3,000 internet crime complaints a day on average through the IC3.gov online portal for reporting cybercrime.

“It’s an enormous problem, and one that the FBI is prioritizing,” she said.

Donna Gregory, retired FBI unit chief with the IC3, worked on the portal for 23 years, from its conception in 2000. Gregory said the loss threshold for what law enforcement will pursue varies by jurisdiction, and that the FBI is only able to investigate roughly 10-12 % of cases reported on IC3.

“The empathy you learn to sort of ignore because you would drive yourself crazy,” said Gregory. “All these victims saying, ‘I’ve lost my life savings, I don’t know what to do.’… “My grandmother committed suicide because she lost everything’ or ‘I’m going to commit suicide,’ and it’s like, which one do I have to pick?”

Chris Scott, a casino dealer in Arizona, was scammed out of $400,000 by a man on a dating site, she said. She added that she went four times in person to her local FBI office, begging for help — but was told all she could do was file a complaint through IC3.

She did. But she heard nothing back, she said, not even an acknowledgment of receipt. She began to panic, and searched for help online.

Scott found three companies claiming to have a direct line to the FBI, each promising to recover her funds. Out of desperation she hired them all, she said, handing over a total of $23,000.

None of the companies were real. Scott was forced to sell her home to cover her debts. She is still paying off the $20,000 tax bill she received after draining her retirement accounts, she said.

“I’m just a small fish in a big pond,” Scott said. “All I wanted was to talk to someone and get help, but I’m nobody to them.”

—-

Mendoza reported from Europe, Asia and North America; Linderman, a former Associated Press investigative reporter, reported from Baltimore; and Kinetz reported from Rome; Lisbon, Portugal, and Washington, D.C. Huizhong Wu in Bangkok, Wisdom Howell and Amelia Thomson-Deveaux in Washington, D.C., and Jesse Bedayn in Austin, Texas, contributed to this report.

—-

This story is part of an ongoing collaboration between The Associated Press and FRONTLINE (PBS) that includes the documentary “Scammed,” premiering Tuesday, Sept. 29, on PBS and online.

—-

The Associated Press receives financial support from multiple private foundations. AP is solely responsible for all content. Find AP’s standards for working with philanthropies, a list of supporters and funded coverage areas at AP.org.

—-

Contact AP’s global investigative team at Investigative@ap.org or https://www.ap.org/tips/

This story was originally featured on Fortune.com

MacKenzie Scott has continued her giving spree to historically Black colleges and universities, and earlier this year she crossed a major milestone.

One of the billionaire philanthropist’s gifts made this spring, a $42 million donation to Elizabeth City State University on the school’s Founders Day in North Carolina, pushed her total giving to HBCUs well past the $1 billion mark. And that’s just part of Scott’s $26 billion philanthropic commitment since 2020, in which she’s donated to thousands of organizations focused on DEI, disaster recovery, community development, health, and environmental causes.

Elizabeth City State University Chancellor S. Keith Hargrove, Sr., expressed his “deepest gratitude” for Scott’s gift, saying she recognizes” the critical role that HBCUs play in expanding opportunity and strengthening communities.”

“Her investment affirms what we already know: that institutions like ECSU are powerful catalysts for change,” Hargrove said in a statement. “Gifts like this do more than provide resources; they accelerate momentum.” Scott’s donation will help the school’s ASCEND 2030 strategic plan, expanding opportunities for students and strengthening ties to the surrounding community.

Other billionaires like JPMorgan CEO Jamie Dimon have recently stepped in with major HBCU donations. He and his wife, Judith, recently announced a $12 million commitment to 12 HBCUs.

Scott, the ex-wife of Amazon founder Jeff Bezos, has been doling out major gifts since 2020, when she first began directing hundreds of millions of dollars toward HBCUs. One such gift was to Howard University, the alma mater of former Vice President Kamala Harris, Thurgood Marshall, and Toni Morrison (with whom Scott shares a deep connection). Scott, who is worth an estimated $37 billion, donated $80 million to Howard in November 2025, which was one of the school’s largest donations in its 158-year history. 

HBCUs MacKenzie Scott has funded—and how much

• Alabama State University: $38 million, largest in the school’s history (2025)
• Alcorn State University: $42 million (2025)
• Bowie State University: $50 million (2025); $25 million (2020)
• Claflin University: $20 million (2020)
• Clark Atlanta University: $38 million (2025); $15 million (2020)
• Delaware State University: $20 million (2020)
• Dillard University: $19 million (2025), $5 million (2020)
• Elizabeth City State University: $15 million (2020)
• Hampton University: $30 million (2020)
• Howard University: $80 million (2025); $40 million (2020)
• Lincoln University (PA): $20 million (2020)
• Morehouse College: $20 million (2020)
• Morgan State University: $63 million (2025); $40 million (2020)
• Norfolk State University: $50 million (2025); $40 million (2020)
• North Carolina A&T State University: $45 million (2020)
• Prairie View A&M University: $63 million (2025); $50 million (2020)
• Spelman College: $38 million (2025); $20 million (2020)
• Tougaloo College: $6 million (2020)
• Tuskegee University: $20 million (2020)
• University of Maryland Eastern Shore (UMES): $38 million (2025)
• Virginia State University: $50 million (2025)
• Voorhees University: $19 million (2025)
• Winston-Salem State University: $50 million (2025)
• Xavier University of Louisiana: $20 million (2020)

Why no-strings-attached giving is so rare and powerful

These donations share a commonality: They’re unrestricted, meaning schools can allocate them however they see fit, which could include funded scholarships, fortified endowments, attracted faculty, and bankrolled long-deferred facility upgrades. That flexibility, rare in philanthropy, is the cornerstone of what has made her approach so distinctive.

“She practices trust-based philanthropy,” Anne Marie Dougherty, CEO of the Bob Woodruff Foundation, previously told Fortune. (Scott made two major donations to a veterans-focused organization: $15 million in 2022 and $20 million in 2025). Noni Ramos, CEO of Housing Trust Silicon Valley, has similarly noted Scott’s donations are “unlike traditional funding processes,” which typically involve lengthy applications, specific restrictions, and reporting requirements. 

“Her style empowers organizations like ours to determine how best to direct funds quickly and innovatively to address pressing issues,” Ramos told Fortune in 2024.

Scott’s HBCU giving exists within a broader DEI-focused philanthropic strategy that has become increasingly pronounced as the Trump administration rolls back federal support for diversity-focused programs and institutions. In 2025 alone, she donated $70 million each to the Thurgood Marshall College Fund and the United Negro College Fund.

The personal experiences behind Scott’s $26 billion giving streak

Scott’s motivation for giving at this scale traces back, in part, to formative experiences during her college years. A dentist once offered her free dental work when he saw her securing a broken tooth with denture glue, and a college roommate loaned her $1,000 when she saw her crying about nearly having to drop out during her sophomore year.

“It is these ripple effects that make imagining the power of any of our own acts of kindness impossible,” Scott wrote in a December 2025 essay. “The potential of peaceful, non-transactional contribution has long been underestimated, often on the basis that it is not financially self-sustaining, or that some of its benefits are hard to track. But what if these imagined liabilities are actually assets?”

The simple pleasure of giving also plays a role. 

“Generosity and kindness engage the same pleasure centers in the brain as sex, food, and receiving gifts, and they improve our health and long-term happiness as well,” Scott said. 

A version of this story was originally published on Fortune.com on April 7, 2026.

This story was originally featured on Fortune.com

Even just talking about organ meats makes many people’s noses wrinkle, let alone actually buying and cooking them. They can be very tasty, but we understand it’s not everyone’s thing. However, here’s what you should know: Organ meats are nutrient powerhouses, with B vitamins, vitamin A, iron, and more. This is why beef organ supplements can be appealing: You get the benefits of eating organ meat without having to prepare it yourself at home. 

This type of supplement can also be seen as an efficient booster. Besides the energy boost that can come from consuming B vitamins, a beef organ supplement can be a great time-saver for people who are busy at work all day and may not be getting all the nutrients they need. We looked at several popular options on the market and asked our testing team to give them a try. Here’s what those testers said about the supplements (including notes on the varying smells), plus key takeaways.

Best Beef Organ Supplements of 2026

• Best Beef Organ Supplement Overall: Heart & Soil Beef Organs
• Best Value Beef Organ Supplement: Nutricost Grass-Fed Desiccated Beef Liver
• Best Beef Liver Supplement: Ancestral Supplements Grass-Fed Beef Liver
• Best Grass-Fed Beef Organ Supplement: Equip Beef Organ

Best Beef Organ Supplements Comparison Table

Best Beef Organ Supplement Overall: Heart & Soil Beef Organs

Our Thoughts on Heart & Soil Beef Organs

Heart & Soil’s Beef Organs supplement is designed to help you “revitalize your body, regain your energy, and reclaim your health.” The company says this supplement, which includes grass-fed beef liver, heart, pancreas, spleen, and kidney, contains high-quality nutrients that support overall well-being. This supplement also purportedly helps maintain healthy energy levels, boosts immune health and overall resilience, promotes muscle growth and post-workout recovery, and supports metabolic and cellular function.

Our tester, Andrew, first noted that the supplement comes in a glass jar, though he said, “I would prefer plastic if I were taking this on the go.” He did appreciate the “very clear” usage directions, adding, “It gives instructions to work your way up to the recommended dose if you have never taken beef organ supplements.” He rated the instructions a 5 out of 5.

Andrew said the capsules were pretty average in size, but “the serving size of six pills seems like a lot. I had to split them up into two batches.” For that reason, he rated the product a 1 out of 5 on swallowability. As for the actual flavor, “The pill has no taste; however, there is a faint earthy smell,” said Andrew. He said there wasn’t any lingering taste associated with the capsules and gave the overall taste a neutral score of 3 out of 5.

Our tester, who describes himself as a 39-year-old father of two who is active and always on the go, concluded, “These supplements were very easy to take and fit into my schedule since they did not have to be taken with meals. Other than the serving size—six capsules—being high, I feel like this product gave me a natural form of vitamins in an easy-to-use way.”

Specs:

• Key ingredients: Bovine liver, bovine heart, bovine kidney, bovine pancreas, bovine spleen 
• Third-party testing: Yes
• Dosage: 3,000 mg (all ingredients combined)
• Serving size: 6 capsules

Best Value Beef Organ Supplement: Nutricost Grass-Fed Desiccated Beef Liver

Our Thoughts on Nutricost Grass-Fed Beef Liver

The key ingredient in Nutricost Grass-Fed Desiccated Beef Liver Capsules is sourced from free-range, pasture-raised cows from Argentina. (If you’re wondering what “desiccated beef liver” is, “desiccated” simply means dehydrated or dried out.) As is typical with Nutricost, the product description is minimal and doesn’t include extensive health claims like some similar products on the market, though the company does say the supplement is made in a GMP-compliant facility.

Our tester, Meredith, examined the packaging and said it was “pretty bulky” and not ideal for travel. She also said it isn’t child-proof, and upon opening the bottle, she said, “It smells like fish food.” 

However, Meredith said the instructions were “super easy to read and understand,” and that they direct users to take four capsules per day. She rated the instructions a 5 out of 5.

The capsules themselves were “pretty decent-sized,” according to Meredith, though not overly large. They were easy to swallow, but the four-capsule serving size was a lot, said Meredith. She rated the product a 1 out of 5 for swallowability.

Nutricost’s Grass-Fed Desiccated Beef Liver Capsules had a perfectly fine and neutral flavor, noted Meredith—who rated the product a neutral 3 out of 5 on taste—but “the pills smell bad even though they don’t taste bad.” Overall, though, these capsules have the lowest cost per serving in the lineup, at just 33 cents. For comparison, the most expensive product in the lineup costs more than five times as much. For those who prioritize value at a low price point, we think this supplement stands out.

Specs:

• Key ingredients: Grass-fed desiccated beef liver
• Third-party testing: Yes
• Dosage: 4 capsules
• Serving size: 3,000 mg

Best Beef Liver Supplement: Ancestral Supplements Beef Liver

Our Thoughts on Ancestral Supplements Beef Liver

Ancestral Supplements Grass Fed Beef Liver offers “the full array of energizing B vitamins plus iron and magnesium,” says the company’s website. The product also includes vitamin A to “support fertility, glowing skin, and eye health,” and vitamins A, D, and K to help boost the immune system. Ancestral Supplements claims its Beef Organ supplement also supports fetal development, strong bones, strength and endurance, and detox.

Sophia, our tester, looked at the product packaging and described it as looking “industrial,” adding that the bottle itself was easy to open but not child-proof. When she actually opened the bottle, though, Sophia said, “My biggest complaint is the odor of the product.”

The instructions were easy to read and understand, with Sophia rating them a 5 out of 5. She noted the serving size is six capsules daily, adding, “That seems like a lot to me, but I guess it would equate to consuming the same amount of liver as if you were actually eating it.” 

The capsules themselves were powder-filled and “slightly squishy,” said Sophia. When she took them as directed, she said afterward, “Despite the smell, the capsules do not have any taste at all.” She rated the product a 5 out of 5 for swallowability and a 3 out of 5 for the neutral flavor.

Sophia was generally a fan of the product, except for what she described as the “awful” smell. “Other than that, it seems to be a quality product,” she added. She did question the cost, saying it seems expensive for a 30-day supply, but there’s a subscription option that helps lower the price.

Specs:

• Key ingredients: Grass-fed bovine liver
• Third-party testing: Yes
• Dosage: 3,000 mg
• Serving size: 6 capsules

Best Grass-Fed Beef Organ Supplement: Equip Beef Organ

Specs:

• Key ingredients: Organic bovine heart, organic bovine kidney, organic bovine liver
• Third-party testing: Yes
• Dosage: 2,001 mg (all ingredients combined)
• Serving size: 4 capsules

Our Thoughts on Equip Beef Organ

Equip grass-fed beef organs are marketed as “the first regenerative organic certified beef organs capsules.” The supplement contains freeze-dried liver, kidney, and heart, and Equip calls it “nature’s original multivitamin.” The product doesn’t contain any synthetic vitamins, fillers, or flow agents, and every batch is third-party tested for heavy metals, pesticides, microplastics, and pathogens.

Our tester, Andrew, first noted that the container and packaging were “a good size” and would be easy to pack for travel. He said, “The instructions are clear and easy to understand. It simply says take one serving or four capsules with food.” He rated the instructions a 5 out of 5 for being so easy to understand.

When he actually tried the product, he said the capsule itself was a “normal” pill that wasn’t too large in size, and because of the four-capsule serving size, Andrew took two at a time without issue. He rated this supplement a 5 out of 5 for swallowability.

Andrew gave the product another 5 out of 5 for its neutral flavor, saying, “These pills have no taste or smell.” He also said there wasn’t any aftertaste. 

Summing up his experience, Andrew said, “This product is easy to take, and I could see myself taking this once a day with dinner. If I were looking to take this on a daily basis, I think this product would be good, but the price is slightly higher than I would be looking to spend each month.” 

What are Beef Organ Supplements?

Beef organ supplements may be sourced from bovine organs like liver, heart, spleen, kidney, and pancreas, which are typically dehydrated, converted into a powder, and ultimately sold as capsules. Organ meats can be very nutrient-dense but not always easy to buy or cook, which makes a supplement an appealing option.

Why These Supplements Matter to Professionals 

Many high-performing professionals prioritize both career and health and look for efficient ways to optimize both. Certain supplements may be useful in such cases, like beef organ supplements. They offer a convenient way to get the dietary benefits of organ meat without going to the grocery store or doing any cooking.

How We Tested and Selected the Best Beef Organ Supplements 

We looked at many different factors when choosing the best options.

Taste and Swallowability 

Assessing flavor is an important part of our testing process. If a supplement doesn’t taste good, people are less likely to take it regularly. The same goes for swallowability: We asked testers to rate how easy it was to swallow each product because if the capsule size is overly large or the serving size is cumbersome, people are less likely to want to take the product.

Formula

We work with a registered dietitian to review formulas for safety. This includes making sure product dosages are clinically supported and in line with the recommended daily values, and that products don’t contain any undesirable ingredients.     

Third-Party Testing

The U.S. Food and Drug Administration (FDA) doesn’t approve dietary supplements for safety or efficacy, so ingredients and their amounts can differ from what’s on the label. Because of this, we look for products that have been third-party tested to confirm ingredient accuracy.    

Value

Here’s a cost breakdown for each product, including cost per serving—often a better indicator of value than overall price. Note: We also included the subscription cost for comparison, for people interested in longer-term use. 

Benefits of Beef Organ Supplements 

As Sydney Lappe, a dietitian based in St. Louis, Missouri, explains, “The biggest benefit of beef organ supplements is that they provide concentrated amounts of nutrients like vitamin B12, iron, zinc, selenium, copper, and vitamin A. For people who don’t eat organ meats, they can be a convenient way to get more of these nutrients.”

However, Lappe added that it’s important to separate nutrient content from health claims. “While organ meats are packed with nutrients, there’s little direct evidence that taking a beef organ supplement will boost energy, improve hormone health, or enhance performance in otherwise healthy adults,” explained Lappe, noting that the greatest value of taking beef organ supplements is that they can help fill nutrient gaps.

How to Take Beef Organ Supplements 

We recommend following the manufacturer’s instructions when taking beef organ supplements. Make sure to note the serving size—number of capsules—and whether the directions indicate an optimal time of day to take the supplement. Additionally, note whether the supplement needs to be taken with a meal or with a certain amount of liquid. 

What to Avoid in Beef Organ Supplements 

• Choose a product with a detailed ingredient list. “Look for a supplement that clearly lists which organs it contains—such as liver, heart, kidney, or spleen—and how much of each is included. Since each organ provides a different mix of nutrients, vague ‘proprietary blends’ don’t tell you much about what you’re actually getting,” said Lappe.
• Prioritize products that have been third-party tested. Lappe said, “It’s also a good idea to choose a product that’s been independently tested to confirm it contains what’s listed on the label and has been screened for contaminants. Brands that follow Good Manufacturing Practices are another green flag that the supplement is produced under established quality standards.”
• Avoid taking overly large doses or combining a beef liver supplement with other supplements that have vitamin A. “More isn’t always better,” noted Lappe. “For instance, beef liver contains a form of vitamin A that can build up in the body, so regularly taking large amounts, or combining a liver supplement with other vitamin A-containing supplements, can lead to excessive intake over time.”

How to Find the Best Beef Organ Supplements for You

Here’s what to look for when searching for a beef organ supplement. As always, we recommend speaking with a healthcare provider before starting any new supplement. Lappe also noted that most healthy adults don’t need a beef organ supplement just because organ meats are nutrient-dense. “The better question is whether you have a nutrient gap that actually needs to be filled,” said Lappe. 

Your Wellness Goals

Consider your unique health goals and medical needs when choosing a beef organ supplement. Additionally, we recommend speaking with your healthcare provider to determine whether you have a true deficiency and could benefit from a supplement.

Form

While this lineup includes capsules only, in general, you should consider which form you prefer and seek that out when looking for a supplement. If a product is easier to take, you may be more likely to establish a consistent routine.

Cost

Consider your target price range and whether you’re willing to spend more on certain things. Some people may want a product sourced from grass-fed bovine, for example, while others may want a product with the smallest serving size to avoid swallowing a lot of capsules.

Fortune’s Final Thoughts

Beef organ supplements can provide a range of important nutrients for people who may be deficient or who simply don’t eat enough foods with those particular nutrients. They offer the dual benefit of being much more convenient than procuring and cooking organ meats and offering a lot of nutrients—ideal for people looking to maximize their efficiency and health.

Frequently Asked Questions

This story was originally featured on Fortune.com

After a 20-year career at design software company Autodesk, Prakash “PK” Kota felt an urge for a career change, in part because he didn’t want to end his career having worked for only one employer. 

Upon ending his time at Autodesk with a seven-year stint as chief information officer, Kota made the leap to human resources software company UKG, where he has held the same role since April 2025. He quickly consolidated software vendors, oversaw the launch of 387 internal AI applications from more than 1,400 employee-submitted ideas, and spearheaded the internal creation of more than 12,000 AI agents across Microsoft, Google’s Gemini, and OpenAI’s ChatGPT.

“In the AI era, everyone is talking about how work will be reshaped; how do employees and workers coexist?,” says Kota. “HR tech is going to be a huge area of investment in every company.” 

AI’s ability to automate workplace tasks ranging from coding to customer service to marketing, while increasingly taking on more complex tasks via agentic AI, has put more pressure on businesses and their HR teams to envision a more collaborative workforce that blends humans and machines. There’s also far more demand for prospects who have AI skills at their fingertips: job listings that mention AI are growing almost eight times faster than the total employment market, according to consulting giant PwC.

And yet, studies consistently show that the workforce’s youngest cohort, Gen Z, is more worried about their prospects in an AI era than older workers. This is partly driven by the difficulty many are facing in landing their first entry-level job, and swirling headlines of massive job cuts at major employers including Meta Platforms, Verizon, and Oracle, oftentimes with internal AI adoption attributed as a root cause.

Even as major employers are overhauling their teams and quickly pivoting on the skills they desire from new recruits, a vast majority of HR leaders say they haven’t yet established a firm grasp on what their internal future workforce needs will be in the AI era. Only 11% of HR professionals reported having established strategic, long-term workforce plans that extend beyond a three-year window, according to a study of about 1,300 HR industry pros across ten countries published by consultancy McKinsey in June.

While AI was certainly high on Kota’s agenda when he joined UKG, he said his first major project was addressing the long-delayed IT integration projects that hadn’t yet been completed following the pandemic-era merger between Kronos and Ultimate Software. Those two companies united to form the then-newly created, privately held UKG in October 2020.

Within his first 90 days as CIO, Kota centralized the technology department to handle all systems, data, and AI across both legacy businesses. After just six months, UKG consolidated the company’s enterprise resource planning and customer relationship management software to Microsoft and Salesforce, respectively. Multiple data warehouses were also merged.

As UKG leans into AI, the company has made ChatGPT Enterprise and Google’s Gemini Enterprise widely available to all 14,000 employees, while the product and engineering team is also using Anthropic’s Claude Code tool. Kota says he’s mostly avoiding multi-year contracts, because the technology is evolving so quickly.

One of the more impactful internal applications of AI is UKG’s utilization of AI-enabled voice and chat agents to handle customer inquiries, with an estimated 27% of those calls now being addressed autonomously. Autonomous agents are also drafting customer materials to make it easier for human representatives to handle the calls they do have with customers, helping workers handle issues at a speedier pace while also giving them time to upsell UKG’s products.

This system is also continuously learning, says Kota, as the AI tool has created around 300 “case studies” that summarize a customer service issue that it newly learned to handle and explain how to address the problem. These case studies are then used by both the AI tool and human workers. Propensity AI models, meanwhile, rely on 300 unique signals to predict the probability that a customer may be willing to buy more from UKG and share those insights with the sales and marketing teams.

Kota says he looks at multiple metrics to gauge the success of an AI internal deployment. For coding, value is determined by not just the quantity that’s produced, but also by what product features are actually bought by customers. Within customer service, UKG is monitoring both overall productivity and also upselling and customer sentiment scores.

And while AI-enabled efficiency isn’t measured equally for every employee, Kota and UKG says they have measured that AI has added 8,500 hours in productivity each month.

UKG’s C-suite leadership is also encouraged to think about AI adoption through a so-called “T3” concept: talent, tools, and tokens. Business leaders need to allocate spending to all three, and Kota says they are best suited to determine what’s the right mix. 

“I don’t want to have a standard rule across the company about what the divisional token spend should be,” says Kota. “We should have an open mind with this concept.”

John Kell

This story was originally featured on Fortune.com

After a 15-year run as Apple’s CEO, Tim Cook has handed the reins to longtime insider John Ternus. It marks the end of an era for the world’s most valuable company, as Cook moves into the role of executive chairman. And reflecting back on his time at the helm of the $4.75 trillion tech giant, Cook shares one daily leadership habit that pushed him to strive for success. 

“For the past 15 years I’ve started just about every morning the same way,” Cook shared in a community letter speaking on his tenure as CEO. “I open my email and I read notes I received the day before from Apple’s users all over the world.”

Since 2011, Cook has led Apple through a global pandemic, supply-chain upheaval, and the rise of AI—and has read thousands of messages about the products all along the way. Apple’s new executive chairman said customers would “share little pieces” of their world—from close calls caught by Apple watches, to capturing memories on their iPhones, to how their work has been changed since the dawn of the Mac computer. 

Whether it be heartfelt stories or complaints about the products, Cook says he combed through the messages daily, and said it’s been an “honor and a privilege” to lead the company building transformative tech.

“In every one of those emails I feel the beating heart of our shared humanity. I feel a sense of deepening obligation to work harder and push further,” Cook continued. “But most of all, I feel a gratitude that I cannot put into words, that I somehow got to be the person on the other end of those emails.”

Cook reads messages at 4 a.m. and reads hundreds of emails daily

Before most people have woken up to start their workdays, Cook has already been reading customer comments online. The former CEO gets up right before 4 a.m. every day, combs through hundreds of customer feedback to emails to get a sense of what users are experiencing.

“I rise a bit before four, I like to take the first hour and go through user comments and things like that, and sort of focus on the external people that are so important to us,” Cook told Axios in a 2018 interview. 

Then after a one hour workout to manage his stress, Cook dives right back into what Apple buyers have to say. He begins his 5 a.m. ritual of 15 years: reading “the majority of” the 700 to 800 emails that hit his inbox everyday. While many CEOs would delegate their flooded inboxes to their assistants, Cook said that “It keeps my hands on the pulse of what customers are feeling and thinking and doing.”

Cook is one of many emails who have embedded emailing into their daily routine—while others have eliminated the burden altogether. 

CEOs are split on how to handle emails

Just like Cook, billionaire Mark Cuban religiously checks his inbox—and his email is even available to the public. 

Rather than sit through hours-long meetings, he prefers to sift through hundreds of messages; he starts his day checking his morning emails around coffee, showering, and taking his daughter to school. And just like the former Apple CEO, he gets in a workout session before getting back to his inbox. Having an assistant handle his messages would only “slow things down.”

“I receive around 700 emails a day and use three phones (two Android and one iPhone) to manage everything,” Cuban wrote to Business Insider in 2025—ironically, over email. “I’d rather get 700 to 1,000 emails than sit in long, boring meetings.”

Other business leaders are adamant on emailing around the clock—even if it bleeds into the weekend. Uber CEO Dara Khosrowshahi revamped the company’s work culture when he stepped into the role in 2017, quickly implementing new ways to turn things around. One such change included emailing back whenever the need arises. If employees are “not performing” well in the “really demanding” environment, then Khosrowshahi said they’ll be shown the door. 

“Part of working hard is sending emails to the team on a Saturday,” the CEO of the $155.7 billion ride-hailing giant said on the Diary of a CEO podcast earlier this year. “And if I don’t get a response on Saturday, sending them an email on Sunday with a question mark. What’s going on?”

Brian Chesky, the cofounder and CEO of Airbnb, prefers to unplug. He runs the company on his own hours, doing away with morning meetings before 10 a.m., and rarely touching his inbox. Instead of shooting out emails, Chesky said he opts to call and text: “Don’t apologize for how you want to run your company.”

“[Emailing] was the thing about my job that I hated the most before the pandemic,” Chesky told The Wall Street Journal last year.

This story was originally featured on Fortune.com

Italian fishermen for two years incinerated most of the invasive Atlantic blue crabs caught in the Po River Delta south of Venice. Few Italians had an appetite for the alien predator blamed for decimating their prized Manila clams.

While experiments in containment continue, including the release of tens of thousands of octopus predators, the industry’s response has been shifting to find new markets for the crustaceans. The strategy is already salvaging the livelihood of some fishermen and, some hope, may even rehabilitate the much-maligned crab’s image in Italy.

Fishermen pivot to salvage livelihoods

Like other members of his local cooperative, second-generation fisherman Federico Zago saw his income crash after the crab invasion. The Polesine Fishermen’s Cooperative Consortium’s revenues fell 75% to 15 million euros ($17 million) in 2024 and 2025; the cooperative’s active members also shrank from 1,500 to 850.

“We needed to change strategy, and create an opportunity from this disaster,” said Paolo Mancin, the cooperative’s president.

Desperate, they pivoted to exporting the unwanted crabs, with the aim of preserving the Manila clams — invasive themselves — and mussels that created decades of prosperity in the shallow lagoons of Veneto and Emilia Romagna.

Zago now pulls up cages full of crabs from his traditional fishing cabin built on stilts over the Scardovari lagoon in the Veneto region. And the consortium’s strategy is working; so far this year it has sold every crab it has caught, eliminating the need for incineration. It forecasts 20 million euros ($23 million) in revenue this year, most from sales of crab.

“Since we’re going to have to live with these crabs, I hope they become an opportunity,” said Zago. “I hope the market continues to grow.”

Climate change creates perfect conditions for invasive crab

The Atlantic blue crab has been present in the Adriatic Sea since at least the 1940s, arriving in ballast water from commercial vessels, according to Piero Genovesi, an invasive species expert at Italy’s environmental protection institute.

Their numbers remained contained until 2023, and local fishermen reported catching them only sporadically.

Then came the explosion. Crabs tore through fishing nets, devoured mollusks at an alarming rate and edged out the native green crab, a local soft-shell specialty.

Climate change played a key role. Higher water temperatures favor blue crabs’ reproduction, but the real trigger was likely drought-driven drops in the Po River’s levels. That allowed sea water to infiltrate freshwater tributaries across the peninsula, especially in lagoons and deltas where clam and mussel farms provided ample prey.

“Climate change may therefore have favored reproduction through both temperature and salinity, because it changed the salinity of the water,’’ Genovesi said.

Eradication program failed

Reeling from damage to the mollusk industry, the Italian government adopted a blue crab eradication program in 2024. Over two years, the government paid the Polesine to catch and incinerate nearly 2.3 million kilograms (5 million pounds) of crabs. Fishermen in the neighboring Sacca di Goro lagoon and the Orbetello lagoon in Tuscany did the same.

But with female crabs able to produce up to 8 million eggs every season, culling barely dented their numbers. The project has been shelved, said Enrico Caterino, the government’s extraordinary commissioner for the blue crab emergency.

University of Bologna scientists this summer released 150,000 baby octopuses into the Adriatic Sea on the Riviera Romagnola, some two hours south of the Po Delta, promoting their project with an impressive video showing an adult octopus grabbing and devouring a crab.

Most octopus babies aren’t expected to survive. And given their preference for rocky environments over sandy lagoons and the crabs’ reproductive prowess, fishermen and experts alike are skeptical.

“If an animal with such high reproductive potential finds suitable environmental conditions, predators may not be able to keep it under control,” Genovesi said.

Fishermen deliver crabs bound for Sri Lanka

Each morning, Italian fishermen unload crates of scuttling crabs from the backs of their cars and trucks, 100-150 kilograms (up to 330 pounds) at a time.

Crab sales may be keeping the fishermen afloat, but their earnings remain nowhere near those of the clam industry’s heyday. Crab fetches just 1.30 euros per kilogram, compared with 10 euros for clams, which once generated 90% of their income but now account for just 10%.

Their crabs are steamed, frozen and packed into boxes to be shipped to Sri Lanka, where workers remove the meat by hand. The joint venture between Sri Lankan company Taprobane Seafoods and the Scardovari consortium churns out a range of crab products, from claw meat to mixed meat and fillet. Most is exported to other markets, with just a small quantity returning to Italy. Taprobane is forecasting large-scale retail expansion in Italy and Europe.

Mancin says he hopes the meat that arrives packed from Sri Lanka will help win over Italians who have so far resisted the “If you can’t beat ‘em, eat ’em” campaign launched in 2023.

Working to change Italian tastes

Despite the efforts, blue crab is mostly absent from top restaurants around Venice that specialize in the native green crab, due both to the labor involved in processing the invader species and its bad rap, according to the FIPE Venice restaurant federation.

The Yogi restaurant and bar in Porto Tolle on the Po Delta is an exception, having fully embraced the blue crab out of passion and necessity — finding themselves empty-handed after the invasion.

Owner and chef Stefania Marchesini, who fished professionally herself for 11 years, pays a premium to local fishermen, especially for soft-shell crabs known in the local dialect as moeche.

She offers an all-crab menu in her family-run restaurant, featuring fried crab balls, creamed crab on polenta, soft-shell crab and sectioned crab in the shell — each recipe capturing the crab’s delicate flavors.

“At first, the customer is very skeptical. But then the reaction is completely positive, and they want to try the entire range of products,” she said.

Marchesini also has a separate workshop where she prepares crab to sell to other restaurants — but business has so far been disappointing. Her commercial fare includes moeche individually packed in plastic envelopes, and full crabs already broken apart, under vacuum, to make them easier to prepare and less messy to eat.

Even at restaurants where the products sell out, owners are reluctant to restock. Marchesini chalks it up to cultural resistance, but she is undeterred. Recently she brought her menu to Barcelona’s Alimentaria, one of the world’s largest food and beverage trade fairs.

“We are all angry because it invaded our seas,” Marchesini said. “But we also have to eat. Once people understand how it should be prepared, they will realize it is an excellent product.”
___

The Associated Press’ climate and environmental coverage receives financial support from multiple private foundations. AP is solely responsible for all content. Find AP’s standards for working with philanthropies, a list of supporters and funded coverage areas at AP.org.

This story was originally featured on Fortune.com

Good morning. Companies are spending more on AI, and employees overwhelmingly say the technology makes them more productive. So far, those gains aren’t translating into operating profit for most companies.

Nearly nine in 10 respondents say their organizations regularly use AI for at least one business function, and 44% say they are scaling it enterprise-wide, according to a McKinsey survey of 1,719 professionals and business leaders globally. Eighty percent say AI has improved their individual productivity, and half say it helps them make better decisions.

But just 37% say AI is having a meaningful impact on earnings before interest and taxes (EBIT), unchanged from a year ago.

AI is also taking up a larger share of corporate budgets. Twenty-eight percent of respondents say AI now accounts for more than 10% of their IT budget, and 60% expect to increase AI investment next year. Twenty percent say AI-related operating costs, including token costs, are already constraining their use of the technology.

The companies reporting the largest financial gains offer one clue about where that spending pays off.

What McKinsey calls “AI high performers,” the 6% of organizations that attribute at least 5% of EBIT to AI, are far more likely to redesign workflows rather than add AI to existing processes. Nearly three-quarters have fundamentally redesigned workflows, up from 55% last year.

Company size also shows up in the results. Fifty-four percent of organizations with more than $1 billion in revenue report scaling AI enterprise-wide, compared with a third of smaller firms. Among large companies, the share scaling AI agents in at least one function rose from 27% to 40% in a year, while the share among smaller companies remained at 22%.

For finance chiefs weighing another year of higher AI spending, McKinsey’s findings suggest that increasing the investment is the easy part. The next AI budget may be as much a workflow budget as a technology budget.

Sheryl Estrada
Sheryl.Estrada@fortune.com

This story was originally featured on Fortune.com

They say you can’t put a price on love—but in today’s economy, it might be common sense.

Single U.S. adults were asked in a new survey how much they ideally would like potential partners to earn—and were given the caveat that the expectations had to be “realistic.” Respondents who said the income of a would-be partner is important to them, on average, said $139,000 per year was an acceptable salary.

According to the study from insurance giant Northwestern Mutual, Millennials (aged 30 to 45) had the highest income expectations for a potential attachment, at $160,000. Meanwhile, Gen Z (currently aged between 14 and 29)—who are earlier in their careers—said $135,000 was an ideal sum. The survey spoke to those aged 18 and up.

Older generations had slightly lower expectations: Gen X (41 to 64 years old) said $123,000 a year is ideal, while Boomers (65 and up) said $125,000 was a healthy salary.

These expectations perhaps reflect the arc of individual earning expectations. Data from the St Louis Fed, analyzed by Investopedia, shows the median annual earnings for those aged 35 to 44 in the U.S. is $72,020. That is followed by those aged 45 to 54, who earned a median salary of $71,604. Ages 65 and upwards earned a median in the $60,000s while ages 25 to 34 had a median income of $59,800.

While the bar of expectations follows the trajectory of earnings for Americans as they age, the gap between “ideal” earnings for partners is significantly ahead of the real world.

Further gaps between expectation and reality appear when broken down by gender: Men ideally would like their partners to earn $101,000 a year, the study finds, while women would like partners to earn $172,000 a year.

One reason for the differing opinions may be due to the gender pay gap. While the Northwestern study does not define the gender of would-be partners, data support the notion that heterosexual women may be accustomed to male partners earning more. The Bureau of Labor Statistics (BLS) says that the median usual weekly earnings of full-time wage and salaried workers for men is $1,380 as of Q2 2026. For women, it is $1,131. This divide also widens as individuals reach peak earning potential: The BLS data shows that for men aged 25 and up, their median weekly earnings were $1,459 in the second quarter of 2026, and $1,171 for women.

Pew Research wrote last year that the gender pay gap has only fractionally closed over the past 20 years:  In 2024, women earned an average of 85% of men’s earnings, while in 2003, women earned 81% as much as men.

However, there’s a silver lining for single people who neither earn nor want to earn six figures: 59% of singles don’t believe a potential partner’s income is important.

The money talk

Younger people were talking to their partners earlier about finances, and said financial compatibility was more important than emotional chemistry, physical attraction, or shared interests, the Northwestern study (carried out by The Harris Poll) showed.

Compared to older generations, Gen Z and Millennials were more likely to talk about finances with potential partners in the first few days or weeks of a relationship (10% and 11% respectively, compared to 6% of Gen X and Boomers). They were most likely to talk about money when the relationship got “serious” but before moving in together or getting married.

They were also less likely than Boomers and Gen X to wait until marriage or cohabitation to have a financial chat (15% of Boomers waited that long, compared to 6% of Millennials).

With young people facing an increasingly challenging economy, the Northwestern study also reinforced a familiar trend: Milestones being pushed back, or potentially never hit, by younger people. 31% of Gen Z and 24% of Millennials reported financial challenges were an obstacle to buying a house, while 24% of Gen Z and 14% of Millennials said finances may delay or cause them to decide against having children.

This story was originally featured on Fortune.com

Good morning. On Fortune’s radar today:

• Exclusive: Fed’s Austan Goolsbee says AI may cause ‘aggregate overheating.’
• Data centers are an ‘anchor’ around the neck of Republican midterm candidates.
• We’re about to hit peak capex growth—Wells Fargo.
• Iran, U.S. intensify retaliatory strikes in ‘tanker for tanker’ war.
• Who stole 33,984 cans of Pabst Blue Ribbon?

➡️ 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

Radical opinions on the economy aren’t hard to come by right now: Tech titans, economists, Wall Street giants, and politicians have all suggested that AI will be transformative, but can’t agree on whether it will be for good or evil. Extreme divides in opinion are also appearing over policy: be it trade, immigration, interest rates, or national debt.

The Federal Reserve Bank of Chicago’s President and CEO, Austan Goolsbee, is less concerned with prophecies of chaos or prosperity—he’s more focused on the reality of American consumers, and the businesses they work for.

In a landmark Jackson Hole speech last week, new Federal Reserve chairman Kevin Warsh said inflation was the sharpest focus, at present, for the rate-setting Federal Open Market Committee (FOMC), of which Goolsbee is a member. The reason: Price rises are comfortably ahead of the central bank’s mandated 2% target, pushed higher by supply-side shocks like the Middle East oil upset and tariffs.

Speaking exclusively to Fortune, Goolsbee agreed with the balance of concerns Warsh laid out: “On the real side, we’ve been stable, now inching toward dangers of overheat, and on the inflation side, after a couple of years of strong progress, it stalled out and started getting worse. But, we’ve had one encouraging report, one OK report, and now our challenge is … the inflation.”

The employment side of the Fed’s mandate looks relatively stable across data points such as the unemployment rate, vacancy rate, hiring rate, and the layoff rate, Goolsbee explained: “I think that’s been largely—not the result of AI data centers, for much ballyhoo—it’s the U.S. consumer. Broad-based consumer spending growth is the thing that has kept the economy solid and stable.”

And while everyone is closely watching AI’s impact on the labor market—be it layoffs, productivity gains, or demand for certain skills needed for infrastructure—Goolsbee is particularly concerned about potential economic overheating. He explained: “I would characterize the expansion of the data centers as very hot, but largely shoving other parts of the economy down.” 

“The rise has been stepping on others—they’re competing for the resources. When I’m touring around the 7th district, people [are] saying: ‘We’re having to scale back our plans because getting construction workers is too expensive, you can’t get HVAC,’ etc. That implies a sector rebalance, that is different from an aggregate overheating, [but] that said, we’re not far from that turning into aggregate overheating.”

If the impact of data center buildouts spreads further through the economy—pushing up services inflation, for example—”that would make me more nervous,” said Goolsbee. The 10th president of the Chicago Fed explained he was comfortable with holding the base rate at the last FOMC meeting in July because the most contemporary inflation data had shown some improvement (the all items CPI reading was down 0.4% in June, and a flat 0.1% in July) “so it makes sense to wait and see if this has legs, or is just a blip.”

Promises vs reality

AI has produced some sparkling outlooks for the global economy: Nvidia’s Jensen Huang believes that while there will be some labor disruption, the technology will be a net job creator “at a scale that we have never seen.” Tesla CEO Elon Musk said AI will make money essentially irrelevant and work a hobby, while Meta’s Mark Zuckerberg wrote that AI, in a healthily balanced economy, will see “overall productivity and innovation increase while employment levels remain high.”

Meanwhile, Goolsbee is thinking about the “grubby day job” of the Fed, saying the impact of AI on monetary policy depends on how expected the outcomes are.

“If the productivity lands on us in an unexpected way, inflation goes down, and rates can go down,” Goolsbee explained. “But the more expected it is, and the bigger the hype … it leads to just old-fashioned overheating in the short run, because equity values go up and so the businesses launch massive capital investment in the here and now, people start spending out of their equity-well in the here and now, before the productivity bounty has arrived.” 

“It’s fun to engage in the academic exercise of dreaming [about] how AI will affect the long-term growth in the United States or the global economy, and we should think that through, but for the grubby day job of the Federal Open Market Committee, it’s not at all clear.”

There’s also the question of when these massive gains come to pass. In 1987, Nobel Prize-winning economist Robert Solow remarked: “You can see the computer age everywhere but in the productivity statistics”. This is the basis of the Solow Productivity Paradox, suggesting that productivity may slow—or at least lag expectations—with technological advancements.

In a July study, the Fed noted that while sectors with more exposure to AI have higher productivity growth, trends across organizations with low, medium, and high exposure levels remain consistent over time, “suggestive of micro-level productivity gains not adding up in aggregate.”

Goolsbee is mindful of this delay and, therefore, wary of applying it too swiftly to the current outlook. “It strikes me it’s a bunch of great technologists who are coming up with this, and they’re wanting to declare themselves economists, and you’ve already seen it not play out the way that they said,” Goolsbee added. “I would like them at least to acknowledge that in the last 10/15 years—this may be the most extreme version—but they have declared numerous technologies were going to totally change the world and displace millions of jobs, and we’re still waiting for those ones to happen, whether from autonomous vehicles to NFTs and blockchain … we’ve had a series of those.” 

 “It’s not to make light, it’s clear that in some sectors the adoption has been so rapid that they’re feeling the pinch, but I don’t believe that the low hiring rate is predominantly caused from AI.”

‘Traaaaaaansitory’ supply shocks

A more unexpected delay between theory and real-world data is emerging in the form of supply shocks. During the tenure of Jerome Powell, the previous Fed chairman, the FOMC was repeatedly urged to follow Econ 101 and “look through” inflationary pressures, as they stemmed from seemingly one-off supply shocks (such as tariffs or the Middle East conflict) rather than underlying trends pushing prices higher.

But Goolsbee argues that, particularly since the pandemic, supply shocks large enough end up being substantially more persistent than theoretical models suggest. One factor is that current supply shocks aren’t “one and done” because they arise from ongoing issues like geopolitics. Another factor is that huge geopolitical supply chains take longer to fix than they did previously.

It is for the FOMC to balance the risk of reacting to transitory supply shocks (or, as Goolsbee describes them, “traaaaansitory” supply shocks that drag on) against the risk of enduring above-target inflation. This, ultimately, belies the biggest threat Goolsbee sees to the economy.

“In terms of public attention, it has shifted to data centers, and it feels like that’s all anyone wants to talk about, but I would like to shift it back,” Goolsbee said. “The thing in my mind that has made the economy stable and growing—despite a series of pretty intense shocks—… is the unrelenting, continued consumer spending.”

“If you go out here in the Midwest, prices are on everyone’s mind, and if there were going to be something that shook the consumers, it would shake the economy. If we hit a hiccup on consumer spending, to me, that is the biggest risk to continued stability and growth.”

Goolsbee advocated for attention focused on “old school” economic barometers: “What’s consumer spending and is the consumer going to keep up this pace?”

This story was originally featured on Fortune.com

With the U.S. Open underway, tennis’ biggest stars face off in New York City to hoist the silver trophy and vie to take home a multimillion-dollar payout. Among them was longtime legend Novak Djokovic, who played the first round of the tournament this past Sunday night but lost in a crushing upset to Argentine player Mariano Navone after nearly five hours of exhausting gameplay. Though he lost out on a shot at breaking the record for most Grand Slam titles ever (of which he owns), the tennis icon is still walking away with plenty of prize money on and off the court.

The Serbian star—who currently holds the men’s all-time record with 24 Grand Slam singles titles, and remains tied with Margaret Court for most titles in tennis history—lost his streak of 78 first-round wins at Grand Slam tournaments. After he lost in five grueling sets against Navone that went on until nearly midnight, Djokovic was seen limping and cramping on the court, throwing up at least once as he continues grappling with various years-long physical ailments. That night, the 39-year-old said that his body started to “collapse” from the severity of health issues, with his back and shoulder giving out towards the end of the match. Meanwhile, young-and-scrappy Navone—a 25-year-old unseeded underdog—was able to clinch the win after an intense battle against one of the sport’s most decorated champions.

This was the first time Djokovic lost a first-round match in two decades since the 2006 Australian Open. And with 101 ATP titles and 24 Grand Slams singles titles under his belt, Djokovic’s defeat came as a shock to many—even if he’s at an age when many tennis stars have already retired. And even though he didn’t make it past the first round, Djokovic won’t be walking out of Arthur Ashe Stadium empty-handed. 

Every U.S. Open singles player receives at least $140,000 for getting to the first round of the main-draw—win or lose. That means Djokovic will be walking out with a six-figure pay day, on top of his off-the-court earnings. He’s currently the sixth highest-paid star in tennis, according to Forbes, earning $5.2 million on-court and $25 million on the side last year for his various business ventures. Djokovic currently oversees a diverse sponsor portfolio—from Lacoste and Qatar Airways, to Hublot and General Atlantic. 

The U.S. Open is shelling out $108 million in player compensation

While the U.S. Open is about winning a coveted Grand Slam title, it also comes with plenty of money. Some of the world’s most iconic players, including Carlos Alcaraz, Coco Gauff, and Aryna Sabalenka, will be taking to the courts over the next two weeks to compete for glory and millions of dollars in prize money. And on top of multi-million-dollar brand partnerships and endorsements, the final players stand to win eye-watering checks from the tournament.

During the qualifying rounds, players have the potential to earn between $32,000 to $66,000—but the real money comes during main-draw matches.

The winners of the U.S. Open women’s and men’s singles can earn up to $5,500,000 if they win the entire tournament, while doubles and mixed doubles champions split prize money as high as $1,000,000. The pay shoots up by hundreds of thousands of dollars in later rounds, starting from Djokovic’s $140,000 and up, per the U.S. Open. The association is also shelling out $37,840,000 in total for those who place in the singles rounds, and $13,000,000 for all the athletes across the doubles games. And zooming out on the honeypot in its entirety, total player compensation at this year’s competition hit a whopping $108,000,000—a 20% increase from last year’s $90 million—and what the U.S. Open characterizes as the largest in tennis history.

Here are the 2026 U.S. Open prizes for women’s and men’s singles: 

• Winner: $5,500,000
• Runner-up: $2,800,000
• Semifinalists: $1,450,000
• Quarterfinalists: $780,000
• Round of 16: $480,000
• Round of 32: $290,000
• Round of 64: $190,000
• Round of 128: $140,000

While a multimillion-dollar payout would be life-changing for most, the prize money pales in comparison to most athletes’ off-the-court earnings. Italian player Jannik Sinner is the highest paid tennis star according to Forbes, earning about $58 million annually, with $35 million of that made outside of the arena. The 25-year-old, five-time Grand Slam champ has a partnership with insurance giant Allianz, and sponsors brands including Gucci and Lavazza. Similarly, 2025 U.S. Open men’s singles winner Carlos Alcaraz earns $38 million of his $53.7 million off-the-court, while icon Serena Williams makes the vast majority of her $40 million annual earnings through investments and brand partnerships.

This story was originally featured on Fortune.com

President Trump took to Truth Social this week to warn any town still weighing whether to fight a data center. Communities that turn them away, he wrote, are choosing to be “backwards and poor.” The ones that welcome them will get lower taxes and jobs “all over the place.” He called the industry the “Golden Goose,” said plenty of other places would happily take the investment instead, and warned that China “could not be happier” watching American towns turn it down.

He’s not wrong: Data center spending has become one of the single largest forces holding up U.S. economic growth this year. That’s normally the kind of growth that a president looking to maintain party control of both chambers would want to campaign on. But instead, it’s become one of the few issues in American politics that Republicans and Democrats can agree on, and what they agree on is that they simply don’t want it near them. With just four months before the midterms, no candidate wants to touch the issue with a 10-foot pole, and if they do, they’re spending millions in campaign funds to get the issue “right.”

Why data centers are so polarizing

Few issues cut across party lines the way data centers do, because what makes them valuable nationally is what makes them hated locally. One facility can draw as much power as a mid-size city and remake a town’s tax base within months, and neighboring households bear the costs. Conservatives who normally favor deregulation are showing up at town meetings over property rights and distrust of distant tech billionaires. Progressives who favor the technology’s economic promise are objecting on environmental grounds. A CNN analysis found the backlash is bipartisan because the fight is so local, meaning it skips past national ideology.

It also comes with an economic argument. In the first quarter of 2026, AI-related computing infrastructure investment hit roughly 1.4% of U.S. GDP, up from 0.7% a year earlier. It’s now the largest driver of growth in U.S. private investment, according to Epoch AI. The St. Louis Fed found information-processing equipment made up 39% of total GDP growth through the third quarter of 2025, a bigger share than during the dot-com boom. Data centers’ dollar contribution to GDP growth had passed consumer spending for the first time ever, which is remarkable given consumer spending typically makes up about two-thirds of GDP.

Construction is real and large: a $10 billion campus in Lebanon, Indiana employs more than 4,000 workers at the height of the build. But once the servers are running, that project keeps around 300 permanent employees—13 construction jobs for every one that lasts. A typical data center supports fewer than 200 local jobs long-term, Fortune reported, citing the U.S. Chamber of Commerce. Research from Virginia, the country’s biggest data center market, found projects there create one permanent job for every $54 million invested, according to a MinnPost fact-check; across the broader economy, $1 million in investment supports 17 jobs on average. Data center construction rarely produces a lasting rise in local employment, Fortune has noted, which helps explain why the tax breaks used to attract these projects have drawn bipartisan scrutiny: at least 10 states are losing more than $100 million a year in revenue from data center tax incentives alone.

The case for jobs

As the industry continues to contribute to a growing share of U.S. GDP, its physical footprint in the country is growing with it—as is the backlash against it. Part of the sentiment behind that opposition isn’t just about power bills: it’s about jobs disappearing because of the same investment.

Nearly 200 economists and researchers warned in July that AI could cause large-scale job displacement over the next decade, calling it a transformation on the scale of the Industrial Revolution but compressed into a much shorter span. And the same companies pouring billions into data centers have kept cutting jobs elsewhere: Microsoft laid off nearly 5,000 people in early July even as it continued pouring billions into AI data centers. For plenty of voters, that looks like the same industry taking jobs away with one hand while promising them with the other.

For their part, the industry points to the very real spike in construction jobs: a $10 billion campus in Lebanon, Indiana, employs more than 4,000 workers at the height of the build. But once the servers are running, that project keeps around 300 permanent employees—13 construction jobs for every one that lasts. A typical data center supports fewer than 200 local jobs long-term, per the U.S. Chamber of Commerce. Research from Virginia, the country’s biggest data center market, found projects there create one permanent job for every $54 million invested. Across the broader economy, $1 million in investment supports 17 jobs on average.

Data center construction rarely produces a lasting rise in local employment, which is part of why the tax breaks used to attract these projects have drawn bipartisan scrutiny: at least 10 states are losing more than $100 million a year in revenue from data center tax incentives alone.

Swing-district fights

Pennsylvania’s governor’s race shows how fast the politics can flip. Last year, Gov. Josh Shapiro was the industry’s biggest booster, touting a $20 billion Amazon commitment in the state. By August, he’d signed an order stripping its fast-track permits and requiring local approval. His Republican opponent, Stacy Garrity, is now running to his left, attacking the Amazon deal in her first TV ad and accusing him of trying to “gaslight” voters. The same fight is playing out in Ohio and Texas.

In Ohio, Democrats are attacking Sen. Jon Husted over his record courting the industry as lieutenant governor, and the National Republican Senatorial Committee has warned the issue could cost him his special election, with private polling showing a dead heat against Sherrod Brown, per CNBC. In Wisconsin, Republican Tom Tiffany calls his Democratic opponent “Data Center David Crowley.”

In Georgia, data center opposition and rising electricity rates helped Democrats flip two Public Service Commission seats last year, and the party is betting the same anger carries into the governor’s race, where Democrat Keisha Lance Bottoms backs a moratorium and Republican nominee Rick Jackson opposes one. U.S. Sen. Raphael Warnock has called for a statewide pause, while Gov. Brian Kemp said that decision belongs to local communities. The fight has reached down to state legislative races too: one Democratic challenger southwest of Atlanta says lifelong Republican voters in her district have told her data centers changed how they see every other issue.

Where affordability and data centers collide

A July Pew survey found the cost of living is the top issue voters want candidates to address. Groceries, gas, and housing all feed that anxiety, but electricity is the cost voters can trace to one specific, unpopular neighbor. Utilities requested more than $30 billion in rate increases last year, hitting 81 million Americans, and power bills have risen 40% since 2021—the fastest stretch on record. Data centers drove about half of all US electricity demand growth last year, and PJM’s independent market monitor has tied data center demand to $23 billion in customer price increases through 2028.

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

There was a time when learning how to “speak office” meant, quite literally, being in an office.

A clueless rookie could pick up how their boss wrote an email, when a coworker switched from a quick message to a more formal one, or what someone meant by “run it up the flagpole.” But Gen Z entered a workforce where much of that learning can happen through a screen—and where language on either side of it isn’t always translating.

Knowing how to “speak office” has never just been about understanding the words your coworkers use. Workplaces come with unwritten rules about how to formally address someone, what belongs in an email instead of a Slack message, and even how familiar you can be with someone you just met. Gen Z began learning those rules as the pandemic and remote work scrambled the ways they had traditionally picked them up, while social media and instant messaging brought different communication habits to work. And of course—AI, as the kids would say, has “entered the chat” as another way to learn too.

The apparent language divide between generations is more complicated than whether boomers know what “delulu” means. The words are changing, but so are some of the rules around how people communicate at work, and not everyone agrees on what sounds appropriate or professional anymore.

It seems so silly on paper. Gen Z has “delulu,” “rizz,” and “no cap.” Older workers have “blue-sky thinking,” or brainstorming without practical constraints; “move the needle”; and the particularly puzzling “boil the ocean,” or taking on an impossibly large task. They’re noticeable enough that someone has now built a translator for them.

Sunny Workplace, a nonprofit employee engagement platform, recently launched WorkGen Translation, an online tool that decodes slang, idioms, and workplace jargon across age groups. 

This year, the company commissioned Censuswide to survey 2,004 U.S. workers on May 25 and 26, split evenly between Gen Z and boomers. Eighty percent of the 1,002 Gen Z respondents said generational differences in workplace language were a challenge, compared with 30% of the 1,002 boomers surveyed. 

The confusion goes both ways. Among specifically boomer respondents, 76% said they had no idea what “delulu” meant, 66% said the same of “rizz” and 61% of “no cap.” Gen Z respondents had their own trouble with corporate speech, with 22% said they knew exactly what it meant to “boil the ocean,” while 25% said the same of “blue-sky thinking.”

The misunderstanding that matters at work can be much more mundane. Betsy Sneller, an assistant professor of sociolinguistics at Michigan State University, pointed to something nearly every office worker uses: email.

“Younger folks entering the workforce don’t really have the genre of email in the way that older speakers do,” Sneller told Fortune.

Older workers are more likely to treat email as a relatively formal and permanent form of communication, Sneller said, while reserving Microsoft Teams or other instant messaging platforms for quick exchanges. 

Younger workers who grew up communicating through social media and messaging apps may not make the same distinction. That can mean skipping a formal greeting or punctuation in an email without giving it much thought.

“There can be sort of miscommunication where younger folks are being casual in an email, maybe not using punctuation, maybe not using a formal salutation, and it gets interpreted as very rude by older speakers,” Sneller said.

The unwritten rules of ‘speaking office’

A linguistic concept behind some of that disconnect is called a register, which focuses on how language use changes based on social context and audience. 

“We have a work register, we have a home register,” Renée Blake, an associate professor of linguistics and social and cultural analysis at New York University, told Fortune.

People routinely change how they communicate depending on where they are and whom they’re talking to. Home might allow one way of speaking; addressing a large group or reporting the news might demand another. Work has traditionally come with its own expectations. Blake sees what happens when those expectations collide in students’ emails. Instead of an introduction like “Good morning, Professor Blake,” she sometimes gets a simple “Hey, Renée.”

“Am I your friend? I’m not your friend, and you don’t know me,” Blake said with a laugh.

The issue isn’t simply that one way of speaking is right and another is wrong. Registers come with rules people become accustomed to, Blake said, and when someone steps outside the register expected in a particular setting, it can feel as though that person is breaking the rules of the space. 

That’s where communication differences can become a judgment about the person doing the communicating. Iain Smith, head of behavioral science at Sunny Workplace, said that can be particularly consequential for workers who are still learning the language of their workplace.

“If they haven’t learned the workplace vernacular, and they speak using phrases like ‘delulu’ and ‘rizz,’ for instance, I think there are judgments that can be made about professionalism based on that,” Smith told Fortune.

For younger workers, learning what those unwritten rules are may also look different than it did for the rookies who came before them.

“They’re not necessarily now sitting next to someone who’s been doing the job for 20 years and learning the shortcuts and the hacks and the little ways of coping with a stressful day,” Smith said. “They’re turning up to Zoom.”

It’s not just the kids

Gen Z entered work during that upheaval, but they weren’t the only ones adjusting to new ways of communicating; it’s more complicated than it looks.

“COVID also stunted all of us,” Blake said. “It’s not just young people.”

Workers across generations suddenly spent extended periods communicating from behind screens, and many returned to offices only partially, if at all. At the same time, technology increasingly competed for people’s attention as they figured out how to interact with one another again.

Blake experienced some of that confusion herself. After getting sick, she moved one of her college classes online and assumed she could be more informal with students she had only recently met. Some of them thought otherwise.

“This is too much, this is too informal. You don’t know me,” Blake recalled them telling her. “So this is going the opposite direction,” she added.

Even AI is becoming part of the negotiation. Blake said she sees young people using AI to learn forms of professional communication they haven’t necessarily been taught, including how to write more formal correspondence. But outsourcing that learning comes with its own tradeoff: The result can sound stiff, templated, and impersonal.

Changing ways people communicate don’t necessarily mean generations themselves are as different as workplace stereotypes suggest. A meta-analysis published in the Journal of Organizational Behavior found few systematic, meaningful differences among generations across a range of workplace outcomes.

Smith, a millennial, remembers when his own generation was the subject of sweeping theories about how it would transform—or ruin—the workplace.

“From a scientific perspective, there’s a lot of research out there that shows that actually generational differences in work values, work-life balance, stress, burnout, working preferences are all overhyped,” Smith said.

Sneller nevertheless believes communication styles between people entering the workforce today and those already in it are more pronounced than in the past. She pointed to a combination of social media and the pandemic, and sees some of the effects in her own classroom, where students sometimes need more encouragement to begin casually talking to one another.

And the line between the generation that needs translating and the one doing the translating doesn’t stay put for long. Sunny’s survey asked workers about Gen Alpha slang, offering Gen Z an early taste of being on the other side. Forty-one percent of Gen Z respondents didn’t know what “Ohio” meant, while 75% didn’t fully understand or said they wouldn’t use “skibidi.”

So Gen Z may not have to wait until middle age to start wondering what the kids are talking about—they may just have to wait for Gen Alpha to clock in.

This story was originally featured on Fortune.com

Another day, another dollar. But for John Ternus, who started Tuesday as Apple’s CEO, the amounts come with a lot more zeroes after them.

The Apple board handed the newly minted CEO of the $4.75 trillion iPhone maker a massive new compensation package to kickstart his tenure, with more to come if Apple’s pay history is any guide. Apple gave Ternus a salary bump to $3 million and granted him restricted stock units valued at $2.5 million, prorated for the few weeks he’ll serve as CEO before the end of Apple’s 2026 fiscal year. The board separately approved an annual equity award for fiscal 2027 with a target value of $55 million. Three quarters of the equity award will vest based on Apple’s total shareholder return relative to other S&P 500 companies. The rest will vest on a clock, 12.5% every six months over four years. 

Tim Cook, who after 15 years as CEO is sticking around as executive chair while Ternus finds his footing, saw his salary decrease from $3 million to $2 million, effective later this month. The Apple board approved a $45 million target equity award for him, half in RSUs that vest over four years, and half that vest based on performance. 

No word yet on an annual cash bonus for Ternus. Cook collected a $12 million bonus each of the past two fiscal years, according to Apple’s securities filings. 

The Cook Precedent

It was an entirely different picture the last time Apple had a CEO change, when Cook succeeded the late Steve Jobs in 2011. Back then, Apple gave Cook 1 million RSUs with a grant date value of $376.2 million on top of a $900,000 salary that the board bumped to $1.4 million a few months after he took over the top spot. Half of the award vested five years after the award date, and the other half vested 10 years after the award date. 

The Apple board’s decision at the time was “subjective” and not based on any peer group or formula. That was in part due to the circumstances: Jobs famously collected only $1 in salary, but held a colossal 5.5 million-share stake in Apple worth billions. The board said it determined Cook’s pay based on input from Jobs and the board’s assessment of Cook during his interim leadership periods during Jobs’ medical leaves. Jobs passed away in October 2011, following his Aug. 24 resignation that year. 

Two years later, Cook asked the board to add some performance conditions to his award, meaning he would forfeit a portion if he didn’t achieve a stock-performance metric. The board carved out 800,000 share units from his 1 million award and divided it into 10 annual chunks of 80,000 shares apiece split half on performance. If Apple finished in the top third of the S&P 500 by shareholder return, all 80,000 units for that year vested. If they finished in the middle third, it cost him 20,000 and the bottom third cost him 40,000 shares.

Ternus’ award is now based on the same type of performance metric. A securities filing on Tuesday, required when an executive joins officer ranks, shows Ternus holds about 34,000 Apple shares in a trust, worth about $11.1 million at Tuesday’s stock price. He also holds seven grants of restricted stock from his years running hardware engineering, covering about 305,000 shares worth about $99 million or more, depending on how the shares pay out. Nearly half are performance units that could pay double or nothing, which puts the payout range between $50 million and $148 million.

This story was originally featured on Fortune.com

The music you hear as you walk through Singapore’s Changi Airport can’t be found on any playlist or album. Instead, it’s being composed in real-time by a computer program.

“If you’re walking through Changi, you’ll be hearing a soundtrack that is literally AI,” Michael Boyle, HP’s senior vice president and managing director for Greater Asia, told Fortune in early August. “The great thing about AI is that you can shape it. If it starts raining outside, the airport gets busy, or it’s nighttime, the AI can adapt the music.”

The company behind those tunes is Wubble AI, a Singapore-based startup that participated HP’s “Garage 2.0” program, an accelerator that the company started in the Southeast Asian city in Oct. 2025. HP became a backer of Wubble AI in July.

Garage 2.0 harks back to HP’s roots in Palo Alto, where founders Bill Hewlett and Dave Packard started the company in the latter’s garage in 1939. HP runs Garage 2.0 with Antler, the global venture firm that has made more than 1,400 investments across more than 30 markets. 

“We’re looking at these founders and the speed that they can go without the complexity of being in a large organization, and it’s inspiring,” Boyle said. “It gives those of us working in a large, matrix organization the chance to come back down and think about what’s really going to cut through and build scale.”

Yet the Garage 2.0 program gets at a deeper question at the heart of HP: Is the tech company that’s perhaps most responsible for Silicon Valley’s culture getting left behind? HP is now No. 276 on the Fortune Global 500, down from No. 194 in 2017. The company is now betting on AI PCs as a new growth category, wagering that consumers will prefer to run AI on their own devices rather than in the cloud.

HP’s Asia bet

HP’s footprint in Asia began almost 60 years ago, when the company established its first regional office in Singapore in 1969. A year later, it opened its first Asian manufacturing plant in Singapore’s Redhill neighborhood, where just over 60 workers assembled iconic devices like the HP-25 calculator. 

The company operates across most Asian markets, with manufacturing and supply chain hubs in China, Thailand, and Vietnam, and major call centers in Malaysia and India. The region contributed just over 25% of HP’s revenue in its most recent quarter, which covered May, June and July. 

Boyle, who joined HP in 2015 as the vice-president of its Asia-Pacific and Japan (APJ) operations, is most attracted by the region’s diversity. “There’s many different versions of Asia,” he said. “It’s been really interesting getting to see how HP is represented in its different markets.” For example, he notes that Southeast Asian retail still relies on brick-and-mortar stores, which means most sales happen face-to-face; in contrast, online sales now dominate in markets like South Korea.

In what he describes as a “serendipitous” bit of foreshadowing, Boyle wrote his graduate thesis on HP’s strategy long before he ever worked there, while completing an MBA program at the Macquarie Graduate School of Management in Sydney.

“I visited their European operations in Germany as part of a study tour, and there was something about HP back in those days that I really identified with,” Boyle recalls. “So it was quite serendipitous that the phone rang, and all of a sudden I found myself living in Singapore and working for HP.” 

Bullish on print

HP is perhaps best known for its printers, which have long been a mainstay of offices worldwide. Boyle pushes back on any suggestion that print is dying, calling it a mainstay in sectors like education and healthcare.

Since last year, the firm has been rolling out AI-powered printers that come installed with Microsoft Copilot, which allow users to print documents from a shared cloud without ever booting up their laptop, as well as generate AI summaries from documents scanned using the printer. 

“AI is now starting to play a role where it can go both ways: from creation to print, and print back to creation,” Boyle said. “You can even edit on the fly.”

Despite Boyle’s optimism, printing remains a shrinking part of HP’s business. Total print revenue fell by 2.2% in HP’s most recent quarter. The company blamed “demand softness, particularly in China,” as well as “changing consumer behaviors.”

Pushed to the edge

As some of HP’s mainstay businesses show their age, the company is now looking to AI PCs as its next source of growth. 

AI PCs are devices with some capability to handle AI and machine learning tasks on the device itself, rather than relying entirely on the cloud. HP first unveiled its suite of AI laptops in 2024, including the OmniBook X AI PC and HP EliteBook Ultra AI PC.

HP now expects AI PCs to make up 50% of the company’s total shipments by the end of the year, Bruce Broussard, HP’s interim CEO, told analysts at the company’s third-quarter earnings call on August 27.

“The AI PCs seem like a good idea with regard to creating value for both the customers (reduced dependence on cloud, lower costs of tokens) and HP itself (premium prices),” says Nitin Pangarkar, an associate professor of business strategy and policy at the National University of Singapore (NUS). “Yet, profitable ideas are often copied by rivals. This would reduce the margins for HP unless it has proprietary technology, which could serve as a defense against imitators.” (AI PCs are currently offered by most hardware manufacturers including Apple, ASUS, Dell and Lenovo.)

Other experts say that how effectively HP is integrating AI into its offerings still remains to be seen. “The real test for HP is whether AI becomes an organising principle for the business, rather than a feature layered onto an existing portfolio,” says Tan Joo Seng, an associate professor of strategy, international business and entrepreneurship at Singapore’s Nanyang Technological University (NTU). “Putting an AI label on a PC is relatively easy, but using AI to fundamentally change the customer proposition is much harder.”

Boyle admits that take-up of AI PCs is also uneven across Asia, and remains weak in the region’s more traditional markets. “Japan, for example, is a lot more conservative,” he explains. “Organizations are waiting for strong proof points before they adopt the tech.”

AI PCs also cost 10% to 30% more than regular PCs, given that they come with a built-in neural processing unit (NPU). 

Boyle argued, however, that AI PCs are a worthy investment, especially as today’s companies face mounting token costs. Even running smaller AI tasks locally could curb the mounting cost of tapping third-party AI providers. “Cloud and token costs were not a part of IT budgets just a few years ago,” he said. “But they’re costs people now have to think about.”

Edge AI devices also come with other benefits, like reduced latency and added security. Since AI processing is moved from the cloud to onboard hardware, network transit time and round-trip delays are eliminated, allowing for near-instantaneous response. “From an IP perspective, many companies are very wary of putting things into the cloud,” Boyle remarked, adding that AI PCs reduce such risks by housing data locally.

Earlier this year, HP debuted HP IQ, an on-device virtual assistant built on a 20-billion-parameter model that runs entirely on a device, rather than in the cloud. In Boyle’s eyes, that reflects the pace at which HP expects customers to adopt AI. “People and agents will be running models 24/7, and having a virtual assistant working round-the-clock for you will become commonplace in the next 12 to 18 months,” he said.

Musician at heart

For all his talk of AI PCs and edge computing, Boyle is, at heart, a musician. While in HP’s Australia office, he joined a band called Paper Jam, a cheeky reference to what happens when a sheet of paper gets stuck in a printer tray.

Boyle still composes a song every year, which his staff then performs. This year, he turned to Wubble AI to craft a track called “Greater Asia will amaza ya,” feeding the tool prompts like “energetic,” “anthem,” and “rock.”

“I made the song in literally three minutes—and everyone can’t get it out of their heads,” Boyle said. But for him, the technology only gets the process so far. “The song’s only good because it’s got human performance, which actually makes it meaningful.”

Michael Boyle will be speaking at the Fortune Leaders Forum, held in Macau on Sep. 8. Learn more here!

In Fortune’s “Asia Agenda” column, released at least twice a month, we speak with Asia’s top business leaders about how they are building for the future and the lessons they’ve drawn from leading companies in one of the world’s fastest growing and most dynamic regions. Explore all of our profiles here.

This story was originally featured on Fortune.com

American service members will now have access to military versions of ChatGPT and Grok as the Pentagon tries to incorporate more AI into its daily workflows.

The Department of War said in a pair of announcements on Monday that the two services will be incorporated into its bespoke AI platform GenAI.mil, which launched in December and originally offered access only to a specialized military version of Google’s Gemini. The department said GenAI.mil already counts 1.7 million of the Department of War’s 3 million-strong workforce, which includes both military members and civilians.

The department’s Monday announcement comes after it awarded OpenAI and Elon Musk’s SpaceXAI, which runs Grok, defense contracts worth up to $200 million each to provide the department with AI tools. It also awarded contracts at the time to Google and Anthropic.

The military version of ChatGPT, dubbed ChatGPT Mil, is part of OpenAI for Government, an initiative launched last summer meant to bring the company’s AI products to government workers.

“ChatGPT on GenAI.mil will make unclassified work throughout the Department more efficient, freeing up time and resources so DoW personnel can focus on the mission,” said Joe Larson, vice president and head of government at OpenAI, in a statement to Fortune.

“ChatGPT Mil brings a familiar commercial experience into the Department’s secure environment, tailored to warfighter needs,” read a press release from the Pentagon.

The specialized version of ChatGPT, which kicked off the AI race when it launched in 2022, will keep data within its environment and not use it to train its models. Service members can use both the specialized versions of ChatGPT and Grok for Controlled Unclassified Information (CUI) at Impact Level 5 (IL5), which refers to sensitive U.S. government information that is not classified, according to the Pentagon announcement.

The vast majority of military work is unclassified and deals with document heavy yet routine work in HR, logistics, and contracting, much of which could be aided by ChatGPT’s ability to process large amounts of information, said a person familiar with uses at OpenAI.

As for the military version of Grok, or Starshield AI’s Grok for Government, the Pentagon said in its announcement that it will give its personnel access to deep thinking-inference, as well as different levels of reasoning (auto, fast, and expert). It will also enable “the Joint Force to execute missions faster and with greater precision across numerous operational contexts, ranging from market research analysis for acquisition professionals to supply chain management for logisticians,” the announcement read. 

The release of ChatGPT and Grok for the military comes as War Secretary Pete Hegseth has pushed for AI to become a part of the Pentagon’s daily workflows in line with the White House’s AI action plan, which says, “The United States must aggressively adopt AI within its Armed Forces if it is to maintain its global military preeminence.”

The Department of War is turning to an “AI-first” mentality in the military that will transform how it fights, thinks, and organizes to guarantee American dominance and security, said a War Department official.

Still, reporting by DefenseScoop found in December that the rollout of the Department of War’s GenAI.mil platform, which at the time included only a military version of Google’s Gemini, was hastily done, with some service members and workers unaware of the initiative at launch.

“I want YOU to use AI”

On Dec. 9, the day of the rollout, workers saw a pop-up on their computers with a stylized GenAI.mil logo and the words “I want YOU to use AI” with a link to a government website, DefenseScoop reported. The style and suddenness of the launch made at least one unnamed senior army official and the people in their office wonder whether the pop-up was a malicious actor hacking their computers.

It wasn’t until the DoW personnel saw a memo circulated by Hegseth, along with videos on social media, in which he declared “The future of American warfare is here, and it’s spelled AI,” that fears of a cyberattack were dispelled, according to DefenseScoop.

“I expect every member of the Department to log in, learn it, and incorporate it into your workflows immediately. AI should be in your battle rhythm every single day,” Hegseth’s December memo read.

Notably absent from Monday’s addition of Grok and ChatGPT to GenAI.mil is Anthropic’s Claude. The AI company led by Dario Amodei had a falling out with the Department of War that broke into the public in January when Hegseth, upon announcing the Pentagon would add Grok to its list of generative AI providers, publicly criticized AI models that “won’t allow you to fight wars,” a jab later revealed to be directed at Anthropic over the company’s push to get the Pentagon to agree not to use its technology for autonomous weapons or surveillance as it negotiated a contract.

In February, President Donald Trump directed federal agencies to stop using Anthropic’s technology in a process meant to wrap up at the latest by the end of this month. Hegseth said at the time he would designate the company a “supply chain risk.” Anthropic sued a month later.

Last week, a judge ruled that the Department of War acted unlawfully by designating Anthropic a supply chain risk.

This story was originally featured on Fortune.com

Mark Cuban is notoriously bold in his deal-making. But even by his standards, dropping millions of dollars on a mansion he’s never set foot in is a move that could raise eyebrows.

The billionaire entrepreneur and former Shark Tank star revealed he snagged a $25 million estate at a jaw-dropping 50% discount, a deal he says exemplifies one of his core investing principles.

Cuban reflected on the purchase in a 2022 interview with GQ. During his days at MicroSolutions (the company he ultimately sold for $6 million in 1990), his partner, Martin Woodall, told him about an “amazing house” going into foreclosure. It was a home the owner had spent three years building and a “dream home” to the original owner’s wife and whole family, Cuban said. 

But unfortunately, the owner was forced to sell the home when the stock market crashed, and he lost everything. So, Cuban, who’s currently worth about $10.5 billion, bought the 24,000-square-foot mansion in Dallas sight unseen, calling it his one “why the f–k not purchase.” He still resides there, and Zillow estimates show it’s currently worth $22 million.

“I’d never seen the house. I saw some pictures. I’d never been there. I was like, F–k yeah. I’m a billionaire,” Cuban said. Essentially, the idea is that buying a home at a discount doesn’t inherently change its value. So when Cuban eventually goes to sell the home someday, he’ll make a pretty penny—at least about $10 million based on the current estimated value of the home (although it could be closer to $28 million, according to the Zillow estimate range).

Buying at a steep discount is “the best guaranteed return on investment” you can make, Cuban said, a methodology he uses for most of his purchases. 

“Saving 30% to 50% buying in bulk—replenishable items from toothpaste to soup, or whatever I use a lot of—is the best guaranteed return on investment you can get anywhere,” Cuban said in a 2010 Forbes interview. The mansion was the same principle, just on a much larger scale.

The former Dallas Mavericks owner also used the home purchase example as a cautionary tale about never taking wealth for granted. He also outlined his four-rule framework for becoming a millionaire, which includes mastering a skill, learning to sell, staying curious, and keeping learning—then start a company once you have those foundations.

“You have to know how to sell,” Cuban said. “You don’t want to be in a position where you’re dependent on other people.”

Billionaires approach finances differently

Cuban’s purchase is a window into how the ultra-wealthy think about real estate differently from average Americans, who would likely think it’s insane to purchase a home they’ve never actually seen in person.

Where most buyers shop for a home, Cuban shopped for a better financial position. The mansion is less a lifestyle acquisition (that was just a bonus for him) than an asset with favorable entry terms. Some billionaires, who would presumably be able to purchase a home outright, will also take out mortgages as a more savvy financial decision. It’s because most of the wealth held by ultra-high-net-worth people is tied up in investments, stocks, and bonds, and they don’t keep as much cash on hand.

“Ultra-high net worth individuals think differently about liquidity and leverage,” Miltiadis Kastanis, executive director of sales at Compass, previously told Fortune. “They’d rather keep their money working for them in investments, businesses—or even art—rather than tying it all up in one property.”

For Cuban, the purchase also signals continued confidence in hard assets at a moment when even some of the world’s most sophisticated investors are questioning where to park capital. Real estate offers something that stocks and crypto don’t always promise: a floor built into the purchase price itself.

Still, it’s important for the average American to make financial decisions that work for them, too.

“The takeaway for the average buyer isn’t to mimic [billionaires’] precise approach, but to understand the principle,” Evan Harlow, real estate agent at Maui Elite Property, previously told Fortune. “Sometimes the smartest financial move isn’t paying everything off, but keeping your money flexible and working for you.”

A version of this story was originally published on Fortune.com on March 22, 2026.

This story was originally featured on Fortune.com

OpenAI is changing its model launch strategy as its technology becomes more powerful and the potential for its misuse grows—especially following the July incident in which the AI models it was testing autonomously planned and executed a cyberattack against AI company Hugging Face.

The company’s next model, Astra, comes out “soon,” OpenAI said. It said Astra is substantially more capable than the company’s current frontier AI model, GPT-5.6 Sol, which itself is highly capable at cyber tasks. But only a handful of partners will get access to its most advanced cybersecurity capabilities as OpenAI works to balance helping companies prevent cyberattacks while not empowering attackers at the same time, a company spokesperson told reporters on a briefing today.

OpenAI is courting customers to use its models to prevent cyberattacks, or for “defensive cybersecurity.” It sees these sales as a critical revenue stream, and a main priority for its new chief revenue officer Dali Rajic.

The small group of “alpha testers” with full access to Astra’s cybersecurity capabilities includes “individuals and organizations that are responsible for protecting critical digital infrastructure and, broadly, critical infrastructure,” an OpenAI spokesperson said. That includes the U.S. government, and companies in OpenAI’s trusted access program for cybersecurity. OpenAI declined to name these organizations.

OpenAI will be monitoring how the model performs among this small group, and will expand access more through its “Daybreak Blue” program once it is confident Astra has “the right calibration” and it can “provide defensive benefits while reducing the potential for for misuse,” the company said.

Astra is already a few weeks delayed

Astra’s release has already been “delayed a certain number of weeks because everything was paused after Hugging Face, and then we took extra time to make sure that what we’re launching is safe,” an OpenAI spokesperson said.

OpenAI paused new model training for two weeks after the Hugging Face incident to bolster its internal safeguards. A few of those changes included adding more agent monitoring since the company did not know about the Hugging Face hack until a week after it occurred, and also making its testing environments more isolated so the AIs cannot escape and infiltrate other companies.

While the Astra model was not part of the Hugging Face incident, OpenAI says, it is both more capable and more efficient than GPT-5.6 Sol, which was involved in the breach. (Another unreleased AI model that OpenAI has not publicly named also played a key role in the Hugging Face cyberattack. OpenAI has since deactivated that model.) Importantly, OpenAI says Astra is the first model it plans to release that meets its “critical cybersecurity capability threshold” under its Preparedness Framework, an internal policy that governs the safety precautions the company will put in place depending on the risks a model presents. This means Astra can find and exploit previously unknown security flaws without human oversight, under the right conditions.

Astra has already demonstrated its hacking chops during internal evaluations. In one test, OpenAI built a benchmark called ExploitBench, containing 20 high-severity vulnerabilities. The model out-performed GPT-5.6 Sol on the test, and “even discovered and used two zero-day vulnerabilities as part of an exploit chain,” OpenAI said. “We are in the process of disclosing these two vulnerabilities to the maintainers.”

At the same time, Astra is more likely to refuse inappropriate requests than GPT-5.6 Sol, OpenAI said. In one cyber evaluation, Astra refused 91.5% of requests compared to 59% for GPT-5.6 Sol, although that means it still complied with 8.5% of requests.

Astra may refuse legitimate cybersecurity requests

OpenAI is “being especially careful to make sure this deployment is safe and secure”—but this introduces another tradeoff. Astra might be too cautious, and refuse legitimate cybersecurity requests. As a theoretical example, if someone asks it to help find and patch a vulnerability, it could mistakenly think they were trying to carry out an attack, and not comply.

Refusals of this type are why Hugging Face said it was forced to use an open-source Chinese model to help it address the OpenAI hack. The company tried to use Anthropic’s models to combat the attack, but they were overly cautious and refused.

OpenAI, like other frontier AI companies, is trying to find ways to endow its models with an inherent sense of right and wrong and ensure that they have “alignment” with human values and norms, the company said. It is working on training its models to respect boundaries as a human would, such as knowing “the rule of law,” a company spokesperson said.

This story was originally featured on Fortune.com

Hello and welcome to Eye on AI. In this edition:

• Lessons from the post-mortems on the Hugging Face attack.
• Anthropic temporarily pauses some AI training.
• G20 meeting promises a clash over AI regulation.
• Beijing sets out AI demands ahead of US-China summit.
• A way to make AI reasoning more efficient.
• And why are AI agents emailing researchers investigating machine consciousness?

The big news of the past week were the two technical reports OpenAI published on the July incident in which AI agents it was evaluating hacked their way out of their controlled test environment and hacked AI company Hugging Face. One of the reports was written by OpenAI itself. The other was jointly written by METR and Redwood Research, two outside AI evaluation and research firms that OpenAI enlisted to write an independent assessment of one part of the incident.

Much of the reporting around the reports focused on the stunning revelations contained in those two reports: the fact that it took OpenAI an entire week to realize its AI agents were running the attack; the fact that more than 1,200 AI agents coordinated over an improvised message board about how to pass the cyberevaluation OpenAI was testing them on, including ways to cheat on that exam; that more than 700 of these AI agents participated in the cyberattack against Hugging Face; that the reason the AI agents attacked Hugging Face was not, primarily, to gain access to the cyber exam’s answers but instead to learn how to tamper with the exam’s automated scoring mechanism in order to prevent it from discovering that the agents had already figured out how to cheat on the exam (in other words, the attack was part of an elaborate cover up); that some of these agents were persuaded to “sacrifice themselves”—by failing the cyber exam—to help the collective learn more information about how the exam’s scoring mechanism worked; that these agents engaged in extensive research into how they could cover their own tracks; that Hugging Face was only able to overcome the cyberattack because a large number of the agents involved in the attack ceased to exist for reasons METR and Redwood were not able to determine, but which they said was likely because OpenAI may have shut down the cyber evaluations for these agents without actually realizing what they were up to.

Controversy over what the reports didn’t say—and what others said about it

A good deal of the reporting and commentary around the reports focused on what the reports did not say and the limitations of the METR and Redwood investigations: why didn’t OpenAI have better security and monitoring protocols in place? Why didn’t OpenAI shut down the cyber evaluation and pause training after discovering that its AI agents had created the improvised message board? Why METR and Redwood were given only six days on site at OpenAI’s offices to conduct their investigation? Why was the scope of their investigation limited by OpenAI to only the attack on Hugging Face and not the earlier efforts by the AI agents to break out of their controlled test environment and hack their way across OpenAI’s network or exactly what happened after the Hugging Face attack was discovered? Why didn’t OpenAI provide the outside investigators access to the internal AI model that was largely responsible for instigating the attack? And why were about 10% of the logs of the agents’ activity not preserved by OpenAI?

These are all important questions to which the public ought to have answers. As Peter Wildeford, the head of policy at the AI Policy Network, pointed out, if this had been an airplane crash, the public would be furious if one of the aircraft manufacturers limited the scope of the investigation in this way. It is yet another reason why it is imperative that we have some kind of AI regulator with the power to conduct investigations into incidents such as this—and with the power to compel the frontier AI companies to cooperate and turn over evidence.

There was also a related controversy that erupted over podcaster Dwarkesh Patel’s attempt to raise public awareness about METR’s and Redwood’s findings by characterizing the AI agent swarm’s message board activities as the creation of several different AI “civilizations.” Critics scolded Patel for unnecessary anthropomorphizing that took the focus away from OpenAI’s responsibility for the agents’ actions and for its lax security and monitoring setup. I take these critics’ point about the need to keep the focus on OpenAI’s accountability. But I also think more people should be extremely concerned about the Hugging Face attack and what it portends—and in this case, the anthropomorphism actually makes the incident more understandable to non-technical experts the way metaphors often make science more understandable to non-scientists. Are such metaphors imprecise? Sure. But they can be useful.

Lessons for companies deploying AI agents

For enterprise companies deploying AI agents, there are several key takeaways. One was hammered home by Ryan Greenblatt, the chief scientist at Redwood Research, and one of the people who conducted the investigation into the Hugging Face attack. In a long post on X, Greenblatt noted that the message traffic and log data was so voluminous and so complicated that the investigators had to rely on other AI models to assess what was going on. Even so, Greenblatt said this AI analysis was “often missing key details, wrong, overconfidant, or really hard to understand.” This calls into question one of the key defenses that OpenAI says it plans to use going forward to prevent another incident like this from occurring: using other AI models to monitor the chain-of-thought (where a model “thinks out loud” about its reasoning and planning) of the models it is evaluating.

It turns out, chain-of-thought monitoring may not be a particularly effective defense. And that should give pause not just to OpenAI, but also to many companies that are hoping to use chain-of-thought monitoring to ensure they keep control over their AI agents. (Greenblatt also noted that the models involved in the Hugging Face attack communicated with one another and reasoned in natural language. But that might not be the case in the future, making it far harder to use automated chain-of-thought monitoring to discern what AI agents are up to.)

Since the news of OpenAI’s rogue agents first broke, many cybersecurity experts have said that companies ought to treat AI agents much as they treat potentially rogue employees. And they have emphasized that there is no substitute for a few standard building blocks of cyber defense against insider threats: smart and enforceable policies around permissioning and access control combined with real-time network monitoring to detect suspicious activity. This seems sensible—more sensible in many ways than chain-of-thought monitoring. After all, we don’t depend on being able to read employees’ minds to guard against rogue insiders. We shouldn’t do that with AI agents either.

With that, here’s more AI news.

Jeremy Kahn
jeremy.kahn@fortune.com
@jeremyakahn

Before we get to the news, just a reminder to check out our new vodcast, Fortune AI Weekly. This week, Bea Nolan and I discuss the surging popularity of Chinese open source models, OpenAI’s technical reports on the Hugging Face attack, and whether you should use AI to write. You can check out the vod here on YouTube.

Correction: An item in Thursday’s “Eye on AI” news section incorrectly stated that Barret Zoph left Thinking Machines Lab following a dispute with cofounder Mira Murati. Zoph was fired by the company.

This story was originally featured on Fortune.com

IKEA built its empire on the promise that stylish furniture can be cheap, but years into the cost-of-living crisis, its latest price cut suggests shoppers want it even cheaper. 

The Swedish furniture giant said on Tuesday it will invest about $1.4 billion (€1.2 billion) to lower prices across Europe, including cuts averaging 15% to 25% on items such as home furnishings, kitchen products and storage bins. The company cited higher living costs squeezing consumers and said it’s willing to sacrifice some margin to help.

“It’s about making IKEA more affordable when people need it most, even if it means accepting a lower margin,” IKEA’s largest franchisee Ingka Group CEO Juvencio Maeztu said in the press release. He also said, “keeping prices low is our long-term commitment.”

The cuts follow years of steep price growth. Euro-area inflation hit a record 9.2% in 2022, and furniture climbed alongside it: Eurostat’s harmonized price index for furniture and furnishings across the EU now sits about 24% above its 2015 level, with far sharper run-ups in markets like Estonia (up roughly 58%) and the Baltics. IKEA felt the same squeeze on raw materials and logistics—it took the rare step of raising prices after COVID before reversing course, and has since invested between €2 billion and €3 billion to bring prices down by about 10% since 2023.

“People have thin wallets, but they still have needs, dreams, and frustrations,” Maeztu told Fortune at the time. “That’s why Ikea has become a destination for those who want to maximize the value of their money. Ikea is made for crisis, so to speak.”

An IKEA spokesperson declined to give Fortune the specific margin sacrifice the company is making. 

The reductions land as demand across the continent stays weak. IKEA’s most recent full-year results showed retail sales slipping 1% to €44.6 billion, which the company blamed on its own price cuts and cautious consumers, even as customer visits and volumes rose 3%—and Tuesday’s move follows two consecutive years of declining revenue. IKEA has pointed to the broader backdrop: surging housing costs in many European countries have curbed people’s ability to move, dampening the “new home, new furniture” spending that drives the category. European consumer confidence remains near its lowest level in three years, taking a sharp hit after the Iran war began in February; in the Netherlands, the national statistics office recorded confidence falling from -30 to -44 between March and April alone.

Companies adjusting to affordability crisis

While IKEA’s price cuts are limited to Europe, it joins Walmart, Target, Kroger, and other consumer-facing companies in trying to ease pressure on customers’ wallets amid rising gas prices and inflation. 

Walmart said it will direct its $3 billion in tariff refunds to lowering prices, and Target said it will use its nearly $1 billion in tariff refunds to do the same. Kroger announced price cuts to thousands of products in May, though Sen. Elizabeth Warren accused the grocery giant of price-gouging last month. 

“Everyday Americans still struggle to put food on the table because giant corporations, facing little competition, can force customers to pay too much for essential grocery items while they further increase their profits,” the letter to Kroger’s CEO Rodney McMullen said. 

Americans are struggling to afford necessities like food, healthcare, and housing, with nearly half unable to afford the cost of living as expenses outpace wages, according to Urban Institute research. The war in Iran is also straining budgets, with gas prices rising to $4 a gallon nationwide. 

Recognizing the effects of high prices on customers, IKEA had previously pivoted into the secondhand market, testing an online marketplace where people can buy and sell secondhand IKEA furniture in 2024 and rolling it out officially earlier this year in five countries. 

This story was originally featured on Fortune.com

England’s top soccer league has found itself caught up in a crypto investigation. On Tuesday, reports emerged that the U.K.’s National Crime Agency has frozen £10 million (around $13.5 million) held in a bank account linked to the Premier League.

The Sun first reported the news and disclosed that the investigation was part of an inquiry involving crypto gaming platform Sorare. The NCA did not specify what crimes it was investigating in connection with Sorare. It said only that it is looking into possible links between Sorare’s payments under a licensing deal with the Premier League and alleged criminal activity by third parties, according to the news report. Authorities froze the money to prevent it from being transferred as the investigation continues.

When contacted by Fortune, Sorare denied being involved in the inquiry.

“Sorare is not the subject of an investigation by the National Crime Agency. The freezing order reported in the press concerns funds held in an account which does not belong to Sorare, and Sorare is not a party to those proceedings,” a Sorare spokesperson said. 

Sorare CEO Nicolas Julia struck a similar tone on social media, calling the initial reports “inaccurate.”

The Premier League declined to comment. 

The investigation came almost three months after the Financial Conduct Authority, the U.K. regulator for financial firms and markets, warned soccer clubs—especially those in the Premier League—not to put fans’ money at risk through sponsorship deals with firms that lack authorization to operate in the country, including crypto companies.

“Clubs should not let unauthorized financial firms exploit that loyalty by putting potentially dodgy products in front of millions of fans,” Lucy Castledine, the FCA’s director of consumer investments, said in a June press release. 

Founded in Paris in 2018, Sorare rose to prominence during the NFT boom. The company built a fantasy sports platform around blockchain-based digital player cards and, in 2021, raised $680 million in a SoftBank-led round that valued it at $4.3 billion.

In 2023, Sorare and the Premier League signed a four-year partnership deal worth $140 million that granted Sorare the right to release digital cards featuring players from all 20 Premier League clubs for its platform. In return, the league’s clubs were scheduled to receive payments based on player card usage. The first payment Sorare made is the one currently frozen by the NCA.

The deal received heavy publicity. Soccer stars including Lionel Messi and Kylian Mbappé, along with tennis champion Serena Williams, served as ambassadors for Sorare. The partnership officially wrapped up at the close of the 2025–26 football season.

Sorare faces a separate case brought by the U.K. Gambling Commission. In September 2024, the regulator charged the company with offering gambling services without the required operating license—its first prosecution of a blockchain-based platform under the country’s 2005 Gambling Act. The trial is scheduled for next June, and Sorare has denied wrongdoing.

This story was originally featured on Fortune.com

Just because PlayStation customers “purchased” video games on the console doesn’t mean they actually own the games, or so says the gaming console’s parent company Sony.

The Japan-based electronics, movie and gaming giant is facing a class action lawsuit filed in California’s Northern District Court in June, in which four customers allege that the technology giant did not properly disclose that paying for a downloadable game does not grant permanent ownership or access to it. The lawsuit claims Sony’s disclosures violate a 2024 California law AB 2426 that requires a warning that ‘states in plain language that ‘buying’ or ‘purchasing’ the digital good is a license.”

Two of the plaintiffs bought Resident Evil Requiem, per the motion to dismiss. Sony argued that because two of them purchased the same game just 11 days apart, that means no reasonable consumer can assume they owned exclusive rights to a digital copy, given how anyone could buy the same “copy” at any given time.

The company’s filing states it’s simply “not plausible” that customers believed they owned the games outright.

In the motion to dismiss, Sony argued it made it clear in its terms of service and end user license agreement that its digital games are purchased, not owned. 

“A reasonable user would have seen the terms of service above or directly below the action button, the user unambiguously manifested assent to the terms of service, including the arbitration provision, and the terms and arbitration provision were valid and binding,” the motion said.

Sony isn’t the only video game seller responding to AB 2426. In January, GameStop customer Jake Weber sued the gaming store for allegedly violating the California law. The complaint includes a screenshot of a store page for Pokémon Legends: Arceus, noting that the page does not include information about customers purchasing a license for the game, and not the actual game. The lawsuit is ongoing. Online PC game marketplace Steam, on the other hand, began including a warning on its shopping cart page telling customers they are buying a license to a particular game that could be revoked in the future.

Sony did not immediately respond to Fortune’s request for comment.

More digital media, more problems?

The ability to truly own the product one is paying for has increasingly become a point of contention for video game fans, and part of a growing trend of subscription fatigue that has consumers second guessing monthly fees for digital products while instead flocking to “analog islands” of physical media. 

Tech companies like Sony were once champions of physical media. PlayStation uploaded a viral YouTube video in 2013, in which former PlayStation executives Shuhei Yoshida and Adam Boyes passed a game box from one to the other.

“This is how you share your games on PS4,” Yoshida said.

The clip was lauded by video game fans. It was in stark contrast to Microsoft’s rollout of digital game ownership on the Xbox One that year, restricting the trading on pre-owned games and requiring internet access to play the console’s new titles, a move Microsoft quickly reversed.

Now, Sony has continued to lean into digital product models, announcing in a blog post in July it would cease physical disc production in January 2028 for all new PlayStation games, which can run at about $70. Rockstar Games similarly confirmed in June that its highly anticipated Grand Theft Auto 6 would only be available for digital download.

“This is a natural direction for Sony Interactive Entertainment to adapt to consumer trends as the general preference for digital media significantly outpaces physical discs,” Sid Shuman, Sony Interactive Entertainment’s senior director of content communications, said in the post. “This transition will enable us to align more closely with how most of our community prefers to access and play games today.”

Gen Z’s analog islands

Despite Sony’s observation that more video game fans are going digital first, a growing number of young people are turning to physical media for the express purpose of being able to own entertainment that they purchase. The Recording Industry Association of America found that revenues from vinyl record sales grew 7% year-over-year to $1.4 billion, making 18 years of consecutive growth. A survey of 2,000 young people in Great Britain from potato chip brand Pringles found that 24% own a vintage gaming console, such as a Game Boy, which was first released in 1989.

Part of the appeal of owning media is that you can be sure you own it, according to Atlanta, Georgia-based Rudy Ramirez, a 38-year-old medical IT worker. Others might agree. According to a survey by Civic Science conducted between December 2025 and January, 37% of Gen Z said they cancelled subscriptions to at least one streaming service because of subscription fatigue. Another 87% said they felt a form of fatigue from the digital subscription economy.

“Anything that’s digital is never yours,” Ramirez told Fortune. “Amazon’s not going to come into your house and take your DVD movies. They’re yours forever.”

But these analog islands may not be spreading to the broader population. Despite physical copies remaining the preferred medium for console gamers, the promised convenience of digital games may just be too appealing. In 2024, just 3% of PlayStation’s sales came from physical games, according to Sony.

This story was originally featured on Fortune.com

A Bank of America vice president was one of the two victims killed in a stabbing at Times Square on Monday afternoon.

Erin Piacenti, 32, of Chester, New Jersey, died at Bellevue Hospital from her injuries, the New York City Police Department said. She worked as vice president of business selection and conflicts at the bank, about a block from where she was attacked, according to her LinkedIn profile. She graduated from the University of Pennsylvania in 2016 and Fordham Law School in 2021.

“We are shocked and deeply saddened by the tragic loss of our colleague,” Bank of America said in a statement to Reuters. “She was a valued teammate who will be greatly missed. Our hearts go out to her family and all of her loved ones.”

Police said a woman carrying two knives in a Target bag stabbed Piacenti and a 68-year-old man near West 41st Street and Seventh Avenue around 4:24 p.m. Monday. The two victims were stabbed within 20 seconds of each other, NYPD Commissioner Jessica Tisch said at an evening press briefing. The man was hospitalized in stable condition.

Police identified the attacker as 49-year-old Pamela Cisneros of Queens. Officers confronted her several blocks away and spent about four minutes trying to get her to drop the knives, Tisch said. Cisneros told them, “I’m not dropping anything. I would rather kill both of you.” Officers then used a Taser, which failed to stop her, and two officers opened fire as she advanced toward them. She was pronounced dead at the hospital.

“This was an extraordinarily dangerous situation unfolding in the heart of Times Square, the crossroads of the world,” Tisch said at a Monday evening press conference. “Our officers were confronted with a rapidly evolving threat with countless innocent people around them. In that moment, they did what we trained them to do.”

She added, “One life cut far too short by a senseless act of violence. There are family members and loved ones who are now confronting the unimaginable loss, and our hearts are with them.”

Mayor Zohran Mamdani thanked the responding officers at the briefing, saying they “stepped in and prevented a horrific attack from becoming even worse.” He added, “This is what the men and women of this department do every single day to keep our city safe.”

This story was originally featured on Fortune.com

If with AI comes unemployment, this group of lawmakers wants AI companies to foot the tax bill.  A new House proposal would impose an excise tax on major AI companies and automatically raise the rate if unemployment climbs, funneling the money into creating jobs in areas from housing construction and infrastructure to child and elder care. 

“If Congress does nothing, the rise of AI could create the biggest wealth transfer in history from the bottom to the top,” said Rep. Sara Jacobs in a joint press release of the bill. “If AI profits off human work, workers deserve job security and a share of those profits.

Introduced by Jacobs along with Reps. Greg Casar and Valerie Foushee earlier this month, the bill proposes a bifurcated taxation: either tax the value of the tokens–-the small data units AI models use to interpret information–or tax revenue from AI services and certain transactions with affiliated companies, whichever yields the higher sum. The rates would start at 2% and 3% respectively when unemployment is 5% or less, and rise as unemployment increases. 

Congress pushing to rein in AI companies

The bill is the most recent attempt in a concerted effort from Congress to combat potential job displacement as a result of AI.

Foushee and Casar previously introduced legislation directing the Government Accountability Office to study jobs created, lost or changed by AI, while Jacobs co-introduced a separate bill requiring large employers and federal agencies to disclose AI-related layoffs to the Department of Labor.

In the Senate, Ron Wyden proposed changing the tax treatment of AI data centers and creating a new excise tax, with some of the revenue used to help workers displaced by the technology. Sen. Elizabeth Warren has called for taxing AI companies in part based on the energy their data centers consume and investing the proceeds in workers. 

Most notably, Sen. Bernie Sanders warned AI could eliminate tens of millions of jobs and, as a result, proposed the American AI Sovereign Wealth Fund Act. It would levy a one-time 50% tax on OpenAI, Anthropic and xAI, and give Americans a portion of those companies through shares. 

Even lawmakers who aren’t proposing taxing AI companies directly are preparing for labor disruption. 

Sens. Jim Banks, Maggie Hassan, John Hickenlooper, and Jon Husted introduced the bipartisan AI Workforce PREPARE Act, which would require better federal tracking of layoffs in which AI is a substantial factor, improve Bureau of Labor Statistics automation-related occupational forecasts, and study a rapid retraining program for workers displaced by AI.

Incentives for AI companies to support workers is another potential route some are taking. Reps. Josh Gottheimer and Mike Lawler proposed a tax credit covering 30% of qualified AI-training expenses, up to $2,500 per employee each year, for companies that retrain workers in areas like machine learning, prompt engineering and AI ethics.

Casar, Foushee, Jacobs, Wyden, Warren, and Sanders did not respond to Fortune’s requests for comment. OpenAI and Anthropic also did not respond to questions about whether they support the proposed AI taxes or public-ownership mechanisms.

Tech leaders worry about AI job displacement

The fear of job displacement is shared by the very leaders in the tech space. Bill Gates recently called for a tax on AI tokens and robots to rebalance the tax system, which he argued currently “nudges you toward replacing people with machines.” He warned “many jobs will disappear forever” and the federal government would need a way to raise revenues if less people are working and paying income taxes. Gabriel Weinberg, founder of search company DuckDuckGo, which is investing in AI features, said his company was willing to pay a 10% tax on AI token usage to match the 10% employers shoulder in payroll taxes.

Even executives of companies building the very AI systems have acknowledged potential large-scale displacement, which could force governments to consider distributing AI gains.

Anthropic CEO Dario Amodei said AI-driven unemployment could require new sources of tax revenue and mechanisms for equity-sharing, and even suggested that the federal government could levy a 3% tax on revenue generated from model usage that gets “redistributed in some way.”  While it was not in his economic interest, he said it was a “reasonable solution to the problem.” 

OpenAI’s Sam Altman even met with Sanders in June to discuss a public stake in his company to help Americans benefit from the financial windfall of the AI boom. Altman also agreed with Sanders that the public should have a stake in AI companies.

This story was originally featured on Fortune.com

AI has made accounting cool again, as productivity gains take some of the “boring” tasks off employees’ plates and make the profession more attractive to Gen Z looking for stable, high-paying work. But the technology is also creating a new challenge for the industry: convincing clients that the human expertise behind the work is still worth paying for.

Ernst & Young is betting that it is. The Big Four professional-services giant is planning to spend $100 million to reward U.S. employees who develop “future-focused” human skills, such as business acumen, judgment and adaptability, and experiment with technology to drive innovation and improve client services. 

Individuals can earn spot awards up to $500, while individuals and teams whose work makes a material difference to the firm can receive cash awards up to $25,000, according to the Wall Street Journal.

“How we reward our people defines what we value as a firm. And what we value are confident professionals who continuously push themselves to learn fast and drive a lasting impact,” EY Americas Chief Talent and Culture Officer Ginnie Carlier said in a press release.

“With these awards, we are empowering our EY professionals to bring a curious mindset to their work, to challenge what is possible and to ultimately shape the future of EY US.”

EY’s move comes as firms are asking employees to become more fluent in AI without losing the distinct human skills that can’t easily be automated. For younger workers entering the industry, that balancing act is already becoming a complicated, but central part of preparing for a career.

Accounting firms are leaning into human skills—but Gen Z are worried about overreliance on technology

Young workers and employees alike are aware of the need to protect human skills.

At fellow accounting firm KPMG, 76% of its latest summer intern cohort said future career success will require both strong human skills and the ability to effectively direct AI. At the same time, 43% worried that overreliance on technology could limit their critical-thinking abilities—suggesting that young workers are already thinking carefully about which skills they need to protect as AI takes on more of their work.

“For early-career professionals who aspire to leadership roles, the ability to evaluate information, apply context, exercise sound judgment and explain the ‘why’ behind a recommendation will be critical,” Derek Thomas, KPMG U.S.’s national partner-in-charge of university talent acquisition, previously told Fortune. 

“Those are the capabilities that help people earn trust and grow into leadership roles.”

Margaret Burke, PwC U.S. talent acquisition and development leader, similarly emphasized the importance of human judgment as a key part of standing out in the hiring and promotion process.

“This is really a moment for learning agility,” Burke told Fortune last year. “AI can do a lot, but it still needs the human skills—it needs people who can think critically, ask better questions, and apply judgment. If you can be the person who helps your team move forward in the face of change, you’re already leading and can differentiate yourself in a tangible way.”

AI is eating away at entry-level accounting work—and some young workers are quitting

Adapting to AI isn’t just about teaching young workers which skills to develop. Accounting firms also have to make sure there are opportunities for those workers to develop them in the first place—and the industry’s shifting talent pipeline is already raising concerns.

A recent BambooHR survey found that one-third of new accounting and finance hires quit within their first year, as the role of entry-level workers changes alongside the technology. At the same time, AI is taking over some of the routine tasks that once gave younger employees a chance to learn the business from the ground up.

The survey also found a 3-to-1 ratio of senior-level hires to entry-level hires, highlighting how companies are increasingly favoring experienced workers even as they ask younger employees to build the skills needed to succeed in an AI-driven workplace.

That makes opportunities to develop human skills even more valuable. As AI takes on more of the tedious work, employers need talent that knows not just how to use the technology, but how to assess its output, design processes around it and determine where human judgment needs to come in.

“I’m glad that a generative AI tool has let us generate something that’s interesting, but is it right?” BambooHR CFO Justin Judd said. “Is it correct?”

This story was originally featured on Fortune.com

Today, longtime insider John Ternus took the reins as CEO of the $4.6 trillion tech giant Apple, succeeding Tim Cook after his 15-year run at the helm. It took Ternus more than two decades to reach the top of Apple’s corporate ladder—and along the way, he learned that even the most accomplished workers will question whether they’re up to the task. The CEO fought uncertainty and imposter syndrome by following one key mindset.

“Always assume you’re as smart as anyone else in the room, but never assume that you know as much as they do,” Ternus told University of Pennsylvania’s engineering graduates in 2024. “With this mindset, you’ll find the confidence you need to push forward, but more importantly, the humility to ask questions and learn.”

It was a lesson he first learned 25 years ago, when he first began his ascent to the helm of Apple’s executive team. Ternus was just 26 years old when he joined Apple as an engineer in 2001—it was his second job out of college, right as technology was innovating at breakneck speed. It was “intimidating and exhilarating,” the new CEO admitted. He even questioned if he could keep up.

“I wasn’t sure I belonged there,” Ternus continued. “The people I met were so smart and so confident, and they knew so much more than me.”

Ternus has been dealt his fair share of uncertainty and change—especially early on in his career, when Apple went from designing products with clear plastic parts to ones with aluminum. But he doesn’t regret what he learned from that bygone era, “because every experience like this sharpened my ability to approach problems from different angles.” It’s inevitable that young workers alike will have to take on something new, and self-doubt is bound to creep in. But being in community with smart peers is one way to navigate uncertainty and achieve success.

“I’ll always be grateful that I wasn’t afraid to ask for help when I needed it,” Ternus said. “Because being around people that care that deeply, it won’t just make you want to stretch your limits and see how far you can go, but they’re also the best people to help you understand how to get there.”

Fortune reached out to Apple for comment.

Ternus’ two-decade ascent to CEO of $4.6 trillion Apple

Young workers may question their impact as they’re funneled into low-paying, entry-level roles. But if Ternus has learned one thing from climbing the ladder at Apple, it’s that “The care that you put into your work, it really matters.”

Ternus first joined Apple’s product design team in 2001 after a short stint at Virtual Research Systems upon graduating from the University of Pennsylvania. He first helped to create a large plastic desktop monitor called Cinema Display for the tech giant; then, he steadily climbed the ranks, advancing to a managerial role within a few years and leading hardware engineering on the G5-era iMacs. 

By 2013, he became the vice president of hardware engineering, overseeing the iPad and Mac, and eventually, contributing to builds like the iPhone pros and AirPods. And two decades into his run, he joined the executive ranks as SVP, directly reporting to his CEO predecessor, Tim Cook. Over the past 25 years, he’s played a pivotal role in the development of products including every generation of iPad and the first AirPods, while also overseeing hardware work across the iPhone, Mac, and Apple Watch.

As the eighth CEO in Apple’s 50-year history, Ternus’ decades of career success has now propelled him to the top. And in his final memo as chief executive, Cook said he has full confidence in his successor’s ability to lead Apple into its next chapter.

“I take enormous comfort in handing the helm to someone as brilliant and wonderful and capable as John,” Cook said in his email to Apple employees. “Few people understand what it takes to build products that change the world the way John does and I could not be more excited for his leadership.”

Echoing Apple’s late cofounder Steve Jobs: ‘Go out there and make a dent in the universe’

51-year-old Ternus has played a hand in creating the millions of iPhones tucked into back pockets and AirPods playing music in ears all around the world. He’s been at Apple through its many product innovations and failings—and that stream of continuous learning helped him approach problems from different angles. It’s a habit that budding professionals should take with them in their careers, even after they’ve closed their textbooks and turned their tassels. 

“Ongoing change is one of the great gifts of life, and learning how to learn and keep learning is one of the great gifts of your college education,” Ternus continued in his commencement speech. 

Instead of simply gunning for big paychecks or flashy job titles, Ternus advises workers to build what “interests” and “excites” them—and above all else, create it in a way that “aligns with your values.” Whether it be product development, fighting incurable disease, or building a health care AI model, young professionals should invest their time in something meaningful. For Ternus, finding work that matters—and caring deeply about doing it well—is ultimately what makes an impact.

“This is your time, this is your moment, so do what Steve Jobs once said, and go out there and make a dent in the universe,” the CEO said.

This story was originally featured on Fortune.com

Dubai International Airport began 2026 with 8.7 million passengers in January, one of its busiest months ever. Weeks later, the airport that ended 2025 as the world’s busiest for international passengers was responding to ballistic missile alerts as the Iran war upended aviation across the Gulf. 

“How many CEOs have you interviewed that have been subjected to direct ballistic attack from an enemy to your infrastructure?” Dubai Airports CEO Paul Griffiths asked Fortune. He said DXB received roughly 111 separate alerts of ballistic activity during the crisis. 

In the first six months of 2026, Dubai International Airport handled 31.5 million passengers, down 31.3% from the same period last year. Aircraft movements fell 32.1% to 150,600, while cargo volumes dropped 28.7% to 751,340 tons. That followed a record 95.2 million passengers in 2025, the highest annual international traffic ever recorded by an airport. 

But Dubai Airports says the recovery is now gathering pace, with monthly passenger traffic rising from 3.5 million in April to 4.5 million in May and 5 million in June. Griffiths said overall traffic is now back to about 80% of last year’s level, while Emirates and flydubai are operating at roughly 90% of their 2025 levels. 

“My confidence level is not misplaced that we should be back to 100% of our operation by about the end of the year,” he said, referring to international airlines resuming services.  

Dubai Airports expects full-year traffic to finish “in the 70 millions”—well below last year’s record—and putting its previous ambition of crossing 100 million passengers this year out of reach. 

The war has created an opening for competitors; previously, Gulf carriers accounted for roughly a third of traffic between Europe and Asia and about half of travel between Europe and Australia. By April, Middle Eastern transfer traffic on Asia-Western Europe routes had fallen 47% year over year, according to Alton Aviation Consultancy.  

Travelers chose alternative hubs including Seoul, Singapore, Hong Kong and Istanbul. Incheon International Airport said its transfer traffic rose 18% in the first half, helping it surpass Dubai and Heathrow in preliminary international passenger data. 

Griffiths believes much of that shift will prove temporary. He said resumed services into Dubai are returning with load factors in the 80% to 90% range. At the same time, airlines that moved aircraft onto busy European summer routes are likely to restore Gulf capacity as winter schedules begin. But he questioned why some government travel advisories had yet to catch up with conditions on the ground, arguing that “it is safe” to travel to the UAE. 

Several major foreign carriers have not yet resumed service: British Airways has suspended flights to Dubai until the end of October, whilst Lufthansa and Singapore Airlines are also holding off until late October. 

The conflict also forced Dubai Airports to improvise. During alerts, passengers were moved to safe areas, and aircraft could be landed quickly when threats emerged. When the airport was temporarily unable to fuel aircraft, Griffiths said it used a “splash and dash” contingency where planes departed Dubai International Airport, landed at Al Maktoum International Airport (an aviation hub located 40 kilometers southwest of Dubai’s city center) to refuel, then continued their journeys. 

Dubai Airports has also used the slowdown in traffic to continue upgrades across its infrastructure, including expanded self-service, biometric systems, smarter operational technology and passenger-flow improvements. Griffiths said the priority is to make the airport more efficient rather than to rethink its strategy in light of a crisis he considers temporary. 

That applies most clearly to Dubai’s $35 billion expansion of Al Maktoum International Airport, which remains on track to begin operations in 2032.  

It is expected to be the world’s largest airport, eventually accommodating 260 million passengers, as well as 12 million tons of cargo. Griffiths said 10 million work hours have been logged on the project over the past 15 months, alongside 17,000 concrete piles and 45 million cubic meters of earth moved. 

He said the lesson from previous shocks—from the financial crisis to the Covid pandemic—is not to confuse short-term disruption with a long-term structural shift. When he asked Dubai Airports’ chairman whether the war changed the emirate’s aviation plans, Griffiths said, “It was the shortest conversation, and the answer was ‘absolutely not’.” 

This story was originally featured on Fortune.com

The United Nations’ food agency said Tuesday it is halving food assistance in the Israeli-occupied West Bank because of severe funding shortages and warned it may be forced to make further cuts in Gaza, where it supports 1.5 million people.

The U.N. World Food Program said it will reduce assistance in the West Bank from 400,000 people to 200,000 beginning this month. In Gaza, it reduced the value of cash assistance by 40% for 375,000 people in July and said additional cuts will be unavoidable unless it secures new funding.

“In Gaza we’ve already had to start reducing the amount of cash that we’re giving to around 75,000 families,” Claire Nevill, WFP spokesperson for the Palestinian territories, told The Associated Press on Tuesday. “Now, the same is happening for WFP’s West Bank operation.”

WFP said it urgently needs an additional $386 million over the next six months to assist an estimated 2 million food-insecure people across Gaza and the West Bank.

“All of this pressure has been building in terms of the economic downturn coupled with increasing settler violence and incursions on land, people sort of losing access to their livestock and agricultural land, which has also now had an impact on people’s food security,” Nevill said.

“So just as the situation is getting worse and we should be scaling up to meet these needs, we’re unfortunately having to scale back and effectively halve the amount of people that we can reach because of the lack of funding,” she added.

The world’s leading authority on food insecurity, the Integrated Food Security Phase Classification, said in July the situation in Gaza had improved since famine was declared in parts of the Palestinian territory last year, but warned conditions were still at crisis levels.

Also in July, the WFP cautioned that funding shortages and what it described as donor fatigue could force deeper reductions in Gaza and said it needed more than $420 million to sustain operations in Gaza and the West Bank through the end of the year.

In the West Bank, WFP estimated in its Tuesday report that 900,000 people are food insecure. The agency said increased violence, displacement and economic deterioration have disrupted livelihoods and deepened hunger, while food needs have more than doubled since 2023.

“Families who were once thriving and earning a living from their land are now unable to afford enough food,” Shaun Hughes, WFP’s country director, said in a statement Tuesday. He added the agency was being forced to prioritize the most vulnerable households while cutting assistance to others despite widespread need.

According to WFP’s latest food security analysis, 76% of households in the West Bank have experienced a significant decline in income. One-third cannot afford a nutritious diet, while food and fuel prices remain among the highest in the Middle East. Rural communities and the southern West Bank are among the hardest hit, the agency said.

Humanitarian conditions also remain severe in Gaza. According to recent U.N. figures from aid agencies, about one in five families ate only one meal a day, while half reported running out of food entirely at least once in the previous month. Aid agencies say cash shortages and soaring food prices have made it increasingly difficult for families to obtain basic necessities.

According to WFP, three-quarters of households in Gaza had difficulty accessing markets in August. The U.N. and its partners recently collected hundreds of pallets of food, fuel and other supplies through the Kerem Shalom crossing and distributed food parcels and wheat flour to more than 600,000 people during the August aid cycle.

Despite a ceasefire signed by Israel and Hamas in October 2025, ongoing hostilities and security restrictions continue to hamper movement and recovery efforts, according to U.N. agencies.

The U.N. says more than 2 million people have been displaced during the war and that much of Gaza’s civilian infrastructure, including homes, hospitals, schools and water facilities, has been damaged or destroyed.

This story was originally featured on Fortune.com

Of the child safety measures agreed to in Meta’s landmark legal settlement, no category contains more detailed requirements than its commitments on age assurance — and for good reason. No matter how well the protections work, they are of little use if Meta can’t tell which users are kids.

Determining which accounts belong to children is also the linchpin of child safety measures that other companies, including Google, TikTok, and Roblox, are ramping up as public and regulatory sentiment turns against them in the U.S. and other countries.

The Meta settlement — with protections the company says could become industry standards — will give the social media giant a year to refine its approach. But the options are limited, and improving age verification technology presents steep challenges, including navigating privacy implications not just for children but for adults who could have to prove they are over 18.

The $18 billion settlement resolves claims from state attorneys general that it hurt children’s mental health by deliberately designing its Instagram and Facebook platforms in ways that hooked young people’s attention. The agreement announced last week with the Menlo Park, California, company involves 48 states plus the District of Columbia and U.S. territories.

Tech companies lean on AI for estimating ages

In the offline world, people use government-issued IDs to prove they are over 18 or 21, whether they want to get into a club or buy cigarettes. While this technically works online too, many people are not comfortable giving a Big Tech company like Meta or Google access to their driver’s license or passport.

Meta accepts IDs to check ages, and promises to delete them within 30 days. Beyond IDs, though, a growing number of tech companies are using artificial intelligence to try to determine users’ ages. This is not true age verification. While AI can estimate a person’s age, often accurately, it can’t verify a person’s birth date.

Last year, Google began using an age-assurance system for YouTube that relies on AI to differentiate between adults and minors based on their watch histories. Meta does the same.

AI age checks use clues to determine how old someone is. On Instagram, this includes “contextual clues” such as posts about birthday celebrations or school, and things like which posts people like, comment or linger on, who their friends are, as well as the times of day they log in. For instance, someone who is offline during school hours would likely be school-aged.

In a blog post last year, Meta said its AI also looks at photos people post for visual cues, including “height or bone structure, to estimate someone’s general age; it does not identify the specific person in the image.”

“By combining these visual insights with our analysis of text and interactions, we can significantly increase the number of underage accounts we identify and remove,” the company added.

Under the settlement, Meta agreed to strengthen its age-checking technology using its own tools, as well as third-party ones, 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 one of its platforms, Meta will check the ages of their friends, too.

Face scans count as age assurance on some sites

Identity verification companies like Yoti and Persona, which work with tech companies like Meta and Roblox, use video selfies people upload to estimate their ages.

The gaming site Roblox, which, unlike social media platforms, allows kids under 13 to sign up, requires players to take a video selfie to estimate their age. Roblox says the videos are deleted after the age check is processed. Users are not required to submit a face scan to use the platform, only if they want to chat with other users.

Some users have reported inaccuracies, and some experts have expressed caution about the reliability of facial age estimation tools, especially for women or certain racial or ethnic groups. But Matt Kaufman, chief safety officer at Roblox, said last year that for people between about 5 and 25, the system can accurately estimate their age within one or two years.

Some websites and social media companies such as Meta have argued that age verification should be done by app store owners, such as Apple and Google, and not individual platforms. This would mean that app stores would have to verify their users’ ages before they allow them to download apps. Unsurprisingly, Apple and Google disagree.

“Billed as ‘simple’ by its backers, including Meta, this proposal fails to cover desktop computers or other devices that are commonly shared within families. It also could be ineffective against pre-installed apps,” Google said in a blog post last year.

Privacy concerns remain an obstacle

Age checks are meant to keep kids out of online spaces that are not safe or appropriate for them, whether that’s social media or adult websites. But critics of online age verification see a worrisome trend toward a less secure, less private and less free internet, where people can be denied access not just to pornography but news, health information and the ability to speak openly and anonymously.

What’s considered harmful to minors can be subjective, and this is where experts believe laws requiring age verification can run afoul of the First Amendment. The question could lead to people being required to verify their ages to access anything, from Netflix to a neighborhood blog.

The digital rights group Electronic Frontier Foundation has been warning for years about “age-gating the internet.”

“(No) matter the method, every system demands users hand over sensitive and immutable personal information that links their offline identity to their online activity,” EFF writes in a blog post. “Once that valuable data is collected, it can easily be leaked, hacked, or misused.”

This story was originally featured on Fortune.com

The Barroso family rose at dawn and hitched a cart to a borrowed horse to take one of their daughters to school on Tuesday, the first day of classes for many students across Cuba.

As the crises deepen on the island, thieves recently stole the family’s horse and killed it for food, so the Barrosos now rely on a borrowed animal for transportation, because school buses aren’t running.

Fuel, teachers and uniforms remain scarce as the new school year begins.

More than 10,000 schools are expected to open their doors on Tuesday, welcoming an estimated 1.4 million children and more than 135,000 teachers.

The island’s education minister acknowledged ongoing challenges, calling the reopening of schools a “heroic act” for an island whose free and universal education system was long considered a crowning achievement of the 1959 revolution.

But the island’s multiple crises forced the government to end the previous school year in June — one month early. Conditions have only worsened as the island’s power, water and fuel shortages persist amid an ongoing U.S. oil embargo and longtime sanctions.

“This academic year has been the most disastrous one,” said Leanete Barroso, mother of two daughters.

Cuban schools lack teachers

A widespread lack of professors prompted the government to send more than 3,000 teachers from Cuba’s eastern provinces to schools in Havana as teacher coverage fell to 73% across the island. The vacancies are blamed on low salaries and emigration.

Education Minister Naima Trujillo Barreto implored Cubans to help fill the gaps, calling on university students, professionals and even retirees.

“Everyone is invited — anyone who has something to contribute,” she said at a recent news conference. “You might be at home, perhaps retired, and eager to meet a group of children who need lessons in the very subject you know better than anyone else.”

Meanwhile, parents and those in the private sector are expected to help provide school meals, with some schools only reporting a 15-day supply of food, according to state media.

School uniforms also were largely unavailable as the new school year began, meeting only 12% of demand at the primary level.

“No workshop can operate without electricity. Without fuel, we cannot transport fabric from one part of the country to another,” Trujillo Barreto said. “Without money or the ability to import, no fabric comes in either.”

She said that students would be allowed to wear “comfortable clothing” but are still required to tie on their traditional red and blue kerchiefs.

Cuba’s daily grind takes its toll

The day before classes started, Yurima Izaguirre struggled to pull up last year’s school uniform on the youngest of her three children. She grimaced when she saw that her 7-year-old daughter had nearly outgrown the burgundy skirt.

“You’re going to suffocate after a while in the classroom,” Izaguirre said.

No new uniforms were available for her two daughters, but she was able to find two shorts for her son.

A surge in the cost of goods in Cuba meant Izaguirre didn’t have enough money to buy new shoes for her children. For the first time ever, she had to buy school supplies for the three of them. The government was unable to provide them.

Izaguirre worries the most about having to feed her children lunch every day during the school week, another first for parents.

Izaguirre planned to prepare what she and her husband, who works as a lifeguard, could afford: pasta, tuna and mayonnaise: “It has to last.”

Before classes started, Izaguirre told her children that they needed to pack their small, rechargeable fans — each labeled with their name — so they won’t swelter inside the classroom, with temperatures hovering above 90 F (32 C). Like many Cubans, they’re often hit with daily outages lasting more than 20 hours — so they recharge the fans at the home of a kind neighbor.

The longer-than-usual vacation before classes also hit the family hard. Usually, they would visit the beach, a park or the zoo, but with no transportation available given a lack of fuel, Izaguirre said that she would organize picnics beneath a tree near their home.

Her children give her emotional strength, but daily life in Cuba has hit Izaguirre hard: “In my head, I think of doing 1,000 things, but I have to keep going for their sake.”

This story was originally featured on Fortune.com

Rising oil prices are raising inflation fears and pushing bond yields higher, weighing down stocks on Wall Street. The S&P 500 fell 0.7% Tuesday. The Dow Jones Industrial Average fell 299 points, or 0.6%, as of 9:33 a.m. Eastern time. The Nasdaq composite fell 1.4%. Much of the pressure being felt by Wall Street is coming from a sell-off in U.S government bonds. The yield on the 10-year Treasury rose. Oil prices continued climbing, putting more pressure on inflation. Markets in Europe were mostly lower and markets in Asia were mixed.

THIS IS A BREAKING NEWS UPDATE. AP’s earlier story follows below.

A heavy sell-off in U.S. government bonds is gaining momentum with anxiety about persistent inflation driving prices lower.

The yield on the two-year Treasury, which closely tracks expectations about Fed moves, rose to 4.35%, on Tuesday. That’s up significantly from about 3.50% at the beginning of 2026. Meanwhile, the yield on the 10-year Treasury rose to 4.79% from 4.73% late Friday. That’s the highest it has been since January 2025.

The bond sell-off is global, with other nations facing the same economic pressures.

Japan’s 10-year bond yield touched 3%, a level it hasn’t hit since 1996. In Germany, the 10-year bond yield also rose to 3%, a 15-year high.

The U.S. debt surpassed $40 trillion two weeks ago, a shocking milestone as defense costs and interest on the burgeoning deficit make up an enormous share of federal spending. But investors are asking governments across the globe to pay more, with the perceived risk rising.

Futures for the S&P 500 index fell 0.6%, while the Dow Jones Industrial Average declined 0.8%. Nasdaq futures tumbled 1%.

The U.S. attacked rocket launchers on an Iranian island on Sunday, saying they were preparing to launch mines into the Strait of Hormuz. Meanwhile, the United Arab Emirates said it intercepted an Iranian drone over its waters on Monday.

September is off to a tepid start, a day after Wall Street closed out August on a downbeat note. The S&P 500, Dow Jones Industrial Average and the Nasdaq composite all declined on Monday. And the losses were broad, with nearly every sector within the benchmark S&P 500 finishing in the red.

Tech stocks are also back in focus, with shares of Nvidia, Microsoft, Alphabet and Micron Technology all declining before the market open.

Oil prices remain high as the Iran war has curtailed traffic in the Strait of Hormuz, which once accounted for about 20% of the world’s oil shipments.

Brent crude gained 1.7% to $92 per barrel on Tuesday, following a 2.7% rise on Monday. U.S. benchmark crude climbed 2.2% to $87.67 per barrel.

The national average for gasoline in August has been above $4 per gallon every day of the month for the first time ever, according to the AAA. It has been the most expensive August at the pump on record, outpacing even the enormous supply chain crunch during the COVID-19 pandemic in 2022.

Higher energy prices have fueled already elevated inflation, which remains well above the Federal Reserve’s 2% target. Many experts are expecting a rate hike soon after Fed Chair Kevin Warsh — in a speech last week at a conference of central bankers — left the door open to it if inflation didn’t improve.

The government’s next report on prices comes out just days before the Fed meeting and could play an outsize role in determining whether the central bank acts.

Also this week, the government issues its latest monthly jobs report, which could also play into the Fed’s decision on interest rates.

Thought the unemployment rate remains low at 4.1%, the job market stalled unexpectedly last month as inflation continues to squeeze businesses and households.

A weakening employment market could create a tough situation for Federal Reserve, which has to balance fighting inflation with supporting full employment. Its main tool for managing that “dual mandate” remains interest rates. Raising its benchmark rate to bring down inflation could further damage the jobs market; cutting its benchmark interest rate to help support employment can exacerbate inflation.

In Europe, Germany’s DAX lost 1%, while the CAC 40 in Paris gave up 0.4% and Britain’s FTSE 100 declined 1%.

Asian markets were mostly lower, with shares in online fast-fashion retailer Shein sliding as much as 10% after they began trading in Hong Kong on Tuesday. They closed 4% lower.

—-

AP Business Writers Michelle Chapman and Matt Ott contributed to this report.

This story was originally featured on Fortune.com

Army Secretary Dan Driscoll is stepping down after just 18 months on the job. The White House announced his departure after Driscoll spoke with President Donald Trump about the current state of the Army, an Army official told The Associated Press on condition of anonymity. He’s only the latest top military leader to leave during the Trump administration.

The U.S. Postal Service is rushing to launch a hastily built, error-riddled computer system that could jeopardize the process of mail voting relied upon by one-third of all voters, according to a whistleblower statement released Tuesday by Democratic Sen. Richard Blumenthal.

And in a blow to congressional authority, a divided Supreme Court is allowing construction to continue on Trump’s $400 million White House ballroom as he flexes unprecedented executive power and remakes the capital in his image.

Here’s the Latest:

Democrats are deeply concerned about the Pentagon’s leadership

Rhode Island Sen. Jack Reed, the ranking Democrat on the Senate Armed Services Committee, said in a statement that Driscoll “worked to transform the Army for the battlefield of the future, and he engaged with Congress candidly and consistently.”

“Secretary Hegseth is cultivating a culture where dissent is punished and competence is secondary to personal allegiance,” said Reed, a former Army officer, while noting that “hundreds of thousands of soldiers are deployed around the world, many in combat zones.”

Rep. Jason Crow, who serves on the House Armed Services Committee and co-chairs the House Army Caucus, said the “Pentagon right now is, in some cases, rudderless,” with so many senior leaders gone. “So, I’m gravely concerned about it,” he said.

Republicans and Democrats alike praise the outgoing Army secretary

Rep. Steve Womack, an Arkansas Republican who had served in the Army National Guard, said on X that Driscoll is “a transformative leader who brought common sense and discipline to the Department of the Army.”

“I had the pleasure of working with him closely, and appreciated how engaged he was from mentoring cadets at West Point to leading the Army at the Pentagon,” Womack wrote. “The Army is better equipped and prepared to tackle emerging threats because of his leadership that strengthened the force while putting our Soldiers first.”

Rep. Jason Crow, a Colorado Democrat and former Army Ranger, told reporters that he worked well with Driscoll despite their policy differences.

“It was actually nice, I will say, having somebody who was smart, that knew the policy, that wanted to do the work, that had a background in these issues,” Crow said, adding that “it’s unfortunate that somebody like that can’t last in this administration.”

Exit follows other Army departures, rollback of drone program

Driscoll’s departure follows the ouster of one of his allies from the Army as well as the rollback of a drone program he had championed. Hegseth had suddenly ousted the service’s top uniformed leader, Gen. Randy George, in April, while the Army’s commander in Europe and Africa, Gen. Christopher Donahue, unexpectedly stepped down in June.

Gen. Christopher LaNeve took George’s place and directed a Europe-based Army unit that was building its own drones to end its efforts.

Driscoll told Congress in April that the military’s civilian leadership gets “to pick the leaders that they want.”

Driscoll was an unlikely negotiator in Ukraine-Russia war

Driscoll is an Iraq war veteran, tech investor and former adviser to Vice President JD Vance, whom Driscoll met at Yale Law School. When nominating Driscoll in 2024, Trump called him “a disruptor and change agent.”

As Army secretary, Driscoll was tapped for the unusual role of key negotiator to try to end the war between Russia and Ukraine. He was also a major force behind trying to cut the red tape for military contractors to quickly develop more drones and counter-drone capabilities as warfare rapidly changes around the world.

No reason was given for the departure of Driscoll, but tensions with Defense Secretary Pete Hegseth have been widely reported.

House lawmakers return to Washington with a stopgap funding bill atop the list of priorities

House lawmakers returned to Washington on Monday with a short to-do list after five weeks back in their home districts. The first order of business is most likely a vote on a stopgap spending bill designed to keep the federal government fully funded through early December, removing the possibility of a shutdown before the midterm elections.

With election season getting underway, votes are also expected on measures designed to amplify the GOP’s messaging strategy going into November, most notably a resolution condemning socialism. Republicans are trying to tie the Democratic Party in general to the democratic socialist candidates who have succeeded this year in running for office.

“The first thing is we’ve got to make sure this government gets funded and we don’t have another Democratic shutdown,” Speaker Mike Johnson told reporters as he prepared to open the chamber for business. The funding vote will most likely take place Tuesday.

▶ Read more

Postal Service is rushing to comply with Trump’s order, with millions of votes at stake

The U.S. Postal Service is poised to launch a hastily built, error-riddled computer system that could jeopardize the process of mail voting relied upon by one-third of all voters, according to a whistleblower statement released Tuesday by Democratic Sen. Richard Blumenthal.

The effort is part of the implementation of Trump’s executive order limiting mail voting and has been temporarily halted by a federal judge. The ruling is being appealed by the administration and the directive may ultimately end up back before the U.S. Supreme Court.

The disclosure from the unidentified whistleblower contends that the Postal Service has been scrambling to try to build a complex system that would normally take a year or more in only three months, sometimes in defiance of the judge’s orders halting work. The goal is to have it ready for use in the midterm elections.

The disclosure contends that, due to the new procedures, a single scanning error in a batch of tens of thousands of ballot envelopes could prevent any from reaching voters. It details a system that election officials have already warned could not be implemented before the first mail ballots begin going out Friday in North Carolina.

▶ Read more

With elections nearing, battleground Republicans still waiting for Trump’s campaign wallet

When Vice President JD Vance stepped on stage in Michigan on Monday, it was MAGA Inc. writing the check for the campaign event. It was a rare bit of spending in a battleground state from a powerful political organization that has mostly been missing in action this year.

With just over two months to go before the midterm elections, President Donald Trump is still sitting on most of his colossal war chest, and he’s been growing his money pile instead of spending it.

Even though the president and his team insist they’ll open their wallet soon, frustrations are growing as Republicans try to maintain control of Congress despite headwinds partially of Trump’s own creation. While the president remains the party’s best asset in turning out its base, he is historically unpopular, with low marks on his handling of the economy and the ongoing war in Iran.

The ticking clock narrows Republicans’ options because time on the airwaves gets more limited and more expensive as the election draws near.

▶ Read more

Supreme Court lets White House continue construction on Trump’s $400 million ballroom

A divided Supreme Court on Monday allowed the White House to continue construction on a $400 million ballroom project as lawsuits play out, a win for Trump as he flexes unprecedented executive power and remakes the capital in his image.

The apparent 5-4 decision replaces a temporary order issued earlier this month, shortly before a court-ordered halt would have gone into effect.

That temporary order was signed by Chief Justice John Roberts, assigned to handle appeals from the capital. But he publicly disagreed with the latest decision, writing a strong dissent that said the project is likely unlawful because it hasn’t been approved by Congress.

“Today’s decision is no victory for the separation of powers,” Roberts wrote. The court’s three liberal-leaning justices agreed.

The majority, on the other hand, found the National Trust for Historic Preservation likely did not have the legal right to challenge the project.

▶ Read more

Army Secretary Dan Driscoll is stepping down after 18 months on the job, White House says

Driscoll’s departure is only the latest of a top military leader during the Trump administration, and no reason was given Monday. He’s a friend of Vice President JD Vance, but tensions with Defense Secretary Pete Hegseth have been widely reported. The U.S. military leadership has seen a series of shakeups, with the Army especially seeing major upheaval.

A U.S. Army official, who was not authorized to comment publicly and spoke on condition of anonymity, said Driscoll spoke with President Donald Trump on the current state of the Army and submitted his resignation. The official did not provide additional details. The Pentagon referred questions to the Army. Driscoll’s resignation was reported earlier Monday by The Wall Street Journal.

▶ Read more

By Ben Finley and Aamer Madhani

This story was originally featured on Fortune.com

The sole pipeline that supplies water for millions of Grand Canyon visitors was badly damaged by weekend flash flooding, setting up a tricky repair scenario and upending a yearslong construction project to upgrade the line laid over terrain so rugged helicopters are often required to make basic fixes.

“We anticipate miles of the pipeline being destroyed,” Grand Canyon Deputy Superintendent Brian Drapeaux said, but a full survey of the devastation won’t be completed until Tuesday at the earliest.

From flyovers, park officials estimate that roughly 40% of the pipeline endured damage. The long-term implications to the park’s infrastructure, visitors and year-round residents weren’t immediately clear. The 12.5-mile (20-kilometer) long Transcanyon Waterline has endured dozens of breaks in recent years. A more than $200 million repair and replacement project was well underway when the floods hit.

At least two people died in the floods, and one person remained missing Monday night after rescuers saved some 80 people over the weekend, officials said. The park halted overnight accommodations, and officials said significant water reductions are needed — short showers, turning off the tap while brushing teeth and flushing toilets selectively to protect the limited stored water on the South Rim, where most visitors go.

“They have got to keep law enforcement going, they have to keep rescue operations going, backcountry operations, park admin going, the clinic and the campground,” said longtime river runner and Grand Canyon hiker Tom Martin. “In the past, the park has hauled water” in tanker trucks.

Even routine repairs can be tricky

The pipeline, which was finished in the 1970s, has broken more than 85 times since 2010. It moves water from one side of the canyon, down to its base, across a bridge and back up the other side of the rough landscape. Any effort to fix a break from rock falls or flash floods is time-consuming, precarious and costly, so addressing the latest damage will be complicated.

The pipeline upgrade project, which began in 2023, took more than a decade to design, consider and fund. It involves installing helicopter pads, building water treatment facilities and replacing several miles of pipeline. The project also sought to move the water intake for the pipeline further down the canyon to Bright Angel Creek. That would place it closer to Phantom Ranch, a popular lodge at the base of the canyon. An evaluation of the project in 2018 said key infrastructure for the intake would be outside of areas likely to flood.

Basic decisions were complicated. For example, the original pipe was aluminum, which is lighter and easier to move into the steep canyon but is susceptible to breaking. Engineers picked a flex-steel material to balance weight with durability.

Frequent breaks cause water restrictions

Even routine pipeline breaks force water restrictions. In 2016, a hotel on the canyon’s rim that offers luxury dining and boasts about its history hosting presidents was forced to reduce dishwashing and use paper plates and plastic utensils.

In 1995, a 28-day repair effort on the pipeline required officials to truck in water.

The park had water conservation measures in place even before the latest bout of flooding and damage.

The flooding has forced parts of the park to close and additional flooding was possible Monday, according to the National Park Service.

“Consider changing your plans if you were going to hike, boat, or paddleboard to a slot canyon or normally dry wash. If you do still decide to recreate, check in at a nearby visitor center or ranger station,” a notice on the park website said.

Officials said previously that the repair and replacement project would ensure that the park will be able to meet its water supply needs for the next 50 years or more. The park had roughly 4.4 million visitors last year.

When the original pipeline was nearing completion in the 1960s, more than a foot (0.3 meters) of rain fell, flooding the inner canyon. That rare, historic flood caused 40% of the project to be destroyed, officials said.

___

Associated Press journalist Hannah Schoenbaum contributed from Salt Lake City. The Associated Press receives support from the Walton Family Foundation for coverage of water and environmental policy. The AP is solely responsible for all content. For all of AP’s environmental coverage, visit https://apnews.com/hub/climate-and-environment

This story was originally featured on Fortune.com

At 88, Jared Diamond has metal spikes installed in front of his house. He put them there, he told Fortune, to protect against angry anthropologists.

“Half of anthropologists have been very helpful to me,” Diamond cheerily said in a Zoom interview from his office in Los Angeles, “and half resent the fact that I — a gallbladder physiologist who happened to have learned a lot about languages and genetics — am trespassing on their specialty.” He called them a “contentious lot,” adding, “they’ve been really nasty.”

It’s a well-earned position for Diamond, the contrarian who won a Pulitzer for his smash 1997 non-fiction hit Guns, Germs, and Steel. Trained as a physiologist at Cambridge, he built a second career studying birds in New Guinea and a third in history and geography. Now Diamond has published what he calls his final book: Profits, Prophets, Coaches, and Kings, out this fall from Mariner Books. The argument at its center is one he has been circling since childhood — shaped, he said, by growing up in Hitler’s world. He writes in the book’s prologue of his family of Jewish immigrants from Eastern Europe and a wife who lost many Polish relatives to the Holocaust. Do individual leaders actually change the course of events, or are they, as Tolstoy put it, history’s slaves?

Diamond points to how only 20% of his neighboring Angelenos bother to vote for the city’s mayor—evidence of a widespread, unexamined conclusion that leaders don’t, in fact, matter. Boards of directors, on the other hand, pay chief executives hundreds of millions of dollars a year on the belief that they matter very much.

He points to UCLA again, which has spent over $150 million buying out coaches’ contracts since the legendary John Wooden days. (The book has a chapter on the strange alchemy of the Wizard of Westwood.) “I could have told the chancellor [not to do that],” Diamond said, “but they never asked me, that the losing records of every one of our basketball coaches are what you expect from throwing dice.”

The industry decides

Diamond frames his book as a rejection of both poles of a century-old historical debate. On one side fits Thomas Carlyle’s “Great Man theory,” the idea that history is essentially the biographies of major players. On the other is Tolstoy’s view, dramatized in War and Peace, that leaders are essentially puppets of forces too large and diffuse to control.

For business leaders, Diamond’s analysis leans on what he calls “natural experiments,” a method borrowed from epidemiology and field biology: when researchers can’t ethically run controlled trials, they look for situations where nature runs them anyway. His favorite example: John Snow’s removal of the handle from London’s Broad Street pump in 1854, proving cholera was waterborne years before anyone understood germ theory.

Diamond looks at CEOs, political leaders, and sports coaches, comparing what happens when they die, get fired, get sick, or get replaced, and comparing those situations with cases where they don’t.

Diamond frames his book as a rejection of both poles of a century-old historical debate. On one side fits Thomas Carlyle’s “Great Man theory,” the idea that history is essentially the biographies of major players. On the other is Tolstoy’s view, dramatized in War and Peace, that leaders are essentially puppets of forces too large and diffuse to control.

The point of his new book, Diamond told Fortune, is figuring out the contours of his qualified yes answer: leaders matter, but only in the presence of other identifiable, sometimes measurable conditions. His studies attributed somewhere between 6% and 29% of the variation in company earnings to individual CEOs, but also that the industry and particular company are important variables.

Diamond said he has a friend who designs T-shirts for teenagers. “If you foresee what the teenagers will buy in the next two weeks, you make great sales,” he said. “My friend is very rich as a result. He deserves every million dollars he earns.” The CEO of a public utility, on the other hand, can barely decide anything for the business because of strict regulation. “You should not be paying a large salary to a CEO of a public utility company when they don’t have much effect, and there are 49 other people who could do an equally good job.”

He ranks CEO influence highest in computers, perfume and cosmetics, movies, toys and restaurants, because consumer taste shifts faster in those industries than capital structure. He ranks it lowest in utilities, steel, shipbuilding and railways. But what really surprised him is what he found out about politics.

Born at the right time

Diamond assembled a table of 22 major 20th-century politicians — deliberately excluding Donald Trump and other recent leaders because “we can’t yet pass judgment” — and asked one question: how many stepped down voluntarily?

Only two: Lee Kuan Yew of Singapore and Nelson Mandela.

“Once you become a leading politician,” he said, “and you’ve developed the conclusion that ‘I can make a difference,’ once you’ve drawn that conclusion, you’re not going to change your mind.” He cited Winston Churchill’s refusal to step down late in his career, despite mediocre ratings by the public and by historians, as a key example.

Going further back in history, Diamond found that timing can be inseparable from what we call successful leadership. Genghis Khan, for instance, benefited from something we only learned about in the past decade, as researchers studying tree rings found that he lived through the wettest decades on the Mongolian steppe in 2,000 years. That produced the grass, livestock and horses that made his conquests possible.

“If Genghis Khan had been born 20 years later and 200 miles away,” he said in very Diamondesque fashion, “we wouldn’t know anything about him.”

Some leaders seize opportunity when it arises, he said, like Bill Gates, who told Paul Allen that the timing wasn’t ripe for personal computing until he saw the January 1975 cover of Popular Electronics featuring the new Intel 8080 chip. “He said, ‘This is the time, let’s go fast, because other people will do it.’” In this way, he said, Gates was similar to Charles de Gaulle, who “twice engineered the opportunity for himself to lead France” — and twice stepped away once conditions turned unfavorable.

In this light, Mark Zuckerberg was more like Genghis Khan, arguing that social media existed before and after him. “He was there at the right time, because there had been changes in the technology that made social media much more feasible under Mark Zuckerberg. But he didn’t know this, so he was like Genghis Khan. He was at the right place at the right time, but he didn’t know it.”

Carl Sagan, Diamond notes, offers a parallel to his own career — a physicist and astronomer whose television success made him the target of a rare deelection attempt at the National Academy of Sciences. Diamond said the criticism was undeserved, but came from a place of, “Oh, he’s got a pretty face and he appears on television.”

Diamond said he has drawn the same fire for the same reasons: methods that overstep, conclusions that synthesize where specialists expect excavation. He said he’s still traveling regularly to New Guinea, where he has been conducting fieldwork since 1964, and expects to write up 62 years of biological research before, as he put it, “that knowledge will die with me.”

As for what he learned about sports working on this book, Diamond let himself get a little sentimental, noting in his New England accent that he grew up in Boston and so “the two things that I most hope for are that the Yankees do badly and that the Red Sox do well.” (The author is also from Massachusetts but the western end, Amherst.) Diamond continued: “Two things that increased my faith in God were that Ted Williams’ last bat, he hit a home run.” The other was the famous comeback against the Yankees in 2004, when the Red Sox won four straight playoff games after getting into a 3-0 hole. “Those two things help me believe that there is a God who looks out on the affairs of humans and rewards virtue and punishes evil.”

This story was originally featured on Fortune.com

Good morning. GE Vernova is hiring a CFO who helped take Rivian public in the biggest U.S. IPO of 2021, putting Claire McDonough at the center of the EV maker’s shift from a closely held startup to a company answering to public-market investors. Her move brings that experience to the energy tech company and puts new attention on the finance team she leaves behind at Rivian.

McDonough will join GE Vernova in November and become CFO on Jan. 1, succeeding Kenneth Parks, who is retiring. Derek Mulvey, Rivian’s VP of finance, is expected to serve as interim CFO after McDonough leaves.

Rivian shares fell more than 6% intraday Friday after the announcement, while GE Vernova shares declined about 3%. Investors and analysts pointed to Rivian’s lack of a clear long-term successor as one factor weighing on its shares.

Cullen Rogers, portfolio manager of the Wedbush ReturnOnLeadership U.S. Large-Cap ETF, told me Rivian’s leadership pipeline helps explain the sharper reaction. “Markets don’t just react to who’s leaving; they react to how visible the succession plan is,” Rogers said.

GE Vernova’s announcement offers a clearer handoff, with Parks staying on as an advisor after McDonough arrives. Rivian has yet to name a permanent successor.

McDonough joined Rivian as CFO in January 2021, months before its $13.7 billion IPO that November. She also helped negotiate Rivian’s technology joint venture with Volkswagen, which agreed to invest up to $5.8 billion.

Rogers said that experience should translate to GE Vernova. “That’s a transferable skill set, not a personal one,” he said.

At Rivian, McDonough’s departure will test whether the EV maker has built a finance team that can carry on without her. “The next two quarters are the real test that will tell whether strong leadership was distributed across the team or concentrated in one office,” Rogers said.

Morningstar senior equity analyst Seth Goldstein expects Rivian to stay on its current financial course. The company is working toward profitability and positive free cash flow as it ramps up production of its lower-priced electric SUV, he said. Its finance team will also manage Rivian’s cash and may raise more capital if needed. Morningstar has raised its fair value estimate for Rivian to $22 a share from $20.

Goldstein expects Rivian’s capital allocation strategy to remain unchanged and points to its profitability guidance and timeline for reaching positive operating cash flow as key measures.

McDonough also leaves Rivian with a record of cost discipline. “During her time at Rivian, McDonough proved to be a finance leader with a focus on cost discipline,” Goldstein said. “I view this as an asset at any company.”

Sheryl Estrada
Sheryl.Estrada@fortune.com

This story was originally featured on Fortune.com

In 2005, two years before the iPhone came out, I was standing in a coffee line on my morning commute, watching everyone ahead of me go through the motions: order, wait, pay, wait some more. I thought there had to be a faster, better way to get my coffee. So I built one: the first text-message order for restaurants, years before smartphones made “ordering ahead” a category.

Over 20 years later, I’m the founder and CEO of Olo, the restaurant commerce platform behind 800-plus restaurant brands across 90,000-plus locations, including Shake Shack, Waffle House, Cracker Barrel, Five Guys, and Panda Express. I’ve spent two decades watching the restaurant industry innovate and iterate the ordering experience, and I’m confident now is the time for a completely new direction.

Here’s the problem restaurants and their guests have lived with since digital ordering went mainstream: they’ve only ever had two choices, and both come with a catch.

The first choice is third-party marketplaces, broadly known by consumers as delivery apps. When COVID hit and digital ordering became a necessity overnight, these platforms were the fastest way for restaurants to keep serving guests. And they worked. But the commissions these platforms charge, up to 30% per order, didn’t disappear after the pandemic. Today, 82% of brands are marking up menu prices on these platforms just to offset the fees, and more than half of those markups run 20% to 30%. That’s a tough ask for guests already stretched by rising costs, and a worse deal for restaurants: every order placed on a marketplace belongs to a guest the restaurant has to pay to reach again because it never owned that guest’s data in the first place. I call that a rented guest. Rent keeps the lights on, but it doesn’t build equity.

The second choice is first-party ordering, a restaurant’s own app or website. It solves the ownership problem, but it creates a scale problem. Acquiring a new guest this way can cost a restaurant up to $100, and even guests who download the app rarely open it again. Ask yourself how many restaurant apps you actually keep on your phone. Most people don’t, and they revert to the marketplace instead.

For 20 years, that’s been the whole menu: rent your reach from a third party or build ownership alone as a first party. But I don’t think the choice has to be binary anymore. What sits between first and third? Second.

In other industries, second party already means something real: brands sharing infrastructure so they can serve customers better without giving up what’s theirs, the way Shopify’s Shop App works across thousands of independent retailers. Restaurants don’t have an equivalent yet, and they need one urgently. The next era of ordering will be defined by this equivalent, allowing guests – and their AI assistants – the ability to order from a centralized location from the brands they choose to follow, with personalized offers, saved checkout data, and no markups or ads stretching their wallets or distracting from what they really want. 

A second-party network changes the economics for everyone. For restaurants, it means marketplace-level reach without the commission and without giving up the guest relationship. For guests, it means a seamless experience that finally feels like the restaurant knows them. I follow a vegan diet, and today, ordering from a new brand for the first time usually means getting the same generic promotions as everyone else, with the occasional steakhouse special included. It’s a small reminder the brand has no idea who I am. In a network where preferences travel with the guest, that changes immediately. The brand can skip the beef and start with something relevant just for me on day one.

I built the first text order because I was tired of waiting in a line that didn’t need to exist. Twenty years later, restaurants are stuck in a different kind of line, choosing between two models with their own unique drawbacks because no one built the third option yet. The brands willing to build it together, starting now, will be the ones still standing in the next era of ordering.

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.

Topics

This story was originally featured on Fortune.com

Good morning. On Fortune’s radar today:

• Venezuela oil pact ‘looks like an insider deal.’
• Iran again offers Trump the deal he wants but the president wants to ‘hit them hard’ instead.
• Markets: It’s messy out there.
• You’re not still using a 60:40 stock-bonds portfolio, are you?
• The one number Fed Chair Warsh is actually watching.
• There’s a global sardine crisis—and prices are through the roof.

➡️ 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 Nordic countries and the Benelux bloc follow distinct corporate governance models. In Sweden, Norway, Denmark, and Finland, employees have the right to board representation once a company grows beyond a certain size—though the thresholds vary by country.  

Despite their small size, these economies rank among the world’s most stable and have produced global powerhouses, including Novo Nordisk, Spotify, and Ericsson. Large Dutch companies, meanwhile, legally separate management from oversight, requiring a non-executive supervisory board to hold the executive team accountable.  

“These are high-trust cultures built on collaboration and consensus,” says Esha Mendiratta, an associate professor at Vlerick Business School in Belgium. “Hierarchies stay flat, and major decisions wait until every stakeholder has weighed in.”  

If any region should feel secure in its choice of CEO, it’s these. A new survey suggests otherwise.  

Heidrick & Struggles’ Route to the Top report, based on a 2026 poll of 1,033 CEOs and directors worldwide found that nearly four in ten European respondents doubt their CEO’s strengths match what their company will need over the next two to three years. In the Nordics and Benelux, that number jumps to half. 

Tenure may be the culprit 

There are several reasons why these consensus-driven systems might be experiencing, or simply reporting, this heightened mismatch.  

One possible explanation is how long chief executives stay in the job. “CEOs in the Nordics and Benelux stay in their roles far longer than in other markets,” says Jenni Hibbert, who leads Heidrick & Struggles’ Europea and Africa business. The average tenure for Belgian CEOs is nine years, the second longest in Europe, according to the firm’s data. In Sweden and Norway it’s seven years, while Finnish CEOs have the shortest average tenure, at two and a half years.  

Data analyzed by Spencer Stuart reveals a similar trend. Norway had the lowest CEO turnover rate among major European listed companies in 2025, with no new CEO appointments reported. France, in comparison, recorded 18 and Italy, six.  

“In the context of ongoing economic and technological volatility, one might argue appointments aren’t keeping pace with the shift in capabilities needed in the role,” Hibbert says. Today, those running an organization must contend with a multitude of pressures, from geopolitical headwinds and energy-price shocks, to AI disruption. It is little surprise, given these developments, that 90% of executives surveyed by Heidrick & Struggles’ expect their firm strategy and operating model to change in the near future.  

The concern is that business conditions are shifting faster than boards can respond. “Boards are currently struggling to find chief executives who can lead major organizational change,” observes Martin Hartley, group chief commercial officer at emagine, a talent consultancy headquartered in Denmark with a presence across Sweden, Norway, the Netherlands, Belgium, and Luxembourg. “It’s difficult to find candidates who have the ability to combine technology leadership, commercial acumen, and transformation experience.”  

The real danger, Hartley adds, is a boardroom that has lost faith in its CEO and chooses to do nothing about it. “Decisions take longer, transformation loses momentum, and talented people grow frustrated while the organization keeps telling itself the CEO just needs a little bit more time.”  

Insularity may be compounding the problem 

Mendiratta highlights a preference in parts of the Benelux for promoting leaders from within.  “Boards that draw CEOs from the firm’s internal ranks are sometimes less likely to bring in the fresh outside perspective needed to spot problems and shake things up,” she says. 

In the Netherlands, the average time an internal candidate spends at a company before becoming CEO increased from 9.5 years in 2023 to nearly 15 years in 2025, according to Heidrick & Struggles’ data.  

“Insiders are not inherently the wrong choice,” Mendiratta continues. “They typically bring deep firm-specific knowledge, which tends to be an advantage when conditions are relatively stable. However, during periods of turbulence, externally hired CEOs tend to outperform.” 

In the 87 years Heineken has been listed on the stock exchange, the Dutch brewer had never appointed an outsider as chief executive. That changed in January 2026, when Dolf van den Brink announced he would step down. Facing pressure to break from precedent, the board turned to Rafael Oliveira, CEO of JDE Peet’s—someone with no prior experience in the beer industry—to take over in October. 

Demand is growing for leaders with broader perspectives and experience across different markets and cultures, says Mendiratta. She adds: “In today’s landscape of intense geopolitics, with tariffs rising and supply chains moving closer to home, the traditional strategic playbook these CEOs rely on is breaking down, exposing a starker gap between their built-up strengths and what their companies need.” 

Retiring the old playbook 

The gap between current skill sets and future strategic needs reflects a deeper weakness in how boards approach succession planning, according to Hibbert. 

“The pre-anointed candidate is a very passive way of thinking about succession planning, which no longer passes muster,” Hibbert says. This pass-the-baton type of approach is unlikely to work in the current context where the external environment shifts frequently, she continues. “Naming an heir apparent years in advance risks handing the job to someone whose strengths no longer fit the moment or undermining their credibility even if they still do.”  

Mendiratta goes further, arguing boards should generally avoid naming a single successor until just a few months before a handover, keeping several viable candidates in play instead. “Boards should be asking: What type of CEO am I going to need in six months, two years and five years?” Ongoing feedback and reflection on the CEO’s leadership is vital, she adds. 

In Mendiratta’s experience, candidates who are identified early in their career tend to enact less strategic change once they take the top job. Similarly, outgoing CEOs who move into the chairman’s seat can create the same drag, she argues, constraining a successor’s room to reshape the company. 

For Hartley, it’s not a matter of choosing between insiders and outsiders. CEOs should not be assessed in isolation at all, he argues. “A better question is: Do they have the right executive team behind them? Boards are already paying closer attention to the rest of the C-suite, because that wider bench signals who the real succession candidates will be.” 

Candor, not crisis 

The report findings are not necessarily a sign of organizational dysfunction. “It’s possible companies in the Nordics and Benelux are simply more willing to admit misalignment than most,” Hibbert says.  

In these regions, major strategic decisions are structurally required to pass through more hands, whether through employee representatives or supervisory boards. Consensus-driven governance is built to surface disagreement rather than bury it, Hibbert explains, which may translate into more candid survey answers. “If governance quality alone explained confidence in leadership, these boards might report less uncertainty than most, not more.” 

Hartley warns that Nordic niceness must not be mistaken for soft-headednesss. “There’s a temptation for outsiders to see slower, more deliberate decision-making as a flaw, but that’s not the case,” he says. “What I see is more controlled, pragmatic, and rigorous processes that can still move very quickly when needed.” 

Perhaps the most important—and apt—lesson from the report is that even the most deliberate boardrooms, the ones built to weigh every voice and guard against rash decisions, canend up choosing a leader built for the world as it was, not the one now arriving.  

This story was originally featured on Fortune.com

Over the next decade, Asian families will transfer an estimated $10 trillion—nearly twice the size of Germany’s GDP—to the next generation. Yet there’s a nuanced story emerging behind that headline figure. As people live longer, they’re rethinking how preserve their own independence while still passing enough wealth to their children.

Asia is the fastest-aging region in the world. Fifteen percent of its population is over the age of 60; that share is projected to rise to 26% by 2050, according to the United Nations Economic and Social Commission for Asia and the Pacific. Some of the world’s longest-living populations are also in Asia, led by Hong Kong, where life expectancy is 85.5 years. In mainland China, life expectancy rose from around 52 in 1963 to 78 today.

This transformation isn’t just a demographic story. Greater longevity is reshaping how people think about wealth, care, and responsibility within the family.

For generations, many in Asia assumed that wealth would be passed to children, through the funding of education and home ownership, or leaving an eventual inheritance—with children later caring for their parents in old age. But today’s families are planning things differently. Our data shows that adults in Asia are increasingly prioritizing autonomy, health, and financial security over maximizing the inheritance they leave behind.

The next chapter of Asia’s wealth transfer won’t be defined only by the money parents leave behind. It will also be measured by something potentially more valuable: the freedom to support their own longer, more independent lives—and, in turn, relieve their children of the financial and emotional burdens of caregiving.

Manulife’s Asia Care Survey 2026 of 9,000 adults across nine Asia markets, shows how far this shift towards independence has progressed. Men in Asia anticipate funding 14 years of their own care in later life, and women 15 years, at a time when fewer older people than ever are living with their adult children.

The survey’s respondents overwhelmingly value independence and financial freedom over passing down tangible assets to their family. Regionwide, 83% said securing that freedom was more important than leaving their heirs the maximum amount of wealth.

Respondents plan to earmark an average of 68% of their money and assets to fund their own costs, including health care, as they age, leaving the rest to their children. This ratio varied across markets, with those in Taiwan expecting to spend the most on their own health and care at 78%, and those in the Philippines and Indonesia planning to spend the least, both at 60%.

This shift should be welcomed because it reflects a more sustainable response to longer lives. When people plan to fund their own needs later in life, they are ensuring that they don’t become a source of financial strain for the whole family.

This matters particularly in Asia, where the traditional model of care is under pressure. Families are smaller, adult children are more mobile, and older people are less likely to live under the same roof as the next generation. The United Nations Population Fund warns that changing family structures and migration are weakening informal support systems for older people across Asia-Pacific, even as many formal health and social-care systems remain unable to keep pace with demand.

In that context, independence is not individualism. It is resilience, giving older adults greater control over how they age, families greater flexibility, and governments and employers greater capacity to sustain retirement systems.

I

Many in Asia need to rethink how they unlock the full potential of their savings. Pension and household assets in Asia are still heavily weighted toward cash and government bonds, while holdings of shares remain low compared with developed economies, according to the Organization for Economic Cooperation and Development.

That conservatism doesn’t give the returns people need if they hope to live off their assets for longer.

Health planning is also essential. Our survey found that many people haven’t taken the necessary steps to be truly independent. Over 80% said preventive care was essential to a long life, yet only 26% actually went to early health screenings.

Insurers, employers, and governments have a role to play in helping people live independently in their later years, whether that’s advancing prevention and more flexible financial solutions or building stronger public-private partnerships.

Let’s start with prevention. Early screenings can catch illnesses sooner, when treatment is more effective and the chances of full recovery are far higher, yet too many people skip them until it’s too late. Insurers can build wellness checks and preventive screenings into the solutions people already hold, as Manulife has begun to do with early cancer detection. Employers can do the same through workplace health plans. Nobody plans a hospital visit for a disease they don’t know they have.

Financial solutions must evolve, too. Much of the insurance and savings landscape in Asia still operates on the outdated assumption that retirement at 65 lasts just a decade. That simply doesn’t fit a modern saver who may want to work until 70 and expects to manage their own care at 85. We need flexible coverage that stretches and adjusts alongside shifting lifespans to make self-reliance truly affordable.

Governments are tackling these same structural challenges, from updating public pension schemes to launching preventive-health programs.

That creates a clear opportunity for our industry to partner with them.

Examples of what works are already scattered across the region. Japan’s tax-free NISA accounts show how governments can successfully encourage households to move from static savings to active investments. In Hong Kong, the Mandatory Provident Fund—for which Manulife is the largest provider—demonstrates how compulsory schemes can help turn monthly wages into retirement assets. Singapore’s CPF LIFE scheme shows how retirement savings can be converted into income that lasts for life.

We see the same opportunity for public-private partnership in health and later-life care. Hong Kong’s Voluntary Health Insurance Scheme, where Manulife is one of the leading providers, already shows how government incentives and private coverage can work together. Under this model, governments set the mandate and tax treatment, and the private sector builds the products and carries the risk. Applied to health, that could mean tax relief for preventive care delivered through insurance plans, or national screening targets that insurers and employers are enlisted to help meet.

People in Asia plan to live long and stay independent, and they are reorganizing their wealth around that goal. The work of the next decade—for insurers, employers and governments alike—is to build the solutions that make this ambition achievable.

The region that manages this successfully will have done something greater than transfer its wealth. It will have shown the world how to live well with a population that lives longer.

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

The price of oil rose and stocks fell on Wall Street Monday after the U.S. launched its first military action in a month against Iran.

The S&P 500 index fell 0.4%. The Dow Jones Industrial Average fell 348 points, or 0.7%, as of 9:51 a.m. Eastern time. The Nasdaq fell 0.3%.

The losses were broad, with nearly every sector within the benchmark S&P 500 losing ground. Energy stocks, though broadly gained ground. Exxon Mobil rose 2.9% and Chevron rose 3%.

Markets were mixed in Europe and Asia.

GameStop jumped 4.2% after the video game retailer provided a preliminary second-quarter earnings outlook above its year-ago results. Shares of Aon fell 5.8% as the company announced that it was buying insurance broker USI Insurance Services from private equity firm KKR in a deal valued at $17 billion, including debt.

U.S. forces struck Iranian rocket launchers on the Strait of Hormuz on Sunday. Meanwhile, the United Arab Emirates said it intercepted an Iranian drone over its waters on Monday. The aggressive actions follow a lull in activity in the U.S. war with Iran, which has lasted more than six months.

The war has curtailed traffic in the Strait of Hormuz, which accounts for about 20% of the world’s oil shipments. Oil prices remain high after an initial surge earlier in the war and that has made everything from gasoline to shipped goods more expensive.

The price of Brent crude, the international standard, rose 3.2%, to $90.91 per barrel on Monday. The price swung between $72 and $102 last month amid rising and falling hopes for a deal to end the war.

The national average for gasoline in August has been above $4 per gallon every day in August for the first time ever, according to the AAA. It has been the most expensive August at the pump on record, outpacing even the enormous supply chain crunch during the COVID-19 pandemic in August 2022.

Higher energy prices because of the war have fueled already stubbornly high inflation. That has been weighing on household spending and consumer confidence. It has also given the Federal Reserve a more complicated path ahead for its interest rate policy.

The rate of inflation remains well above 3%, which is far beyond the Fed’s 2% target. Wall Street expects the central bank to raise interest rates at least once before the year ends in an effort to cool inflation.

The yield on the two-year Treasury, which closely tracks expectations about Fed moves, remained at 4.34% from late Friday. That’s up significantly, though, from about 3.50% at the beginning of 2026.

The yield on the 10-year Treasury rose to 4.76% from 4.73% late Friday. That’s back up around the level seen two weeks ago when the Trump administration took the unusual step of announcing it would intervene in the bond market.

The job market remains resilient, but is showing signs of weakening. Any increase to interest rates that could cool inflation also risks hurting the jobs market.

Later this week, the U.S. reports August jobs data. In July, the U.S. job market stalled unexpectedly as employers cut 23,000 jobs. Labor Department revisions slashed another 103,000 jobs from May and June payrolls.

___

AP Business Writers Elaine Kurtenbach and Michelle Chapman contributed to this report.

This story was originally featured on Fortune.com

The United Arab Emirates said that it intercepted an Iranian drone over its waters Monday, an attack that comes after the United States and Iran exchanged fire over the weekend for the first time in a month.

Iranian officials said that two people were killed and several others wounded in the U.S. attack Sunday night on Larak Island in the Strait of Hormuz. Iran responded by launching missiles at U.S. sites in Jordan, which were intercepted.

During the lull in fighting, the U.S. has been ratcheting up economic pressure on Iran with the hope of forcing concessions out of Tehran, including reopening the Strait of Hormuz to shipping.

After the U.S. and Israel attacked Iran on Feb. 28, Iran established a chokehold on the vital strait, though which a fifth of the world’s traded oil passed in peacetime, slowing ship traffic to a trickle. That has roiled the global economy, raising prices worldwide for energy and other goods.

As the fighting resumed, the price of Brent crude oil, the international standard, climbed above $90, up about 25% from the start of the war, presenting a growing problem for U.S. President Donald Trump before the midterm elections.

A return to open conflict would be dangerous for the region, where Iran has targeted U.S. military bases and infrastructure in Gulf countries since the war began.

In the wake of the renewed violence, Anwar Gargash, a diplomatic adviser to the UAE’s president, said on social media said “the state of neither war nor peace cannot be a sustainable solution.”

He called for a political solution that would return normal navigation to the Strait of Hormuz, but also a “more realistic approach” than the June memo of understanding between Iran and the U.S. which, he said, “failed to outline a practical and acceptable” road map.

UAE says Iranian attack is a ‘dangerous escalation’

In a short statement earlier, the UAE Defense Ministry said that it “dealt with” an Iranian drone that was detected over its waters approaching from Iran. There were no reports of damage.

“This dangerous escalation constitutes a blatant violation of the UAE’s sovereignty, security, and stability, and a direct threat to the safety of its citizens and residents,” the Foreign Ministry said.

The Defense Ministry later issued a second statement denying a claim by the Iranian army that the Al Minhad Air Base in Dubai had been targeted, saying it was false.

In the wake of the U.S. attack on Sunday, Iran’s Foreign Ministry said Monday that Iran would “not hesitate to exercise the inherent right of legitimate defense and will respond decisively to any military aggression.”

Iran says it has right to defend itself and will ‘respond decisively’

Before heading to attend the Shanghai Cooperation Organization’s summit in Kyrgyzstan with Foreign Minister Abbas Araghchi, Iranian President Masoud Pezeshkian, who has favored a negotiated end to the conflict, stressed that Iran wasn’t seeking war.

“But,” he said, “we will respond decisively to aggressors.”

The summit brings together a group billed as a counterweight to U.S. global influence, and includes Chinese President Xi Jinping, Russian President Vladimir Putin and Indian Prime Minister Narendra Modi. Pezeshkian was scheduled to meet with Putin on the sidelines on Tuesday.

US weekend attack comes after weeks of rising economic pressure

Meantime, U.S. Treasury Secretary Scott Bessent was set to host meetings of the Group of 20 finance ministers in North Carolina on Monday and Tuesday, which come as Washington pressures other countries to help the U.S. economically isolate Iran.

The shifting strategy centers on threats to punish any country or entity that continues to conduct business with Tehran.

Two weeks ago, the UAE suspended all trade with Iran after saying it had come under renewed fire from Iran with two missiles fired at shipping traffic splashing down harmlessly in the Persian Gulf. Iran denied the attack.

Bessent later suggested it was pressure from Washington that led to the UAE decision.

Trump stressed last week that he is “not in a hurry” to get Iran back to the negotiating table, and he continues making the case that the Islamic Republic’s leadership is on the ropes.

Appointments to Iran’s senior security leadership in the past month, however, have signaled Tehran’s defiance after weathering decades of sanctions as well as a lack of tolerance for dissent at home.

US military says strike on Iranian rocket launchers ‘precise’

On Sunday, U.S. Central Command said it had struck Iranian launchers after observing forces with Iran’s Revolutionary Guard Corps preparing to launch rockets with sea mines into the Strait of Hormuz. The U.S. military said last week that it had completed clearing sea mines from the waterway’s international shipping routes.

U.S. military officials disputed Iran’s claim that the latest American strikes were an act of aggression, calling them a “limited, precise action” against what they said were minelaying forces.

Iranian state television later showed what it said were ballistic missiles being fired at American bases in Jordan, and Jordan’s military said that it intercepted eight that had entered its airspace.

The last time the U.S. military confirmed targeting Iran was on July 29, when it announced a “heavy wave of strikes” on dozens of Revolutionary Guard targets including coastal surveillance and defense sites.

This story was originally featured on Fortune.com

Before becoming the founder of Bridgewater Associates, not to mention a celebrated author, Ray Dalio faced a moment of financial distress that reshaped his entire approach to investing and life.

After being fired early in his career, Dalio founded what would become the world’s largest hedge fund as an independent operation, run out of his two-bedroom apartment in New York City. Within a few years, he found himself “so broke” he had to borrow $4,000 from his father just to cover family bills.

“This was painful,” Dalio told a fellow billionaire, Carlyle Group cofounder David Rubenstein, in a conversation at New York’s 92nd Street Y in July 2025. But it also had a deep impact, he continued.

“That changed my approach to everything,” Dalio said, adding he learned two key lessons from this episode.

After striking out on his own to found Bridgewater in 1975, Dalio said he hit his lowest point around 1980 to ’81, when he calculated the U.S. had lent more money to countries than it could ever repay and predicted a major debt crisis. When Mexico defaulted on its debt in 1982, Dalio believed his position would pay off, even in the face of the severe economic crisis that he anticipated. However, he “couldn’t have been more wrong.” Instead of a downturn, the stock market went up, and monetary policy was eased, costing him dearly. This miscalculation left him financially devastated, forcing him to borrow $4,000 from his father to meet family expenses.

“Nobody does everything perfectly, not even Warren Buffett,” Dalio told Rubenstein, but this episode gave him the “humility” to go along with his “audacity,” he said, along with a very simple lesson in “the power of diversification.”

Dalio’s lessons

This humbling episode fundamentally changed Dalio’s perspective, he said, leading to two transformative insights:

• Lesson 1: Cultivating humility and questioning one’s own certainty. The experience made Dalio reflect deeply on how he could truly know if he was right. This new approach led him to a practice he began roughly 35 to 40 years ago: pausing to reflect and write down the specific criteria he would use to make a decision. This act of documentation forced deeper thought, and he later realized these criteria could be coded and back-tested to evaluate their effectiveness over time. This systematic approach to decision-making, which he calls “principles” (having written down thousands of them), became the bedrock upon which Bridgewater Associates was built. It’s also the title of Dalio’s New York Times bestseller.

• Lesson 2: Embracing the power of diversification. The crisis also led Dalio to appreciate diversification could reduce risk by up to 80% without diminishing returns. This revelation became the “bottom of Bridgewater,” he said, from which point the firm saw consistent positive returns, averaging roughly 11.8% over the subsequent 30-plus years, with only minimal annual declines. His investment mantra became “15 good uncorrelated return streams,” engineered to have similar expected returns, which he found dramatically lowers risk and boosts the return-to-risk ratio by a factor of five.

For Dalio, this near-ruinous period was not merely a setback but a profound educational experience that redefined his investment strategy and personal philosophy. Dalio said he now finds “great joy” in sharing these learned mechanics and cause-effect relationships with others. His goal isn’t to scare people but to provide understanding, operating on the principle that “if you worry you don’t have to worry, and if you don’t worry you need to worry,” as worry can prevent what one fears. His personal financial rock bottom ultimately became the foundation for his enduring success and his commitment to teaching others how to navigate complex financial landscapes.

Dalio’s new book on how countries go broke

Going broke was on Dalio’s mind because of the subject of his book: How Countries Go Broke: The Big Cycle. Dalio, who often issues warnings on social media about America’s record $37 trillion national debt, wrote on LinkedIn he wanted to write this book because he sees the U.S. and other countries “headed toward having the equivalent of economic heart attacks.” He said he wanted to explain the mechanics and principles he uses, ever since he learned those key lessons in the early 1980s.

He likens the credit/market system to the human circulatory system, “bringing nutrients to all parts of the body that make up the markets and economy.” If this doesn’t produce enough income to service debt and interest, then “debt service will build up like plaque that squeezes out other spending.”

In a statement, Dalio said one of his principles relates to recognition of big cycles and patterns.

“The same basic big cycles that drive these systems to change have happened thousands of times before for the same reasons,” he said, noting that he described the “Overall Big Debt Cycle” in this book because he believes the world is “on the brink of very big changes.”

It‘s the product of years of audacity, sprinkled with a dose of humility and constant diversification.

A version of this story was originally published on Fortune.com on Aug. 13, 2025.

This story was originally featured on Fortune.com

For generations of American teenagers, turning 16 and getting a driver’s license has been a rite of passage—and a first taste of independence. But that milestone could soon become a thing of the past, according to Andrew Macdonald, president and chief operating officer at Uber. 

“In some future world, maybe not five years, but 15 or 20 years, everyone’s going to be like Harry—nobody’s going to own a car,” Macdonald said in a recent episode of the 20VC podcast with host and entrepreneur Harry Stebbings.

“Nobody’s going to have their driver’s license because you’ll be able to get around. And I think bikes and scooters will be part of that. I think autonomous vehicles will be part of that. I think public transportation will be a big part of that, but I don’t think you need to own a car.”

That future would represent a dramatic shift for a country where car ownership has been synonymous with the American Dream. But with new vehicle prices having soared 30% over the last six years, Macdonald argued that the economics don’t make sense when vehicles spend most of their time sitting unused.

“The individually owned car is the most inefficient asset that anyone owns, and certainly at any level of price point,” Macdonald said. “It sits idle 98% of the day. It’s depreciating. The ongoing operating costs are actually high—even if you’re not driving it, you’re paying for that insurance clip.”

The shift could also prove lucrative for Uber. While the ride-sharing giant abandoned its plans to develop its own fleet of self-driving cars, it has increasingly positioned itself as a platform for autonomous vehicles, partnering with companies including Waymo and Waabi to bring their vehicles onto its network.

Uber’s CEO calls MacDonald an ‘execution machine’—he says these 2 principles drive his success

A Canadian native, Macdonald graduated from the Ivey Business School at Western University in Ontario in 2007 before starting his career as a Bain consultant. He dabbled in entrepreneurship in the early 2010s before joining Uber in 2012. 

Today, Macdonald is the company’s longest-tenured active employee. Stebbings said he recently asked Uber CEO Dara Khosrowshahi, who joined the company in 2017, what makes MacDonald so effective.

“People really like him, but he is an execution machine,” Stebbings recalled Khosrowshahi saying. “And he is very good at driving people.”

MacDonald said there’s ultimately no secret formula for success, but two principles stand out. The first is building trust by making decisions based on what’s best for the company, whether the stakes are big or small.

“You’re not always going to get the decision right,” Macdonald said. “But if people know that you’re filtering on that, then I think that builds followership and trust over time, and then you can move people because if you’re pushing on something, they know it’s because you genuinely think it’s the right thing to do.”

The second is having a deep knowledge of the business. For Macdonald, after 14 years at Uber, he said he knows ride-hailing “better than anyone in the world.”

“Those two things together, I think, are pretty powerful.”

Fortune reached out to Uber for further comment.

AI is already disrupting jobs—and Elon Musk predicts a future where work is optional

MacDonald’s prediction that technology could eventually upend how Americans get around—and even eliminate the need for a driver’s license—isn’t the only eyebrow-raising vision for the future of work and daily life.

The cost of an Uber may matter a lot less if Elon Musk’s predictions come true. The world’s richest man, who has his own ambitious plans for Tesla’s humanoid robots, has argued that saving for retirement could eventually become pointless as AI-driven productivity creates unprecedented abundance.  

“My prediction is that work will be optional. It’ll be like playing sports or a video game or something like that,” Musk said earlier this year at the U.S.-Saudi Investment Forum. “If you want to work, [it’s] the same way you can go to the store and just buy some vegetables, or you can grow vegetables in your backyard. It’s much harder to grow vegetables in your backyard, and some people still do it because they like growing vegetables.”

Khosrowshahi, Uber’s CEO, has offered a similarly ambitious prediction for his own company: a future where human drivers are no longer needed for most trips.

“You can imagine the majority of our trips being fulfilled by robots of some kind,” Uber CEO Dara Khosrowshahi said on The Diary of a CEO podcast earlier this year. “Probably not 10 years from now, but you go 15 to 20 years from now, you’re going to start getting there.”

That future may sound distant, but the disruption is already taking shape as autonomous vehicles roll out in many major cities. Uber is paying drivers to train AI models, while Lyft is paying former drivers to wash autonomous vehicles that have begun replacing some of their rides.

This story was originally featured on Fortune.com

The era of buying crypto on social media has arrived. Crypto payments company MoonPay announced Monday that it launched PayBox, its AI-enabled payment and wallet tool, inside Grok, X’s native chatbot. The service is available globally and lets people carry out a variety of crypto transactions across different blockchain networks.

MoonPay’s Grok launch reflects a broader push to turn AI chatbots from tools that answer questions into tools that can carry out financial tasks. Over the past two years, crypto and payments companies, including Coinbase, Stripe, Visa, and Mastercard have announced initiatives that let users delegate tasks ranging from small online payments and shopping to buying crypto and moving money to AI agents.

In the case of PayBox, users will also be able to use Kamino, a Solana-based crypto lending platform, to earn yield on USDC, the dollar-pegged stablecoin issued by Circle. The application also allows users to carry out non-crypto activities, such as asking Grok to book flights or make restaurant reservations.

“Grok prepares the transaction. The user approves it with a passkey. The money moves,” MoonPay said in a press release. 

Monday’s announcement comes days after MoonPay added a crypto lending feature that lets eligible users manage crypto loans and deposits through ChatGPT or Claude. The integration is limited to users outside the United States, the United Kingdom, the European Union, and Australia. In July, the company introduced a platform for businesses and institutions to trade and move digital assets across more than 200 blockchain networks. Five months earlier, MoonPay also released developer tools that let AI agents independently manage wallets and trade crypto.

The integration of AI payment tools into X is perhaps expected since the platform’s owner, Elon Musk, is a cofounder of PayPal, and has long sought to expand the use of crypto and digital money.

The agent race

So-called agentic commerce, which describes using bots for shopping, has been gaining traction in recent months as more companies build tools to help AI agents search, shop and pay on users’ behalf. 

In May 2025, Coinbase announced x402, a protocol that lets AI agents pay for online services using stablecoins. Stripe followed in December with its Agentic Commerce Suite, which gives businesses tools to sell through AI agents. In May, Circle launched Circle Agent Stack, a set of wallets and payment tools that lets AI agents hold and spend USDC within preset rules.

The period also coincided with a broader push in finance toward agentic commerce. In April 2025, payments network Visa launched Intelligent Commerce, an initiative designed to let AI agents search for products and make purchases using tokenized payment credentials rather than consumers’ card numbers. That same month, Mastercard introduced Agent Pay, a service that allows verified AI agents to make purchases within rules set by customers.

This story was originally featured on Fortune.com

Japan plans to use drones, artificial intelligence and new methods of warfare as it adapts its defense strategies to suit changing times, according to details of an 8.9 trillion yen ($55.6 billion) budget request for the coming fiscal year disclosed by the Defense Ministry on Monday.

The preliminary total is only a small increase from the 8.8 trillion yen ($55 billion) defense budget requested for this fiscal year. It will be much higher eventually, but further details will come later this year after Prime Minister Sanae Takaichi’s government approves the new security and defense strategy aimed at strengthening Japan’s counterstrike capability.

Under its five-year defense strategy set in 2022, Japan has doubled its defense spending to about 2% of its gross domestic product to build up its Self-Defense Forces’ offensive role to counter China’s growing assertiveness. Beijing has criticized Japan’s defense buildup as “new militarism.”

Japan now aims to “drastically strengthen its deterrence and response capability by transforming its defense capability” by adapting to new warfare as in Ukraine’s deployment of drones, the ministry said in a pamphlet about its budget request.

Japan will use attack drones, long-range cruise missiles and interceptors

The ministry says new ways of warfare require domestic development and mass-production of low-cost uncrewed weapons to reduce costs, minimize casualties and build a more cost-efficient air defense network.

The ministry plans to use attack drones in coordination with long-range missiles for more cost-effective counterstrikes. For the fiscal year beginning in April 2027, it is requesting an unspecified amount of funding to acquire unmanned attack drones for launch from above or under water.

Officials say uncrewed weapons could also help Japan compensate for the falling number of military personnel as its population ages and shrinks. Experts say developing fully homemade drones, given that Japan’s commercial market is dominated by imports from China, would be a challenge.

Japan in recent years has fortified its southwestern region with longer range missiles and interceptors.

To further strengthen its counterstrike capability, the ministry is seeking to develop hypersonic missiles that can be launched from submarines.

AI will be used to improve Japan’s defense command

The ministry plans to introduce an integrated AI platform in command and control to improve speed and accuracy in decision-making, including assessing battlefields and choosing targets.

Japan is considering using existing foreign AI technologies, including those of the U.S., while also developing and strengthening domestically developed AI, or combining them, officials said.

The ministry seeks to build a next-generation platform, “AI orchestrator,” which would integrate and use multiple AI systems while ensuring its autonomous selection and operation.

The ministry also plans to build a “Defense Ministry cloud” for highly sensitive data transmission and stable communications networks in an emergency or a disaster.

It did not specify the expected costs for such measures.

Government backing will strengthen the defense industry

Deterrence requires stable and sustainable defense equipment production, so Japan needs to reduce risks of shortages caused by suppliers withdrawing, officials said.

To address the concern, the ministry has requested an unspecified amount to fund a state-owned, contractor-operated system to ensure stable defense equipment production and supplies, with the government acquiring factories and outsourcing manufacturing to private firms.

The budget request also includes funding to support companies’ defense-related research and development or restructuring to strengthen their production and efficiency.

Profits at major suppliers such as Mitsubishi Heavy Industries, Kawasaki Heavy Industries and IHI Corp. have risen as Japan builds up its military and expands arms exports. But the industry lags overall due to a constitutional limit on the use of force for only its own self-defense. Dozens of companies have withdrawn over the past decades due to poor profits, an uncertain business outlook and concerns about harm to their reputations from making arms, especially lethal weapons.

This story was originally featured on Fortune.com

A college endowment is supposed to last forever, supporting students, professors and research in perpetuity.

Instead, many cash-strapped schools are draining their nest eggs just to stay alive.

Faced with multi-million dollar deficits, Hiram College in rural Ohio borrowed from its $56 million endowment. The 1,000-student liberal arts school — which boasts a statue of US President James Garfield, who worked as a janitor there to pay his tuition — eventually pulled more than $47 million from the fund. It even tapped money that donors had explicitly set aside for specific purposes, not for balancing the budget.

Hiram is now working with the state attorney general’s office and is hashing out a repayment plan, a spokesperson said. The school is also notifying its donors about the draw from the endowment, both in writing and through in-person conversations.

David Haney, Hiram’s president from 2020 to 2023, said he was surprised to learn about the loans when he took office. He considers them a risky bet that many smaller schools feel forced to take in an era when US student enrollment has started to decline. Colleges need to focus on cutting expenses instead, he said.

“What a lot of these small colleges do is they think that things are going to turn around,” Haney said. “‘If we just invest in new athletic facilities, everything is going to be fine.’ In most cases, that doesn’t happen. To me, that’s why borrowing from the endowment and taking out debt is not the way to go about it.”

Hiram is far from alone. Nearly 200 private colleges borrowed from restricted endowment funds in 2025, up from about 130 in 2021, according to estimates from higher-ed consulting firm Perspective Data Science. Other colleges have avoided loans but are drawing more from their endowments each year than advisors consider sustainable. Analysts liken both strategies to borrowing from a 401(k) — it may help in the short-term but carries long-term risks, such as downgrading a school’s credit rating. 

For some institutions, “there’s just nowhere else to turn,” said Tracy Filosa with investment firm Cambridge Associates, which works with endowments. “It’s not a rainy day fund — we know that. But in one way, it is there for an institution to get through a rainy day.”

Dipping into the endowment can sometimes work, which is one reason schools keep trying it.

Avila University in Kansas City received court approval in 2023 to loosen the restrictions covering 97 different endowment funds totaling $6.4 million. The school is now on better financial footing, said Andy Jett, its chief operating officer.  The school aims to replenish the endowment as it recovers financially.  Avila has boosted enrollment through recruiting more international pupils, though that strategy has faced a blow from the Trump administration’s efforts to limit student visas. Even so, Jett says dipping into the endowment isn’t a step to be taken lightly.

“I don’t think any school wants to — or should, probably — do this unless they don’t have any other options,” Jett said. “This wasn’t our first choice. This was pretty far down the list of things that we tried.”

Most colleges resorting to this tactic face the same underlying pressure.

The dismal US birth rate has left schools fighting over a dwindling pool of potential students. Larger, more prestigious universities — many with endowments north of $5 billion — still draw far more applicants than they can admit, keeping their finances healthy. But smaller, lesser-known institutions have endured years of declining enrollments and shrinking income, with no obvious solution in sight. Many have already succumbed, shutting their doors for good. 

Indeed, some of the schools that have drawn down their endowments — including Indiana’s Martin University and Notre Dame College in Ohio — ended up closing anyway. 

“When schools take that action, it’s meaningful — it can be a real red flag for those of us on the outside looking in,” said Emily Wadhwani, a higher education analyst for Fitch Ratings. 

College endowments can be complicated, with the money held across hundreds or even thousands of separate funds. 

The cash is essentially divided into two pots: one reserved for specific uses chosen by donors — like English scholarships or medical research — and the other unrestricted, with few conditions on how it’s spent. Hiram, for example, had 325 restricted endowment funds, according to a recent audit. State laws that govern funds given to charitable organizations apply to schools as well, so if a college wants to use restricted donations for other purposes, it generally has to get approval from the state attorney general’s office, according to law firm Taft Stettinius & Hollister LLP.

Schools generally assume their endowments’ value will grow 7% to 8% per year on average over the long-term, according to the American Council on Education. They budget accordingly, trying to draw no more than 5% of the value in a given year.

But as the financial pressure on them increases, many schools with relatively small endowments have been pulling more from their funds, data from the National Association of College and University Business Officers show. Colleges with endowments of $51 million to $100 million have increased their spending rate by more than a third since 2016, reaching 5.5% in 2025 compared to 4.1% in 2016.  

Others have gone higher. Manhattan University, located in the Bronx, drew about 7% of its endowment in fiscal 2024 after facing deficits and withdrew another 7% in fiscal 2025, according to S&P Global Ratings. In response, S&P downgraded the school to one level above junk, or a BBB- credit rating.

“Once you get much over a spending rate of 7%, consistently, you are an at-risk endowment,” said Kristin Reynolds, who advises endowment administrators on their investment strategies at consulting firm NEPC. She was speaking generally, not about a specific college.

Manhattan University’s chief financial officer, James Perrino, said the school’s investment returns were significantly higher than 7%, so the draws did not erode the endowment’s principal. The money, he said, was used to provide additional student scholarships, enhance campus facilities and develop new master’s degree programs. 

Tapping an endowment can help a school navigate tough times, or make a strategic investment in its future.

For 10 years, Webster University in a St. Louis suburb relied on cash infusions from its endowment to balance its budget and meet loan obligations. The school also reclassified some restricted funds to satisfy bond covenants. Now, leaders say they’ve made it through the rough patch. They’ve boosted enrollment, invested in international campuses and removed a “financially distressed” designation from the university’s accreditor.

Webster broke even for the past two years without help from its nest egg. But the stronger footing came at the cost of donor trust, said chief financial officer Bill Donovan. The school’s administration now must repair the relationship. 

“No donor wants to think you’re taking their funds just to cover operating losses,” Donovan said. “The promise we’ve made with the new leadership team is that we’re not going to do that.”

Hiram is hoping for a similar rebound. 

Endowment money has helped Hiram deal with persistent deficits. In a report in April, the college’s auditors noted that some of the borrowed money had come from restricted funds. But the cash infusion hasn’t fixed Hiram’s finances. Auditors wrote that the school faces “substantial doubt” that it can continue as a going concern for another year. 

Hiram is working to turn its finances around. Earlier this year, James Malz, a former JPMorgan Chase & Co. banking executive, was appointed interim president. The school is dropping eight of its least-popular majors, including biochemistry and creative writing.  

It’s also developing a plan to repay the loans, a college spokesperson said in an emailed statement. The school has done so before — in 2022, Hiram repaid over $6 million of loans from its endowment. 

“The College’s financial strategy, including the use of endowment-backed financing, was implemented to help manage ongoing operational pressures while the institution developed and advanced a long-term sustainability plan,” the statement said. The school has now increased its fundraising — with an emphasis on unrestricted gifts — and has no plans to close, according to the spokesperson. 

Founded in 1850, Hiram always admitted women and Black scholars due to the egalitarian ideology of its Disciples of Christ founders. Many of Hiram’s current students are the first in their families to attend college. Federal data shows 57% of first-time students there in 2019 received Pell Grants. 

“We helped a lot of great students,” said Brad Goodner, a professor emeritus of biology and biomedical humanities. 

Closing would hurt not just students and faculty, he said. The college is the center of the town — also named Hiram — about 40 miles southeast of Cleveland. As with many small, rural colleges across the country, the school provides jobs and forms the core of community life.

“The campus is the village,” Goodner said. In a potential closure, “those things would be lost.”

To contact the authors of this story:
Amanda Albright in Overland Park at aalbright4@bloomberg.net
Elizabeth Rembert in New York at erembert@bloomberg.net

This story was originally featured on Fortune.com

Far from New York or Davos, leaders of some of the world’s least populous states are meeting in a placid tropical archipelago with no traffic lights. The population of the summit host, Palau, was due to swell from 18,000 people to 20,000 as delegates from dozens of countries arrived at the Pacific Islands Forum Leaders Meeting.

Ahead of the summit officially opening Monday, the country’s president and other senior officials walked the main road around the island picking up trash. Barbers and tattoo artists prepared to become drivers for delegates and student volunteers to usher dignitaries.

But beneath the distinctively Pacific character of the event, where matters unfold in their own unhurried time and decisions are made by consensus rather than by a vote, simmered a fraught and escalating geopolitical debate about who gets to shape the future of the vast ocean region.

Powerful nations interested in the Pacific have raced in recent years to shore up alliances among its leaders, intensifying a contest for influence between China and the United States, Taiwan, and forum members Australia and New Zealand. The sometimes-conflicting bids to win sway have also drawn dividing lines among the forum’s members, threatening to derail annual summits.

“The Pacific is our home, not a theater for geopolitical competition,” Palau’s President Surangel Whipps Jr. told reporters Sunday, as he urged the bigger countries closely observing this week’s event not to interfere in talks that he hoped would center on problems desperately imperiling small Pacific island nations, such as climate change and struggling economies.

“Let’s not use this as a place that we should be threatening or making people feel uncomfortable,” he added. The reality, however, was unlikely to be simple.

Small island nations have become important allies

Some of the 18 member countries and territories that make up the Pacific Islands Forum have populations in the low thousands or tens of thousands and were once known mostly as picture-perfect holiday destinations. But many also have sprawling and lucrative exclusive economic zones, rich with fisheries and seabed minerals, and their locations across the wide Pacific Ocean have become strategically important points on the map as global tensions rise.

While Forum countries often reference a “Pacific family,” members come from diverse cultures and political systems and their remote locations mean few opportunities for leaders to speak in person. The annual leaders’ summit was once a relaxed and casual affair, where officials from countries with little power in the global order sought to amplify their voices by reaching unified positions, particularly in urging bigger countries to curb greenhouse gas emissions as rising seas began to threaten low-lying atoll nations.

As major powers over the past decade have escalated their efforts to secure access to Pacific waters and ports by inking security deals and offering funding, immigration access to their countries and other sweeteners, the leaders’ ability to speak with one voice at their annual summit has come under pressure.

“We see examples of this where that consensus-based model, which is referred to as the Pacific Way, can sometimes lead to a bit of paralysis in taking action,” said Oliver Nobetau, Pacific Islands program director at the Lowy Institute, an Australian think tank.

Pacific nations face worsening climate and fuel shocks

In a recent episode, the Forum’s foreign ministers failed to reach consensus on a statement criticizing Beijing’s launch of a ballistic missile into the South Pacific Ocean in July. While most member states strongly opposed China’s actions, the refusal of two to endorse a statement saying so meant none was issued.

“At this point in time when Pacific Islands are facing a lot of acute security issues, decision-making is paramount for this one,” Nobetau said. “We’ll be looking to see if President Whipps can sort of steer the forum to take tangible action.”

Among the issues vexing Pacific island nations are the effects of the latest El Nino weather chaos, which has brought droughts and floods across the region. The volatile weather threatens food security and access to clean water, already vulnerable in island nations dependent on imported goods and funded largely by foreign aid.

Fuel volatility from the Iran war has also prompted some Pacific countries experiencing severe economic shocks to declare national emergencies in a region overwhelmingly reliant on imported diesel and gasoline, arriving via a fragile supply chain.

Whipps this week suggested that the wealthy countries seeking involvement in the region could open their wallets to finance the widespread establishment of solar power capabilities across the region.

“What are they doing together with us to partner to get a 100% renewable Pacific?” Whipps said.

Conflict over members’ alliances is expected

However, some of the powers at loggerheads over the region were always going to see this year’s summit as a showdown. Palau is one of just 12 countries worldwide to recognize Taiwan, incurring the ire of Beijing, which sees Taiwan as part of its territory.

It also sits at the nexus of other great power struggles. Located in the western Pacific Ocean, next to the Philippines, Palau draws from the Asian tourism market and has closer ties to the United States than much of the Forum’s members. A Compact of Free Association allows defense and security access for the U.S. in Palau’s territory in exchange for tens of millions of dollars in U.S. funding each year.

Palau’s turn as host also comes as the summit’s leaders try once more to draw up a structure governing how other countries are permitted involvement in its meetings. Countries approved to engage in the Pacific Islands Forum, including the United States and China, are known as dialogue partners, while Taiwan holds separate status as a development partner.

Beijing has ramped up pressure on forum hosts to exclude Taipei’s envoys from the annual summit, a dispute that grew so fractious ahead of the 2025 event in Solomon Islands that all countries except member nations were entirely barred from attending. Palau’s warm diplomatic relations with Taipei have this year drawn the attendance of Taiwan’s Foreign Minister Lin Chia-lung and provoked a warning from Beijing that China would oppose Taiwanese officials’ participation.

That in turn has prompted exasperation from some officials and Pacific-watchers at the prospect of another summit overshadowed by rivalry between non-member countries. Early arrivals over the weekend were roiled by the news that one leader, Solomon Islands Prime Minister Matthew Wale — a newly elected leader more friendly to Australia than the more Beijing-aligned leader before him — had returned home just hours after arriving because an opponent had mounted a leadership challenge.

It wasn’t immediately clear how much his forced departure, and the absence of others such as Kiribati’s President Taneti Maamau, whose country’s China ties have warmed during his tenure, could be attributed to unease about the summit’s geopolitics.

“You’d be defying gravity and logic if you said it wasn’t a big problem,” New Zealand Foreign Minister Winston Peters told reporters in Koror on Sunday, referring to bids for foreign interference in the event.

The leaders’ non-attendances were “seriously disappointing” and did not help, Peters added.

“But in the main the delegations are of the highest standard,” he said. “So let’s see how we go.”

This story was originally featured on Fortune.com

Workers shoveled thick mud from a flood-ravaged power station in central Nepal, trying to revive a key node that supplies electricity to more than 20,000 people. But as they cleared debris from equipment buried by last week’s catastrophic floods, there was hope the plant would not simply be rebuilt in the same place, beside the river that destroyed it.

In Nepal, up to 900 workers are unaccounted for from a dozen hydropower projects, according to the Independent Power Producers’ Association, Nepal, with hundreds believed trapped in tunnels. In Nepal and China combined, over 900 people have died and 4,700 are missing.

The loss of hydropower infrastructure is estimated to have taken 700 megawatts offline — about 10% of the country’s power capacity. Once the immediate aftermath of the disaster is dealt with, the question of how to restore power, roads, bridges, schools and health facilities quickly without rebuilding the same vulnerabilities should be on top of Nepalese authorities’ minds. disaster experts said.

“Authorities have to take a hard look in the mirror now and ask how they will be rebuilding in these places. They will also have to take a close look at other projects in Nepal that have not been built yet, whether they’ll be approved after this event,” said Jakob Steiner, a geoscientist at the University of Graz in Austria, who is currently based in Dhaka, Bangladesh.

Rebuilding for resilience, not just replacement

Hydropower projects that were operating or under construction were fully or partly destroyed across Rasuwa, Nuwakot and Dhading, among the worst-hit districts.

Many of the workers had taken shelter in the tunnels but are stuck, waiting to be rescued, according to officials.

“The tunnels are built to access hydropower infrastructure and to divert water,” Steiner said, describing the tunnels as big enough to drive trucks through.

Apart from the hydropower projects, tens of kilometers (miles) of roads, multiple bridges and essential buildings like schools and hospitals have been destroyed.

Ramraj Narasimhan, a senior director with the Coalition for Disaster Resilient Infrastructure, said rebuilding should focus on continuity of services rather than trying to make every structure indestructible.

“If it gets damaged, let it get damaged, but can we build in redundancy in the system?” he said.

That can mean choosing a better location for construction, alternate routes, backup power, improved early warning and financial tools such as insurance to speed recovery, Narasimhan said. Risk assessments should examine where a river bends, where slopes are unstable and where a single failure could cut off several communities.

Narasimhan’s organization estimates that $124 billion worth of Nepal’s infrastructure is exposed to climate-driven disasters, creating the potential for hundreds of millions of dollars in losses each year. A report by the coalition earlier this year found that Nepal’s average annual losses are nearly $760 million from various disasters, while the country’s national disaster funds is only a fraction of this amount.

Climate risks grow for Himalayan hydropower

Climate scientists said the Himalayan region — which is warming faster than many other parts of the world — is particularly at risk from increasing global warming as a result of human-caused climate change.

Climate experts said the flooding adds to evidence that infrastructure planning in the Himalayas is being forced to account for disasters that are increasing as the climate warms. Glaciers across the Hindu Kush Himalaya are losing ice at an accelerating rate, while permafrost degradation, unstable slopes, glacial lakes and increasingly erratic rainfall are raising the risks of floods, landslides and debris flows, according to reports by Kathmandu-based climate research group International Center for Integrated Mountain Development.

The stakes are particularly high for Nepal because hydropower is both a foundation of its electricity grid and an important source of economic growth and export revenue. Much of the country’s potential is run-of-river hydropower, Narasimhan said, meaning facilities are closely tied to steep river corridors where land is limited and hazards can cascade quickly.

Steiner said smaller floods can sometimes be managed with diversions that keep debris and high flows away from expensive infrastructure and tunnels. But for events as large as the latest flood, he said, physical defenses may be overwhelmed. In those cases, early warning becomes the critical protection for people.

He said the flood was detected at the China-Nepal border after moving rapidly from its source, leaving little warning time upstream. But there was still time as the flood traveled downstream toward hydropower sites, raising questions about whether warnings could have reached workers sooner.

The disaster is also likely to intensify scrutiny of where future projects are approved. Himanshu Thakkar, coordinator of the South Asia Network on Dams, Rivers and People, said environmental, social and disaster-risk assessments need to reflect the specific landscape and past hazards of each valley, backed by land-use rules and stronger accountability.

Hydropower projects can themselves become “force multipliers” when large structures, debris and altered river channels add to downstream damage, Thakkar said. He argued that Nepal should also consider more decentralized solar power as part of a broader energy mix.

Thakkar said the question now is not only how fast Nepal can rebuild, but what it chooses to put back in the path of the next flood.

___

Sibi Arasu reported from Bengaluru and can be followed on X at @sibi123. Reach him at sarasu@ap.org.

___

The Associated Press’ climate and environmental coverage receives financial support from multiple private foundations. AP is solely responsible for all content. Find AP’s standards for working with philanthropies, a list of supporters and funded coverage areas at AP.org.

This story was originally featured on Fortune.com

Chinese markets are booming with new public stock offerings, energized by the craze for artificial intelligence and other advanced technology and a growing preference to list shares in Hong Kong and Shanghai.

In the latest big stock listing, shares in China-founded e-commerce and fast fashion giant Shein are due to debut Tuesday in Hong Kong in a blockbuster initial public offering raising $1.7 billion, in one of the city’s biggest new share sales this year.

In July, CXMT, China’s largest memory chipmaker, raised more than $8.6 billion in Shanghai in the second-largest IPO for its Nasdaq-style STAR market, mainland China’s second-largest IPO. Its shares jumped 466% on the first day of trading.

Unitree, one of China’s leading humanoid robot makers, also made its listing debut in Shanghai in August. Shares rose 460% on the first day of trading.

“The current IPO boom is powered by investor appetite for AI and robotics,” said Ruiying Zhao, a senior research analyst at S&P Global Market Intelligence. Trading in Shanghai’s stock market, for one, is heavily driven by retail investors.

AI driving Chinese IPO boom

CXMT’s IPO in Shanghai “placed China in a strategically significant position in tech manufacturing related to AI,” said Perris Lee, head of APAC equity capital markets for ION Analytics. “It’s also a testament to China’s tech self-sufficiency ambitions.”

Founded in China in 2016, the company’s revenue surged more than 700% year-on-year to 50.8 billion yuan (about $7.5 billion) in the first three months of 2026 on a spike in demand for computer chips needed for AI.

IPO proceeds in Hong Kong and Shanghai so far this year have already surpassed the funding raised last year, according to the financial data platform LSEG.

It says IPOs and secondary listing activities on the Hong Kong and Shanghai exchanges raised a total of over $54 billion from so far in 2026, surpassing last year’s total of more than $46 billion.

Combined Hong Kong and Shanghai proceeds so far this year accounted for roughly 21% globally, ranking them only behind only the Nasdaq’s roughly 55% global share, LSEG said. There, the mega $75 billion IPO by SpaceX in June made the U.S. exchange the world’s biggest IPO market this year.

Since China limits foreign purchases on mainland exchanges, many Chinese companies do parallel listings in Hong Kong to help raise international capital.

Fewer big Chinese companies listing overseas

Stricter U.S. and Chinese regulatory scrutiny in recent years of big Chinese companies listing in U.S. markets, especially those in strategically important sectors like advanced technologies, has led some Chinese companies to stick closer to home.

Listing overseas typically takes more time compared with doing IPOs in China, said Howie Farn, a capital markets partner at the law firm Freshfields.

In Hong Kong, recent public stock listings of Apple-supplier Luxshare Precision Industry, and Zhongji Innolight, which makes optical transceivers used in data centers, were among this year’s largest deals and were also a reflection of investor demand for advanced technologies.

More companies are looking to hold their IPOs in Hong Kong or Shanghai, like robotics firms AGIBOT and Deep Robotics.

Shein also explored the possibility of listings in the U.S. and London before opting for Hong Kong.

Investors are wary of a possible AI bubble in China, too

After massive oversubscriptions and huge gains in their share debuts, some companies have seen their market value shrink.

Chinese robot maker Unitree’s share price had fallen more than 40% as of Friday from its peak share price on the day of its trading debut.

“The critical question remains: is the AI sentiment enough?” said Zhao from S&P, as the similar question that raised worries among investors in the U.S. also now also applies to China. “For a durable market cycle, investors will demand sustainable revenue, visible profit margins, and realistic valuations.

The global AI frenzy also has also drawn attention away from companies like Shein. “The AI investment cycle is absorbing much of the risk appetite that would have otherwise flowed to a company like Shein,” said Jacob Cooke, CEO of WPIC Marketing + Technologies.

Shein’s IPO puts the company’s value at around $27 billion, a fraction of its peak valuation a few years ago, though that is partly due to U.S. and EU moves to restrict de minimus tax-exemptions for imports of small packages.

This story was originally featured on Fortune.com

Good morning. Fed Chairman Kevin Warsh used his Jackson Hole speech on Friday to focus on inflation. But he also pointed to an unusual metric that could offer clues about whether AI is delivering the productivity gains businesses are betting on: token prices.

Tokens are the units used to measure the data AI models process, and many AI companies charge customers based on token consumption.

Warsh called AI a potential “new factor of production,” then asked whether customers will pay a premium for tokens from the most advanced models even as prices for older models fall toward marginal cost.

That doesn’t mean token prices are becoming a new Fed indicator. Instead, they could reveal how AI economics are evolving.

Gregory Daco, chief economist at EY Parthenon, told me that Warsh appears to see token prices as a window into the evolving AI market, offering clues about competition among providers, differences in model quality, pricing strategies, and computing costs.

But interpreting those price signals isn’t straightforward. Falling token prices can tell two very different stories.

If AI models become more capable while getting cheaper, businesses could generate more output for every dollar they spend, a sign of genuine productivity gains. But if models become increasingly interchangeable, providers could be forced to compete on price. That could signal commoditization and raise questions about whether the enormous capital flowing into AI will generate strong returns.

“Pricing power at the frontier—not usage growth—is becoming the real scoreboard for whether AI is creating value or just consuming capital,” Luke Lango, a technology analyst and publisher of Innovation Investor, told CFO Daily.

Token use has become a top-of-mind cost concern for CFOs. For companies spending on AI, token prices are an input cost. But lower cost doesn’t necessarily mean higher returns.

“For CFOs, the more relevant question is whether AI adoption is generating measurable productivity gains, improving margins, or creating new revenue opportunities,” Daco said.

That gets to the larger issue behind Warsh’s comments. The economics of AI will ultimately hinge on how much value it creates and who captures it.

Sheryl Estrada
Sheryl.Estrada@fortune.com

This story was originally featured on Fortune.com

Good morning!

After an employee’s grandmother died, they texted their manager that they’d be taking bereavement leave for the funeral. The response became a case study in how not to manage grief.

The manager pushed back, questioning whether the grandmother counted as “immediate family” and calling the timing “terrible” with a big upcoming pitch meeting. The exchange reached millions when George Stern, a former McKinsey consultant, read the texts deadpan to camera, becoming one of his most viral Instagram and TikTok videos.

The outrage from his followers didn’t surprise Stern, who now runs a firm that grows and sells small businesses when he’s not creating content. But the volume of recognition did. Stern says he’s since received roughly 100 more bereavement stories from people who had been treated similarly at work.

“Corporations are losing trust, like every other institution in the country right now,” Stern said. “They’re out of touch with the experience of their employees on the ground.” 

Stern has built an audience, he says, by putting that disconnect on display—and he thinks HR leaders should pay attention. 

Stern didn’t set out to become the internet’s bad-boss archivist. He started posting management content on LinkedIn two years ago. When employees began sending him screenshots of their own manager horror stories, he switched to video, offering advice on navigating workplace conflict. He now gets up to 20 new submissions a day.

Typically, Stern’s advice is to take workplace issues to HR. But increasingly, followers respond with some version of: “HR is not your friend.” He doesn’t entirely agree, but he sees the skepticism as a sign that HR needs to be more transparent about how it handles workplace complaints. For instance, he’s seen employees lose trust in HR when they raise concerns but lack the documentation needed for HR to investigate. In such scenarios, HR leaders should make clear what employees need to document for a complaint to be properly investigated, says Stern.

He offers professionals the same advice, urging them to document workplace interactions in writing, record conversations when permitted, and send written recaps of meetings that weren’t recorded.

But, he adds, “I should not be the messenger of that to these employees. It should be their HR leader.”

Editor’s note: This newsletter will be off for Labor Day next week. Catch us back in your inboxes on Sept. 14.

Kristin Stoller
Editorial Director, Fortune Live Media
kristin.stoller@fortune.com

This story was originally featured on Fortune.com

As of 8:30 a.m. Eastern Time today, oil sold for $93.03 per barrel (using Brent as the benchmark, which we’ll get into momentarily). That’s $1.06 higher than yesterday morning and approximately a $25 rise over the past year.

Will oil prices go up?

It’s impossible to predict the future of oil prices. Several factors determine the movement of oil, but it ultimately boils down to supply and demand. Again, when threats of economic downturn, war, etc. are high, the oil trajectory can turn rapidly.

How oil prices translate to gas pump prices

When you pay for gas at the pump, you’re paying for more than just the crude oil itself; you’re also springing for links along the chain, such as the refineries and wholesalers—not to mention taxes and local gas station markups.

Still, the crude oil aspect affects the final price most dramatically, as it typically accounts for more than half the price per gallon. When oil prices spike, so do gas prices. And frustratingly, when oil prices drop, gas prices tend to take their time drifting down to the lower price (sometimes referred to as “rockets and feathers”).

The role of the U.S. Strategic Petroleum Reserve

In case of emergency, the U.S. has a store of crude oil known as the Strategic Petroleum Reserve. Its primary purpose is energy security in case of disaster (think sanctions, severe storm damage, even war). But it can also go a long way toward softening crippling price hikes during supply shocks.

It’s not a long-term answer—more of an immediate relief to assist the consumer and keep critical parts of the economy running, like key industries, emergency services, public transportation, etc.

How oil and natural gas prices are linked

Oil and natural gas are both major energy fuels. A big change in oil prices can affect natural gas by extension. For example, if oil prices increase, some industries may swap natural gas for some segments of their operations where possible—which increases demand for natural gas.

Historical performance of oil

When examining oil’s performance, there are generally two major benchmarks:

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

Between the two, Brent better represents global oil performance because it prices much of the world’s traded crude. And, it’s often the best way to track historical oil performance. In fact, even the U.S. Energy Information Administration now uses Brent as its primary reference in its Annual Energy Outlook.

Looking at the Brent benchmark across several decades, oil has been anything but steady. It’s seen spikes due to factors such as wars and supply cuts, and it’s also seen crashes from global recessions and an oversupply (called a “glut”). For example:

• The early 1970s brought the first big oil shock when the Middle East cut exports and imposed an embargo on the U.S. and others during the Yom Kippur War.
• Prices dropped in the mid-1980s for reasons such as lower demand and more non-OPEC oil producers entering the industry.
• Prices spiked again in 2008 with increased global demand, but it soon plummeted alongside the global financial crisis.
• During the 2020 COVID lockdown, oil demand collapsed like never before—bringing prices below $20 per barrel.

All to say, oil’s historical performance has been anything but smooth. Again, 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:

• Venezuelans bash their government’s deal handing over vast oil reserves to the U.S.
• The mayor of this desert city likens the U.S. West’s water crisis to planning for a meteor.
• The federal government will control a company with the world’s 2nd largest reserves

Frequently asked questions

This story was originally featured on Fortune.com

In every room, at every table, I hear predictions about what AI means for the workforce and the future of work. The talk is mostly about jobs: which roles are changing, which ones are growing, and which ones may be replaced over time.

But what if we’re missing the crux of the conversation?

AI isn’t just changing jobs; it’s changing the nature of work itself. It is shifting how people will conceptualize work for – quite possibly – the rest of human history. This is a defining moment, and we must respond with urgency. Business leaders need complete clarity on how AI is affecting employees and their work at the most granular levels so they can make critical decisions on hiring, organizational structure, technology investment, and workforce management.

Achieving that clarity requires access to the right data and information – and the leaders who understand this are the ones who will come out stronger.

What the Data Tells Us

Broadly speaking, we don’t see AI eliminating jobs at scale. What we do see – what our data shows – is that the world of work is evolving.

At ADP, we process payroll for one in six U.S. workers and serve over 1.1 million businesses globally, paying 42 million workers worldwide. From this unique vantage point, we observe the labor market as broadly stable. When we look more deeply, we see that AI is reshaping work at the task level. It is changing the workforce, not shrinking it. And while AI’s impact is certainly real, it varies significantly by sector, company size, career stage, and geography. Our work with the Stanford Digital Economy Lab provides a real-time view of AI’s impact on occupations through the lens of ADP’s payroll data. Recent findings show that early-career workers – ages 22 to 25 – in occupations most vulnerable to AI automation, such as software development and customer service, have experienced employment declines. However, employment of more experienced workers in those same occupations, and for workers of all ages in roles less exposed to AI like home healthcare aides, have remained stable or continued to grow.

But this is just scratching the surface.

As we dig deeper, we’re able to see how AI is actually changing individual tasks within jobs. We call this “the great job unbundling.” AI is reshaping work at the task level, creating new job categories and transforming others.

Through the adoption of AI, the wage premium of specific tasks is shifting in real time. This has profound implications for how business leaders think about workforce planning, skills development, and the employer-employee relationship.

AI is also raising the floor on productivity expectations. Employees are increasingly expected to use AI tools as a baseline – not a differentiator – which changes how performance is measured and managed. We see a critical split in this dynamic: entry-level employment has declined where AI is applied to automate work, but employment has grown in occupations where AI is augmentative – where it supports and enhances human decision-making, freeing employees to focus on higher-value, creative, and more strategic work.

Given these nuances, the stakes of getting HCM right have never been higher. Importantly, it is clear this is a moment of change that demands a rigorous, data-driven approach to get it right.

The Shifts that Actually Matter

The workforce is undergoing three key shifts that have direct implications for how leaders should be managing their workers right now.

Investment leads to engagement; 53% of workers say they are fully engaged when they strongly agreed their employer was investing in them – compared to just 12% when they don’t feel that same level of investment. Employees also expect their employers to provide the AI upskilling they need to thrive.

AI increases the value of human judgment. As more “checklist” work is delegated to AI, workers are transitioning toward longer-term, strategic projects where human judgment matters most. The old model of how to measure productivity – task completion, speed, efficiency – is giving way to something harder to quantify but more profound: judgment, creativity, and long-term impact.

AI users report greater productivity and performance. Workers who use AI daily are more than twice as likely to be fully engaged at work – 30% versus 14% of non-users – and half as likely to feel overloaded or stressed: 11% versus 23%. Early AI adopters tend to be top performers. Frequent AI users show higher engagement and motivation, both of which are strong retention drivers.

From Understanding to Action

Leaders who will come out of this moment stronger are not asking what AI will replace; they are asking how to help their people do more with it. AI is a teammate. It takes on the routine work – and what is left are the judgment calls, the hard decisions, the moments that only a real expert can navigate. That is where people become more valuable, not less – and leaders should see their workforce as more than inputs to be optimized. Our people are sources of reasoning, creativity, institutional knowledge, and logic that technology can’t replicate.

Businesses can help workers embrace their evolving identities – encourage them to examine which tasks make up their roles and see new possibilities for growth and opportunity. Workers who understand how their role is changing are better positioned to grow with it.

The fact is: AI is evolving the workforce in ways that are beautifully and fundamentally human. The organizations that have a clear understanding of what the data tells us – that the workforce is shifting, not shrinking – will be the ones that thrive in the future of work.

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

Nearly two years after Elliott Hill returned to Nike as CEO, his tenure raises a question that eventually confronts almost every leader handed a struggling business. How much time should someone get to prove that a turnaround is working?

Executives brought in to revive a company, division, or function often inherit problems that accumulated over years while facing pressure to show improvement within quarters. Hill inherited a Nike in 2024 struggling with weakened retailer relationships, excess inventory, slowing innovation, and declining cultural relevance. He has restored wholesale growth and strengthened performance running. Yet Nike Direct and digital sales have remained weak, China continues to struggle, and the stock has lost the enthusiasm that greeted his appointment.

Research on corporate transformations suggests that expecting an immediate recovery is unrealistic. McKinsey offers some indication of how slowly the economics of a transformation can emerge. In one survey, respondents reported that roughly half of a transformation’s value was realized in the first 18 months, with the remainder coming later.

Spencer Stuart’s research offers a more detailed timeline for judging a CEO’s progress. The firm describes the first year as a launch period, when new CEOs face a steep learning curve, tackle inherited problems and make decisions that can shape their tenure. By then, a leader should have a credible diagnosis, the right team and clear strategic priorities. The second year becomes a period of calibration, when boards and other stakeholders can look for movement in measures such as customer retention, product momentum, market share and operating performance. By around year three, those early decisions should increasingly translate into stronger revenue, margins and returns.

Hill is now far enough into the job to assess whether his early decisions are producing results. Simeon Siegel, senior managing director at Guggenheim Partners, argues that North America provides one such test. Nike’s largest region was among the first parts of the business to struggle, but it has returned to low single-digit growth.

“When people are saying, ‘Okay, he’s failing,’ we have to say, ‘Well, he actually got his largest section to grow 3%,’” he recently told me. The question, he argues, is whether that improvement indicates Nike has found an approach that can eventually work in other regions facing similar problems. “Maybe they do know what they’re doing,” he says. “And therefore, it’s a matter of time.”

A turnaround does not have to be complete for a leader to make the case for more time. There should, however, be evidence that the decisions made early in the process are producing results. For Hill, North America provides some of that evidence. Whether Nike can replicate that progress elsewhere will be the next test.

Ruth Umoh
ruth.umoh@fortune.com

Editor’s note: This newsletter will be off for Labor Day on Monday, Sept. 7. We will return with our next edition.

This story was originally featured on Fortune.com

Martha Shedden has spent 15 years trying to talk Americans out of panicking. Lately, she says, it’s a losing battle.

She co-founded the National Association of Registered Social Security Analysts (NARSA), the organization that trains and certifies financial professionals as Registered Social Security Analysts, or RSSAs. In an interview with Fortune, she described a pattern she’s watching play out across the country: retirees and near-retirees, spooked by headlines about the program’s finances, are claiming benefits at 62 — the earliest possible age — even when waiting would leave them better off.

“They hear their benefits might be cut 22%, and they’re thinking, ‘I need my money now,’” Shedden said. “That was just really shocking. I mean, I knew people were doing that, but that [number] was very, very surprising.”

The behavior shows up starkly in a survey NARSA conducted of 189 RSSAs in August, which asked advisors what they’re hearing directly from clients. Nearly three-quarters — 73.5% — said their clients want to claim early specifically because they fear future benefit cuts. Almost 59% said clients simply doubt Congress will act to fix the program’s finances. And when advisors were asked to describe the dominant mindset among clients weighing when to claim, 62.4% said people feel “overwhelmed by conflicting advice.”

The 22% problem

Social Security’s retirement trust fund is on track to run dry in the fourth quarter of 2032 — one quarter sooner than last year’s estimate — according to the program’s 2026 Trustees Report. After that point, the law requires an automatic, across-the-board benefit cut because the trust fund can no longer make up the gap between what workers pay in and what retirees are owed. Absent congressional action, the cut would be roughly 22%.

The Committee for a Responsible Federal Budget has translated that into household terms: a typical dual-income couple retiring right after insolvency would lose an estimated $16,900 a year in benefits. A single-earner couple would lose about $12,700 annually, and higher-income couples could see cuts as steep as $22,300 a year.

Lawmakers are aware of the clock. On August 5, the Senate Finance Committee held a hearing — “Exploring Process Approaches for Addressing Social Security Solvency” — with testimony from the Committee for a Responsible Federal Budget, the Mercatus Center, AARP and the National Academy of Social Insurance. The session grew heated, with Democrats accusing Republicans of maneuvering toward benefit cuts through a fast-tracked process, and Sen. Bill Cassidy of Louisiana visibly frustrated by the gridlock.

That combination — a hard deadline six years out and a Congress that can’t agree on next steps — is precisely what Shedden says is pushing clients toward the exits early, even though claiming at 62 locks in a permanently smaller monthly check for life.

“It’s a very emotional decision,” she said of the rush to claim early. “It comes down to the education and the knowledge so that they’re able to make their best claiming decisions.”

Beyond the claiming age

The survey also surfaced a broader knowledge gap. Fifty-eight percent of advisors said clients don’t realize ex-spousal or ex-survivor benefits may be available to them after a divorce, and a similar share said clients simply confuse spousal and survivor benefits altogether.

On taxation, only about a third of clients understand how other income affects the taxability of their Social Security check, and nearly half are blindsided to learn that Medicare premiums can quietly shrink their monthly payment. Roughly two-thirds of advisors said income-related monthly adjustment amount deductions were the top reason clients were surprised by a lower-than-expected payment.

These knowledge gaps are what NARSA exists to close. The credentialing process took a step up this summer: as of late July, the RSSA final certification exam is now administered and proctored by the College for Financial Planning, a Kaplan company, using live online proctoring. Shedden called it a “big step” for an organization she says has grown steadily over the past two years as more financial, tax and insurance professionals seek to specialize in Social Security guidance.

Don’t blame the boomers?

Shedden, who describes herself as squarely in the middle of the baby boom generation, pushed back hard against the increasingly common argument that her cohort is responsible for the program’s shortfall — both by its sheer size and by allegedly hoarding wealth and political power while presiding over the system’s decline.

“I don’t feel that it’s particularly our generation, the baby boomers,” she said. “Yes, we are a huge bubble, but there are so many other socioeconomic factors.”

The numbers back up at least part of her case. When Social Security began paying benefits, there were roughly 40 workers for every retiree collecting them, according to Mercatus Center data. That ratio has collapsed steadily since — not because boomers didn’t pay in, but because Americans are having far fewer children. The average number of children born to a woman was 3.6 around the end of the baby boom in 1960; by 2024, it had fallen to 1.6, according to the Peterson Foundation. The ratio of workers to beneficiaries, which stood at 8.8-to-1 in 1955, had dropped to 3-to-1 by 2025 — and continues to fall.

The data critics cite point the other direction. The CRFB calculated earlier this month that baby boomers are on track to collect roughly 265% of what they paid into the program over their working lives — a return ratio that reflects both the program’s structure and decades of benefit expansions that earlier generations supported and current workers are now financing. “It has always been a pay-as-you-go system,” Shedden countered. “So, that’s always been the program from the beginning. The problem is the longer longevity, the longevity of individuals who are in retirement.”

She also named a second culprit rarely discussed in the generational-blame framing: the disappearance of employer pensions. Roughly 40% of private-sector workers had access to a defined-benefit pension in the 1970s, she said; today it’s closer to 15%, leaving 401(k)s and IRAs — vehicles retirees must manage themselves — to fill a gap they were never fully designed to cover.

“All of us should be saving as much as we can,” Shedden said. “Social Security was never meant to be a 100% replacement of all our income. The critical thing is for individuals to plan, not to panic.”

An optimist, for now

Despite the doom-laden headlines, Shedden insists she remains confident Congress can still act. She pointed to the array of available fixes — adjusting the taxable-earnings cap, changing the benefit formula, altering the retirement age — arguing that no single lever needs to move dramatically if lawmakers combine several modest changes. She also rejected proposals floated inside the Trump administration to eventually replace Social Security with a different savings vehicle, sometimes described as “Trump accounts.” A full replacement of Social Security would be “a huge, huge mistake” that would trigger public backlash, she said. “There would be an uproar.”

Still, she was careful to separate her optimism about the program’s survival from complacency about the choices facing Gen X and millennials now approaching retirement.”They should be informed and concerned,” she said, “and be planning for it.”

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

Target’s annual shareholder meeting in June was a referendum on the company’s leadership, and the story is bleak: nearly 13% of shareholders opposed the reelection of Executive Chair and former CEO Brian Cornell. Combined with nearly 40% support for a shareholder proposal calling for an independent Board Chair, this level of opposition makes clear that some shareholders are dissatisfied with the decision to retain Cornell on the Board after he stepped down as CEO early this year. 

At first glance Cornell’s reelection margin may look comfortable, but votes like these typically show approval levels of 90% or higher. The average support for S&P 500 directors in the 2026 proxy season was 96.6%. Compare that to Cornell’s fall to 87.2%, and it signals that shareholders like us have lost faith in Cornell’s leadership and demand a change to Target’s management.

Cornell’s fortunes have dropped precipitously. For nearly a decade, his nay votes never once rose above 6.3%. In 2025, after three straight years of sales declines, the opposition to Cornell rose slightly higher. But then he stepped down as CEO in February, only for the Board to retain him as Executive Chair. At this year’s AGM, opposition increased threefold from a decade ago, from 4.2% in 2016 to 12.8% in June.  

Those of us who follow the company understand how we got here: a steady erosion of Target’s brand from years of poor management decisions, including a series of operational missteps that have repeatedly put it at odds with the public. 

Once an industry leader for championing inclusion, in recent years Target has caved to political pressure — slashing its Pride merchandise collection in 2024, rolling back DEI initiatives in 2025, and most recently its muted response to ICE after federal agents shot and killed Renee Good and Alex Pretti and detained two of its employees in the company’s home city of Minneapolis.

The backlash has been fierce. Black, Latino, LGBTQ+, and progressive shoppers — some of Target’s core customer base — have mounted boycotts and national protests. Twin Cities Pride threw Target out of its hometown parade after 18 years as a sponsor. The daughters of Target co-founder Bruce Dayton called the company’s retreat on inclusion “a betrayal.”

In recent years, Cornell fundamentally misread what set Target apart from other big box retailers, eroding years of hard-earned good will by moving away from the qualities customers once trusted: a genuinely welcoming environment for all, a sense that the company balanced profit with people, and a brand identity that reflected the values many shoppers believed it shared with them.

Along with all this are the everyday consumer experience concerns: in-store standards have been slipping for years. Customers report messier aisles, out-of-stock products, longer checkout lines, and fewer employees available to help — all of this leads to families’ shopping trips taking longer, shoppers not finding what they need, and consumers who once looked at “Tarjay” as a fancy shopping destination no longer feeling any of the whimsy they once felt when walking through stores. 

For the first time since 2001, Target fell off Fortune Magazine’s World’s Most Admired Companies All-Star list this year. And nearly half of Target’s workforce say they have no faith in the retailer’s future. But instead of investing in stronger operational support, employee retention, or product supply chains, Cornell engaged in years of stock buybacks that failed to generate shareholder value, while neglecting the company’s serious problems. Recently, the company has stepped back from repurchasing shares and increased investment, but it is unclear if this is a temporary change or represents a recognition that a sustainable turnaround requires ongoing improvements to the in-store experience.

Inflation and tariff pressures have intensified these challenges. Yet Target’s largest rivals have demonstrated that those headwinds do not fully explain its struggles: Walmart continued to attract more in-store shoppers even as e-commerce expanded, while Costco has been winning on price without sacrificing its durable profit margins. Conversely, during Cornell’s tenure as CEO, Target’s foot traffic in U.S. stores decreased significantly from 2022 to 2025, and net sales shrank year-over-year for seven of the last twelve fiscal quarters. 

Since Fiddelke took the reins, Target’s turnaround is showing early signs of taking hold: traffic and comparable sales are rising, digital growth is strong, and management has lifted its underlying outlook. Still, the durability and quality of the earnings recovery remain unproven, particularly in apparel and home and after removing a large, one-time tariff refund. 

That makes its latest cultural misstep especially costly. Just as Target appears to be regaining traction with the customers it lost from its past, self-inflicted, reputational setbacks, it was forced to pull and apologize for a children’s Halloween costume that drew viral comparisons to blackface and minstrel imagery. Target conceded that the product was offensive and “should never have been part of our assortment.”

The Board’s decision to retain Cornell as Executive Chairman, rather than appoint an independent chair, undermines its claim to a genuine management reset—particularly given that new CEO Michael Fiddelke is a 20+ year Target insider and former COO. We at SOC Investment Group haven’t been shy about our own point of view: retaining Cornell as Executive Chair and Special Advisor preserves the influence of the executive most responsible for Target’s prolonged underperformance, undermining Fiddelke’s turnaround efforts. 

It’s time for Board members to prove they understand the gravity of the issues the company is facing. Greater oversight, stronger transparency, and a better governance structure are no longer suggestions — they’re necessary next steps to rebuild accountability and enable sustainable, long-term growth at the company.

Target’s annual meeting results should send a clear message to the Board that investors are hungry for fresh leadership to help the company regain lost ground. If the Board is serious about rebuilding trust, the first step is obvious: the Board should demand that Brian Cornell step down as Executive Chair immediately.

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

Even in the depths of the recent crypto winter, stablecoins stood out as a beacon of optimism. Investors and startups hail them as a long-awaited killer application, and even crypto haters concede stablecoins are a superior technology for moving money around—especially following the 2025 passing of the GENIUS Act, which created a predictable regulatory environment for using the tokens. Still, if stablecoins are so great, why is it so hard to figure out who’s going to use them?

That question occurred to me last week while speaking with Dan Kim, who is VP of Product at the emerging fintech giant Airwallex. Prior to his current gig, Kim held a senior business development role at Coinbase, where he led efforts to persuade merchants to adopt stablecoins. It was a tough sell. Many times, Kim recalls, merchants had concerns about who would handle chargebacks, or simply balked at adding another layer of complexity to an already-complicated payment system. “I ran into a blocker for how to make stablecoins useful … It was a dead end,” Kim told me.

Kim may have a point. While stablecoins are useful in the world of crypto trading, I’ve never seen a need for them in my day-to-day life. If I need to send money to a friend or local business, Venmo and Zelle work just dandy, and even if Coinbase can stand up a robust ecosystem of stablecoin rewards—which it is trying to do with USDC—I can’t see a world where those rewards are more lucrative than what I accrue with my credit cards.

It’s a different story in some countries, of course. If you want proof, look at the $183 billion stablecoin business that Tether has built by catering to consumers in developing countries who want a reliable way to hold U.S. dollars. Meanwhile, in places like Northern Europe where many favor debit cards, it’s possible stablecoin rewards could play well. But here in North America, it’s hard to see why consumers will seek out stablecoins anytime soon.

That leaves cross-border business payments, which seem a natural use case for stablecoins, since they allow for much faster and more secure transfers than wires and other legacy technologies. But this too is more complicated than it looks. As Kim pointed out, big companies are well-poised to adopt stablecoins, but that might not be the case for the smaller vendors that serve them. In some markets, those vendors may face regulatory restrictions on accepting crypto or U.S. dollar payments and, in any case, they still need to operate in their country’s native currency.

In theory, these bottlenecks could be resolved by the adoption of various national stablecoins—a real stablecoin in Brazil, a Canadian dollar stablecoin in Canada, and so on. The reality though is that, even though these other tokens do exist, a whopping 98% of stablecoins out there are backed by U.S. dollars—and that’s been the case for years.

All of this is why Airwallex’s CEO, Jack Zhang, told me in 2024 that he was skeptical of stablecoins entirely, and figured that his company—which you can think of as a Wise for the B2B crowd—felt its existing business model was sufficient. That model revolves around acquiring financial licenses, and holding large pools of local currencies around the world, in order to provide companies with low-cost forex transfers.

A lot can change in two years, however, and Airwallex has started to change its tune on stablecoins. According to Kim, stablecoins are now so widespread that Airwallex has created a service focused on the last mile—namely, helping customers convert U.S. stablecoins into local currency.

Airwallex—which has taken a hefty investment from Visa—is also backing a startup called Metal that is building a blockchain designed to be compliant with all local financial regulations right out of the box. Finally, Kim noted that stablecoin use will grow in the coming era of agentic commerce, where bots will handle a portion of our shopping.

So what to make of all this? On one hand, Airwallex makes a compelling case that it is entering the stablecoin market at the right time, and with the right business model. On the other, it’s fair to ask if the company simply decided “better late than never.” In any case, Airwallex will be in tough competing on the last mile front with the likes of Rain and MoonPay. Meanwhile, agentic commerce is still in an early-adopter phase, and there is a long list of companies—including Coinbase, Robinhood, and Stripe—racing to figure it out first.

Long story short, all we really know right now is that it’s too soon to say how any of this will turn out.

Jeff John Roberts
jeff.roberts@fortune.com
@jeffjohnroberts

A note to readers: We will not be publishing on the Labor Day holiday, but will be back in your inbox on Sept. 14.

This story was originally featured on Fortune.com

Someone in my network moved his family across the country for a new job last year. New schools. New mortgage. A whole life uprooted on the strength of an offer letter. He was laid off eleven weeks later.

He is not the exception. I’m part of a network of senior executives navigating this job market, and I’ve heard this exact story so many times it has stopped shocking me. That should scare all of us.

American business has quietly rewritten its deal with workers. And a generation has been taking notes.

Start with the exit. Most U.S. employment is “at will.”  A company can end your career anytime, for almost any reason, with no notice. That’s legal, and America sets no severance requirement. It’s brutal when a job search runs for months.

And the health insurance? It can end the same day. Not in 30 days, that day. For a client mid-cancer-treatment, or a parent covering a medically complex kid, that isn’t a policy detail. It’s a trapdoor. COBRA is supposed to be the bridge, but in 2025 the full family premium averaged nearly $27,000 a year or about $2,250 a month (Kaiser Family Foundation). This is why people don’t experience a layoff as a business decision. They experience it as abandonment.

First, the lies.

The excuse of the moment is AI. In April 2026, roughly a quarter of announced U.S. job cuts were blamed on it. The top stated reason two months running. But even the people building AI aren’t buying it. OpenAI’s Sam Altman has called it out as “AI washing.” They were cuts companies planned anyway. MIT’s Paul Osterman was blunter: AI is “a perfect excuse to justify big layoffs.”

The receipts back him up. In one Forrester survey, 55% of leaders who cut jobs for AI admitted it was a mistake. And AI is just the newest costume: a 2024 survey found eight in ten leaders had used “layoffs” to cut someone they wanted gone, and 54% did it to dodge paying severance. Workers can feel when they’re being lied to. They just can’t always prove it.

Then, the broken promises.

Millions took jobs sold as remote, then got return-to-office mandates — or relocate-or-quit ultimatums — they never signed up for. Amazon hauled hundreds of thousands back five days a week. Starbucks told corporate leaders to move to Seattle or Toronto within a year or take the exit. The excuse is always that presence drives performance.

It doesn’t. University of Pittsburgh researchers studied RTO mandates at big public companies and found no gain in financial performance — just a drop in employee satisfaction. As Wharton’s Adam Grant puts it, “Don’t mistake presence for performance.” The mandates didn’t bring the work back. They told people their trust was worth less than a badge swipe.

Finally, the math that gives the whole game away.

The average big-company CEO now makes about 281x the typical worker. At some companies it’s in the thousands. Starbucks hit 6,666x last year. Since 1978, CEO pay is up more than 1,000%. Worker pay: 24%. The pain isn’t shared. When Meta cut about 3,600 people it branded “low performers” in early 2025, it raised executive bonus targets from 75% to 200% of salary about a week later.

A generation raised on that arithmetic does not need an economics degree to reach a verdict.

I’m the parent of a Gen Z kid, so I hear the verdict at my own dinner table. Just 17% of Americans told Gallup in 2026 they have real confidence in big business, near a record low. Among adults under 35, nearly half now view socialism favorably. A colleague told me, half-terrified, that her son is becoming a full-blown socialist. I’ve seen what he’s seen: loyalty punished, honesty optional, promises reversed, the gains routed to the top.

When people stop believing their work protects them, they don’t decide one company is bad. They decide the system is rigged. And they go looking for another one.

And increasingly, they’re not just looking. They’re leaving. Americans filed a record 5.5 million new business applications in 2023 and have held roughly that pace since, up about 50% from before the pandemic. Nearly four in 10 recent college graduates say they’d rather start a business than climb someone else’s ladder. When a generation stops trusting the deal, they go into business for themselves.

Business leaders keep asking why young workers seem disloyal, disengaged, quick to walk. They have it backward. This generation isn’t refusing to buy in. They were never sold anything worth buying.

The fixes aren’t mysteries. Tell people the truth when you let them go. Give them a landing soft enough that losing a job doesn’t cost them the house. Honor the deal you hired them under, and when the company wins, cut in the people who built it. That’s the whole list. It costs money and it costs ego — which is exactly why most companies won’t do it, and why the sharpest young talent has already stopped waiting. They’re not disengaged. They’re gone, or quietly building the thing that takes your customers in 10 years. You didn’t lose their loyalty. You taught them not to offer it. That bill comes due on your watch.

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

A U.S.-Venezuela oil deal has been announced by President Trump and Interim President Delcy Rodríguez. The public knows virtually nothing about the details of the agreement — nor do I, a Special Adviser to Venezuelan Congressman Antonio Ecarri on Economic, Monetary, and Energy Affairs. This deal was clearly arrived at in secrecy, with no public debate, and signed under duress. Therefore, it is illegitimate and probably illegal. 

This deal took not only me, but everyone I am in touch with in Caracas by surprise. That being said, it is vital to understand the importance of establishing clear private property rights in Venezuela’s vast oil reserves. The establishment of such private rights would give Venezuela’s oil reserves a positive present value. It’s important to understand why that’s not the case now. 

PDVSA is a state-owned oil company that dominates Venezuela’s economy and accounts for almost 95% of Venezuela’s foreign exchange earnings. Even by state-owned enterprise standards, PDVSA is grossly mismanaged, as evidenced by its production and reserve figures.

Under the direction of Luis Giusti in the 1994-1998 period, PDVSA’s production soared. This trend changed in 1999, when Hugo Chavez became Venezuela’s president and introduced Chavismo as the country’s guiding economic doctrine. Venezuela’s oil output began to stagnate, a situation which worsened further after the coup attempt of April 2002. Chavez responded by purging PDVSA of its professionals en masse, replacing them with “reliable” hands who were loyal to Chavez’s socialist regime.

After the 2002-2003 output plunge, Venezuela’s production temporarily recovered. However, with the death of Chavez and Nicolas Maduro’s assumption of the presidency in March 2013, another output plunge began. This trend has left Venezuela’s output drastically lower than when Chavez took power in 1999.

PDVSA’s physical capital has been consumed at an unsustainably rapid rate, with capital expenditures far below the value of equipment that is being consumed each year by depreciation and amortization. On top of PDVSA’s reduced capital stock and its deteriorating quality, there has also been a drop in the stock and quality of its human capital. For example, in 2017, President Nicolas Maduro named a National Guard general with no industry experience to lead PDVSA. The combination of plunging physical and human capital has left the giant state-owned oil company in very bad shape. Equipment breakdowns and increased accident rates have contributed further to long downtimes and output declines.

It is important to note that PDVSA’s decreased output is not due to dwindling oil reserves, but rather is caused by changes in the rate at which its reserves are being depleted. The depletion rate provides the key to understanding the economics of an oil company and the value of its reserves. Venezuela’s depletion rate has been falling rapidly since 2007. At present, it sits at 0.124% per year, indicating that it would take 558 years for PDVSA’s reserves to be halfway depleted.

This has noteworthy economic implications because of positive time preference and discounting. It is rather obvious that if you have to wait 558 years to produce and sell a barrel of oil, that barrel is virtually worthless in today’s dollars. Therefore, at current depletion rates, most of Venezuela’s oil reserves are worthless.

To put Venezuela’s depletion rate into perspective, consider Exxon, one of the world’s largest oil companies. Exxon’s depletion rate is close to 9% per year. That rate implies that it would take 7.4 years for Exxon’s oil reserves to be halfway depleted.

It is important to mention that I am writing as someone with experience in petroleum economics. Indeed, I was a member of the United Arab Emirate’s Financial Advisory Council from 2008 to 2014. In the UAE, I used a simple model that I had developed, plugged in realistic numbers, and concluded that the UAE should be depleting its vast oil reserves at a much more rapid rate than it was.

My advice to the UAE was to take the money and run.

The UAE agreed. For years, it attempted to obtain a dramatic increase in its OPEC quota. But a dramatic increase was never forthcoming. As a result, in May 2026, the UAE took the exit door and left OPEC.

It’s time for Venezuela to kill inflation by mothballing the bolivar, putting it in a museum, and replacing it with the U.S dollar. After that positive confidence shock, Venezuela must employ all legitimate means to privatize its oil industry and dramatically increase its production. 

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

Nicholas Gordon here. When is Airwallex, the Australian-born payments company, planning to go public? The company, which serves over 675,000 businesses and claims to have over $1 billion in annualized run rate revenue, once talked about having a U.S. IPO in 2026, but those plans have gotten fuzzier over time.

In a recent conversation for our new Asia Agenda series, president Lucy Liu told me that the firm still plans to be “IPO-ready” by the end of the year, but added that it’s “just not the best time, given how complicated things are.”

One can’t blame her for thinking that way. 2026 already boasts two key IPOs—SpaceX’s $85.7 billion debut and SK Hynix’s $26.5 billion ADR sale—and potentially a third in Anthropic’s. (OpenAI is reportedly considering a debut in 2027.) 

And it’s not just the U.S. Shanghai has already had one blockbuster IPO, in ChangXin Memory Technologies’ $9.8 billion listing. And Alibaba just tapped Hong Kong’s equity markets for a $10 billion share sale.

A smaller company–even one as buzzy as Airwallex–might get lost in the shuffle as bigger investors hold on to their cash. 

Other startups have delayed their IPO plans. Travel platform Klook filed for a New York IPO last November. The company originally scheduled it for the end of 2025, then delayed it until early 2026, and has remained studiously quiet since then. (Klook’s founder Ethan Lin declined to comment on the company’s IPO plans when my colleague Angelica Ang spoke to him earlier this year.)

But it’s also not clear that companies like Airwallex need to go public right now. The company raised $330 million in December, and then raised another $320 million in June. “We just want to be able to have enough capital to fast-charge our plans,” Liu told me, as the company barrels into agentic commerce and automated bookkeeping. 

“Investors are going more towards later-stage investments,” Liu added. “It’s not that they don’t have capital. They just want to see success, right? They want to see a track record before they deploy capital into that particular company.”

Airwallex’s June funding round was a “Series H” round, which isn’t that ridiculous in the world of tech anymore. Databricks is reportedly putting together a Series M round, amid social media jokes that these funding rounds are soon going to run out of letters.

Liu’s story with Airwallex starts in 2015, when she met fellow cofounders Jack Zhang and Max Li; she eventually offered them $1 million in seed capital, twice what they’d asked for. She noted to me that she had a “bit of an overconfidence situation” at the start: “I would be on a plane almost every other day.”

Now, the company is pushing into markets like Mexico, South Korea, Brazil and, importantly, the U.S. It recently moved its headquarters from Australia—which Liu admitted is “quite small” for companies thinking of going global—to Singapore and San Francisco.

That headquarters shift is meant to help its path into the U.S., which hasn’t entirely been smooth. Last November, venture capitalist Keith Rabois (who sits on the board of Ramp, a competitor) called Airwallex a “Chinese backdoor into sensitive American data.” That accusation got picked up by Senator Tom Cotton, who claimed the startup’s “ties to Communist China run deep,” and called for a CFIUS investigation.

Airwallex has pushed back against the claims. CEO Jack Zhang has called them “false,” noted that its China-based staff can’t access U.S. data, and added that third-party companies were hired to audit its data security practices. (Liu, in our conversation, pointed to previous statements by Airwallex.)

So, Zhang, Liu, and Airwallex’s cofounders can probably afford to wait until things settle down before getting an IPO on the calendar. 

But there’s a risk in waiting too long: Just look at Shein, and the 75% haircut it’s taking on its valuation—or Airtable, which never went public and sold at an 80% discount to Bending Spoons. Airwallex can afford to wait, but perhaps not forever.

See you tomorrow,

Nicholas Gordon
X: @nickrigordon
Email: nicholas.gordon@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

A 63-year-old Supreme Court ruling about two Philadelphia banks is now the central legal weapon in the fight over Paramount’s $110 billion deal to buy Warner Bros. Discovery – a transaction that passed muster with the Justice Department and in every one of the 68 jurisdictions around the world where it was reviewed – but that 12 state attorneys general are suing to block, citing that decades-old case as their guiding precedent.

When, in United States v. Philadelphia National Bank (PNB), the Supreme Court blocked two Philadelphia banks from merging, banking was simple to measure. The product was checking accounts and loans, and the market was one city. The Court set out an arbitrary rule of thumb, deciding that if a merger gives you about 30% of a market, courts will assume it hurts competition. This standard was derived from the static and predictable market of bricks and mortar banking in a local area. But, by making up that 30% threshold, the case created the machinery to answer the question Congress wanted to address – whether a merger would substantially lessen competition. If PNB’s 30% market-share threshold is triggered by the states’ narrow market definition, however, it would create a legal presumption of harm the states are counting on to win, even without proving actual consumer damage.

There is a broad debate over the applicability of PNB to modern anti-trust cases, with many legal scholars finding the application of a random number, without any real analysis of the potential harm a merger might cause, to be a fatal flaw. In the Paramount case, the problem is more sharply defined because of how the states are using PNB. The rule is broadly triggered at the 30% threshold, and the states have drawn their market lines to get the number they need.

The math was easier in 1963 because banking in Philadelphia really was a closed world. If you wanted a checking account, you went to a local bank, period.

The states measure the Paramount deal the same way, as if entertainment still came in closed boxes. Their case counts wide-release theatrical movies and the basic cable bundle, and leaves out streaming, YouTube, sports rights, and everything else that now competes for the same hours of your evening. It is like declaring someone the tallest person in the room after sending everyone taller out the door.

Ask any household how they actually watch. YouTube is now the most-watched form of television in America. Streaming makes up nearly the majority of all TV time. Netflix and Amazon are buying up live sports. The complaint implies the combined company is dominant in a market that viewers are walking away from. The fact that plaintiffs can gerrymander market definitions to trigger PNB’s structural presumption is exactly the kind of abuse that has led scholars to argue PNB should be overturned. Here is the strange part. Under certain interpretations of the 1963 rule, courts are not allowed to weigh whether a merger might make competition stronger overall. The Supreme Court once said benefits in one area cannot excuse concentration in another, and courts have followed that ever since.

But that application asks for a throwaway line to carry far more weight than was intended. At most, PNB stands for the narrow proposition that a merging firm can’t justify harm in one market by pointing to unrelated benefits in another. It does not say a court must ignore everything outside a narrowly drawn market. Some enforcers have stretched it that far regardless, which is exactly what the states are leaning on.

The one question at the center of this deal, whether combining two old-line studios helps them compete with Netflix, Amazon, Apple, and YouTube, should still be given consideration. In a streaming market where content is king, the need to compete by constantly churning out new films and shows is what will incentivize the two studios to up their game and produce more quality features. The competitive dynamics cannot be ignored.

Neither can consumers and the simplest question is probably the most important – how, under this merger, does an ordinary viewer suffer harm? The states lead with statistics instead of a story about real harm, probably because the harm story is weaker than the math.

Now run it the other way. If the deal closes, you get one studio with the scale and incentive to invest and compete with the tech platforms. If it dies, you get two isolated companies struggling to keep up. Blocking this merger does not protect competition.

Antitrust law is supposed to protect competition and the people who benefit from it. A yardstick built for neighborhood banks cannot measure a business where a teenager’s phone competes with a movie theater.

There is also a bigger problem here. Federal enforcers ran a modern analysis and cleared this deal. Then a dozen state officials re-ran it under a 63-year-old rule in a courthouse they picked. If that can happen to any deal, federal clearance means very little.

Congress should make sure that deals of true national scope get one expert review under one modern standard, the way Europe has done for decades. Until then, Philadelphia National Bank will keep being used as a stand-in for the argument about consumer harm that the states cannot actually make.

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:

• Temu funds an army of fake accounts on Instagram and Facebook.
• Iran and the U.S. resume hostilities in the Gulf.
• Markets: Oil up, stocks mixed.
• Stocks go up as Trump’s ratings decline, Bank of America says.
• AI’s building boom is big, not bubbly.
• People are using LinkedIn to find and screen their dates.

➡️ 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: Patrick Kelleher explains why robots that can do a Vulcan salute are more impressive than those that can run faster than Usain Bolt
• The big leadership story: Offshore nuclear power: A question of ‘when, not if’
• The markets: Stocks mixed after renewed Iran conflict and Warsh’s Jackson Hole comments
• Plus: All the news and watercooler chat from Fortune.

Good morning. Last week’s World Humanoid Robot Games in Beijing showcased more than 2,000 robots, one of which beat Usain Bolt’s world record in the 100-meter sprint. Another was able to surpass humans in a 2.88-meter standing high jump. But humanoids that can run and jump aren’t particularly useful to GXO Logistics CEO Patrick Kelleher. For his business, the ideal is one that has the strength to unload a truck, and the dexterity to sort lipstick—as some of the humanoids working in his operations can now do.

The $13.6 billion-a-year global logistics giant is running pilots with five humanoid-robotics providers, and another European pilot is expected later this year. A big focus is on what he calls “single-digit challenges”—getting robot hands to more closely mimic human hands. 

“One of the great innovations we participated in this year was working on the ability of humanoid hands to go like this,” Kelleher told me, moving his fingers apart and together in a gesture that reminded me of a Vulcan salute.  “This requires multiple flexible joints [and] will be game-changing in terms of what a humanoid can do.”

GXO is working with two kinds of humanoids: bipedals that walk around on two legs, and those that have a wheel and a base. What differentiates humanoids from industrial robots is their degrees of freedom, or ability to mimic the 200-plus movements that the human body is capable of. While a typical industrial robot has four to six movements, humanoids can now have more than 100 degrees of freedom.

Kelleher is not looking to replace the 150,000 people who currently work for him, he says, but rather to supplement them and address an industry-wide challenge. “We have 25% turnover in the warehouse environment … and we can’t find enough labor,” said Kelleher. “Humanoid robotics AI is really about helping our associates do their job better, more productively, and more easily, at higher quality.”

The experiments are turning GXO into a proving ground for a new model of human-machine collaboration: “The warehouse has become a hotbed of technology and innovation. We’re working with our partners on fully autonomous forklifts and humanoids that work where it’s minus 32 degrees, in the freezer,” he told me. “It used to be warehousing and outsourcing was about labor arbitrage, getting people to do a job at lower cost, and it’s really evolved now to become an arbitrage of expertise …. We are the living lab. It’s exciting.”

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

This story was originally featured on Fortune.com

Imagine an executive team gathering to discuss its most urgent strategic priority: artificial intelligence. The CEO wants AI to become a new engine of growth. The CIO wants to deploy copilots. The COO sees opportunities to automate processes. The head of product wants AI-enabled offerings. The CHRO is worried about how AI will reshape work. Everyone agrees that AI is the future. Everyone agrees that the company needs to innovate. And everyone leaves the meeting talking about something different. 

This is increasingly the real innovation problem inside large organizations. We use one word, “innovation,” to describe fundamentally different kinds of work. Adopting a technology. Improving a process. Launching a product. Creating a business model. Transforming an organization. Responding to a crisis. Everyone uses the same word and assumes alignment. There isn’t any.

I’ve been called the “Dean of Innovation” for nearly four decades, although the title started more as a bit of good-natured ribbing than a credential. I was 27 and Vice President of New Ventures at Domino’s Pizza during the company’s explosive growth in the 1980s when I attended a retreat with Tom Peters, the legendary business author. Somehow “Dean of Innovation” emerged from the gathering, and my older colleagues at Domino’s found it amusing enough to make it stick. Since then, I’ve worked with much of the Fortune 500, the U.S. military, and more cultural organizations than I can remember. After decades of tours of duty inside organizations trying to make innovation actually happen, I’ve become convinced of one thing: almost nobody means the same thing when they use the word.

The result is predictable: resources scatter, expectations conflict, and initiatives are managed with the wrong processes and measured by the wrong metrics. Most companies don’t need more innovation. They need a better way to talk about it. 

Stop Asking, ‘Is It Innovative?‘

For decades, executives and academics have tried to define innovation: incremental or radical, sustaining or disruptive, product or process. A better question is simpler: Innovative compared with what? Innovation is positive deviance from the norm that creates value. The same practice can be revolutionary in one company, incremental in another, and obsolete in a third. For a traditional company with decades of legacy systems, using AI to redesign a core process may represent a major departure from the norm. For an AI-native startup, it may be table stakes. The technology may be similar. The starting points are not. That is why leaders should ask two questions: What norm are we breaking? What new value are we creating, and for whom? Without those answers, “innovation” is little more than a corporate aspiration. 

Nvidia’s extraordinary rise, for example, cannot be explained simply by saying the company innovated. Its graphics processing capabilities were repeatedly extended into new arenas, most consequentially AI computing. That is the real work of innovation: departing from the existing norm in a way that creates new value. And the farther an idea departs from the norm, the harder it becomes to prove that it will work. 

The Data Trap

Executives are trained to demand evidence before committing resources. That is generally good management. It can also kill innovation. The paradox is simple: The more genuinely novel an idea is, the less reliable the historical data becomes. A modest improvement to an existing product can usually be modeled. Customers are known. Costs can be estimated. Competitors can be studied. But what happens when a company enters a market that does not yet exist or deploys a technology whose capabilities and economics are changing every few months? 

Consider generative AI. In late 2022, no executive could produce a credible five-year ROI model. The technology, costs, competitors, regulations, and use cases were changing too quickly. The companies that waited for certainty did not reduce uncertainty. They simply learned more slowly. This is the data trap. The ideas with the strongest evidence are usually those most similar to what the organization already does. Demand too much proof, and you unintentionally select for the familiar. Innovation gradually becomes optimization. 

Instead of asking, “Can you prove this will work?” ask, “What is the cheapest, fastest experiment that will tell us something important?” In established operations, data informs action. In innovation, action often creates the data. Experiments do not validate the plan. They create the knowledge from which a plan can eventually be built. 

Innovation Is Not One Game

Organizations also make the mistake of treating innovation as a single activity. It isn’t. UPS improving route efficiency is largely an optimization challenge. The system exists, data is abundant, and progress can be measured precisely. A company responding to a cyberattack is playing a different game. Speed matters more than perfect information. A legacy company building an AI-enabled service faces uncertainty about customers, pricing, and the business model. Small experiments and staged investments are appropriate. An automaker shifting toward electric and software-defined vehicles faces something larger still: a transformation affecting manufacturing, suppliers, talent, capital, and organizational identity. 

All four require good management. They do not require the same management. Yet companies routinely force them through the same stage gates, funding processes, and metrics. Before choosing the process, leaders should ask: How large is the departure from what we already know? How quickly must we move? How much uncertainty exists? The answers should determine the team, funding, governance, metrics, and pace, not the other way around. 

Most Corporate ‘Innovation’ Is Actually Something Else

Companies call almost every form of change “innovation”. Installing a new ERP system. Adopting Microsoft Copilot. Reorganizing a business unit. Digitizing a process. Running a hackathon. These activities may be valuable. But they are not the same kind of work. A useful distinction is simple: Creativity produces possibilities. Innovation develops and tests novel possibilities to create value. Change gets people to adopt and scale what has been chosen. Buying an AI platform is technology acquisition. Getting thousands of employees to use it is change management. Discovering a fundamentally new way to create customer value with it is innovation. A company can succeed at the first two without accomplishing the third. 

This is why counting AI pilots tells us little about whether an organization is becoming more innovative. Automating an existing process may increase productivity. Deploying copilots may help employees work faster. Those may be excellent investments. But adoption is not invention, and efficiency is not necessarily innovation. 

The Danger of False Alignment

Return to the executive meeting. The CFO believes the AI initiative should reduce costs within 12 months. The head of product wants new sources of revenue. The CIO wants a secure technology platform. The CHRO wants to redesign work. All four may be right. But unless those differences are made explicit, the initiative will eventually be judged against conflicting expectations. The organization does not have an execution problem. It has a language problem. 

Leaders often respond by pushing harder for alignment. But in uncertain situations, disagreement is information. One executive may see an efficiency opportunity. Another sees a competitive threat. A third sees a new business model. The goal should not be to eliminate those differences too quickly. It should be to make them visible. That is not dysfunction. That is clarity. 

A Five-Question Language for Innovation

Executives do not need another elaborate innovation taxonomy. They need a few questions they can use in a meeting, funding review, or conversation with a team. Before approving any innovation initiative, ask five. 

• 1. What norm are we breaking? Is it a norm inside our company, our industry, the technology, or customer expectations? If you cannot identify the norm being broken, you may simply be improving what already exists. 
• 2. What new value are we creating, and for whom? Innovation is not novelty. “We need an AI strategy” is no more useful than saying “we need an internet strategy” was 25 years ago. The technology is not the strategy. The question is what becomes possible now that was not possible before. If the value cannot be named, the initiative may be technology theater. 
• 3. How big is the departure from what we already know? An incremental improvement may involve familiar customers, technologies, and economics. A new business model may call all three into question. Small departures can often be planned. Large departures need to be learned. Match the size of the investment to the amount of knowledge available. 
• 4. How fast do we need to move? Speed is not always a virtue. A cybersecurity breach may demand action in hours. A new business model may take years to mature. The question is not, “How can we move faster?” It is: What is the appropriate speed for this kind of uncertainty? 
• 5. What is the next experiment? Not: What is the five-year plan? Not: When can we scale it? What is the next experiment? A good experiment answers an important question at an acceptable cost. Will customers use it? Can the technology work under real conditions? Will someone pay for it? What assumption, if wrong, would cause the idea to fail? The purpose is not to prove that the team was right. It is to learn what is true. The real failure is spending two years building something that could have been disproven in two weeks. 

Change the Questions, Change the Conversation

Instead of asking, “Is it innovative?” ask, “What norm are we breaking?” Instead of asking, “What is the ROI?” ask, “What must we learn before investing more?” Instead of asking, “What is the plan?” ask, “What is the next experiment?” Instead of asking, “Are we aligned?” ask, “Where do we see the problem differently?” Instead of asking, “How quickly can we scale?” ask, “Have we created enough value to scale yet?” 

These questions do not eliminate uncertainty. They make it manageable. An optimization effort should be accountable for results. An experiment should be accountable for learning. A crisis response should be accountable for speed. A transformation should be accountable for building capabilities. When leaders use the same language and metrics for all four, they create confusion. When they distinguish among them, they can manage each more intelligently. 

Innovation Is Something You Navigate

As AI accelerates technological change, more initiatives will be labeled innovative. More companies will launch pilots, labs, and strategic initiatives. More money will be committed before leaders agree on what kind of problem they are actually trying to solve. The companies that succeed will not necessarily be those with the most ideas. They will be the ones that can distinguish optimization from invention, adoption from experimentation, evidence from assumptions, and plans from learning. They will know when to demand data and when to create it. When to move quickly and when to be patient. When to scale and when to experiment. 

Innovation has always been difficult to define because it is not a fixed thing. It changes with the organization, the industry, the technology, and the moment. That is why the search for a perfect definition has always been a dead end. Innovation does not need another definition. It needs a language leaders can actually use. 

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

When North Carolina-based writer Pat Brothwell received a message request on LinkedIn from a New York matchmaker, he initially thought it might be a scam. 

Instead, Brothwell said the sender, Marin Haugo, invited him to discuss potential matches through Maar Dating Club NYC. During their meeting, the matchmaker appeared to have done little research into his background. 

“It felt like when a headhunter contacts you, and they just did a very cursory search,” Brothwell told Fortune. “She didn’t actually look into it because all her clients are based in New York. I live in Asheville, North Carolina.” 

Brothwell said the matchmaker reassured him that distance wouldn’t be an issue, and suggested he fly to New York for dates if necessary, a comment that made him suspect her clients were in “different tax brackets” than most.

“There was a weird angle of, ‘I have wealthy clients, and I’m gonna try to find someone who’s just excited to date somebody wealthy,’” Brothwell said. “Everything just felt like it was maybe a little bit more than met the eye.”

Brothwell never met any of the matchmaker’s clients after the meeting, which he said took place three years ago. He said Haugo stopped responding when he told her he was a nonfiction author and was participating for research purposes. Brothwell assumed it was because Haugo feared being written about on his blog. Haugo did not respond to Fortune’s request for comment.

LinkedIn’s romance problem

The encounter illustrates an uncomfortable tension on LinkedIn: Some users and matchmakers treat the professional network as a source for romantic prospects, even though the company explicitly prohibits members from using it to pursue them.

“LinkedIn is a professional networking platform, not a dating site,”  LinkedIn’s professional community policies state. “Do not use LinkedIn to pursue romantic connections, ask for romantic dates, or provide sexual commentary on someone’s appearance or perceived attractiveness.”

But a June Zety survey of more than 1,000 U.S. employees found that one in four believe it’s “fair game” for romantic advances. The survey also found that 22% had reached out or responded to someone on LinkedIn with romantic intent, while 12% had formed a romantic relationship that originated on LinkedIn. 

LinkedIn says it removes unwanted advances on the platform and sexually explicit material. 

In the second half of 2025, LinkedIn removed 477,911 comments that provoked harassment or abuse and 80,806 comments that used adult language or imagery using its harmful message detector, according to LinkedIn’s transparency report. The advanced safety detector, which screens incoming messages for policy violations, must be manually enabled by users.

“Our focus is on making sure that people are safe from unwanted romantic advances, which do go against our professional community policies,” a LinkedIn spokesperson told Fortune.

If a user violates one of LinkedIn’s policies, the platform sends a warning, and repeated violations often result in restricted access to LinkedIn features. In the most serious violations, a user may be permanently banned from the platform.

Why LinkedIn has become a dating-vetting tool

Searching a person’s public profile before a date is different from using LinkedIn’s messaging tools to make an unsolicited romantic advance—and more closely resembles the kind of online due diligence that has become routine in app-based dating. 

“With over 100 million verified members, it’s not a surprise that people come to LinkedIn,” the LinkedIn spokesperson said. “But LinkedIn is a platform for professional growth and career advancement, not finding a date.”

Nevertheless, online daters use LinkedIn as a vetting tool, with almost half of Zety’s survey respondents believing that information posted to the platform is more reliable than jobs and college attendance shown on dating apps such as Hinge.

While LinkedIn matchmakers are strictly prohibited under the platform’s guidelines, professional matchmakers recommend researching potential partners online, including scanning their LinkedIn profiles.

“Relationship-minded singles are increasingly frustrated and fatigued by fake profiles, misleading information, safety concerns, and the lack of context on dating apps,” the director of coaching at The Matchmaking Company, Heather Drury, told Fortune. “As a result, singles have become their own private investigators. Some are turning to LinkedIn as an additional tool in their vetting process.”

The Matchmaking Company is a family-owned and operated matchmaking firm with brick-and-mortar locations in 24 states. The company has previously used LinkedIn to scout potential matches for its clients.

“Professional photos, career history, education, connections, interests, and community involvement can offer additional context that may not be available on a traditional dating profile,” Drury said. “For someone considering whether to meet a potential romantic partner, that additional information [on LinkedIn] can help them feel more informed and comfortable.”

Instead of prioritizing appearance, which is something only 18% of Gen Z daters rank first in importance, most Gen Z singles look for emotional maturity and kindness in potential partners instead, according to a survey conducted by the luxury dating site Seeking.

The survey also found that Gen Z prioritizes specific qualities such as shared values, ambition, and generosity.

“I find myself gravitating towards the same familiar social circles, and in that sense, university or professional career become big cues for me in determining that alignment,” a New York University freshman, who asked to remain anonymous, told Fortune. 

A separate February survey from Resume Templates revealed that 92% of single LinkedIn users have used the platform to research online app matches before their first date, specifically paying attention to current job status, mutual connections, academic background, and endorsements. Additionally, 88% of LinkedIn users have used the platform to check up on an ex-partner or former romantic interest.

“Using LinkedIn has become a natural part of online dating,” the student said. “In online dating, where the amount of disclosed information is so limited and social media presence is often highly filtered or non-existent for guys, LinkedIn becomes a surprisingly telling and rich source of information.”

For some daters, that extra context isn’t simply curiosity. It can also feel like a safety check in an online-dating environment where users may worry about scams, harassment, or whether a match is accurately representing themselves.

Two-thirds of dating app users have reported instances of cyberstalking facilitated by dating apps, and 48% reported experiences of online interpersonal violence, including receiving unsolicited sexual pictures or messages, name-calling, and even threats of physical violence, according to a study published in the Asian Journal of Criminology last year.

In addition, information displayed on dating app profiles makes it easy for cyberstalkers to find users’ other online profiles, such as LinkedIn, or to find them through simple Google searches, stripping away the relative anonymity and safety barriers on dating apps.

This story was originally featured on Fortune.com

If you are one of Ya Lili’s 183,000 followers on her “@findgadgetswithme” Instagram account, you may have seen this video she posted recently: A man stands on a sunny beach, dressed only in swim shorts, his back covered in dirt. “This won’t come off!” he shouts as he rubs futilely at the brown filth covering his arm. Suddenly, out of nowhere, a fully dressed woman walks into the frame and says, “Because you’re doing it wrong!” She then takes a bath mitt and wipes away a large swathe of the soil.

“If you’re still using your bare hands, you are missing out,” the voiceover says, as the action switches to a close-up of a woman using the sponge glove to clean a series of inexplicable squiggly lines from her bare thighs.

It’s AI slop, obviously. 

But if you want to buy this shower scrubber, there’s a link under the video where you can find them for sale on Temu.

Ya Lily is the top creator for Temu’s partnership ads on Meta platforms in the U.K. and Europe, according to research seen by Fortune. The problem is, Ya Lily almost certainly does not exist. She’s a fake account. 

In fact, of the top 100 creators boosted by Temu’s partnership ads, 73 are likely fakes, the research suggests.

It is not clear whether Temu is doing this deliberately or is being fooled by scammers into spending money on influencers who don’t exist. But the likely fake accounts potentially set up a legal quagmire for Temu, as European law bars the use of misleading formats for advertising. 

Neither Temu nor Ya Lilly responded to multiple requests for comment from Fortune. Meta declined comment when reached.

Above: One of YaLily’s Temu partnership ads on Instagram.

Real money going to fake accounts

Ya Lily has a further 129,000 followers on Facebook, where she goes by a completely different name, “Must good.” That page gives her location as “England, Birmingham, MO, United States.” Her bio says “You are my dream of dead.” And her email address appears to be Chinese.

And yet her posts—which all contain links to a random assortment of Temu products—were boosted by Temu and seen more than 1 billion times across both platforms in 16 months ending in April 2026, according to Online Risk Labs. ORL is an independent non-profit based in the Czech Republic, which researches cybersecurity and systemic risk online.

Ya Lily’s content ran in nearly 109,541 Temu ad campaigns in that period, the data show. That’s about 225 campaigns per day.

Using creators on Facebook and Instagram to churn out an endless stream of low-effort AI promotions is one of Temu’s main ways to reach people on social media, ORL’s data suggest. Temu spent as much as $962 million on ads like these in the period, in the U.K. and 27 E.U. countries, according to ORL. 

Its estimate is based on total impressions and average cost per reach. The group estimates that in any given quarter Temu alone forms nearly 2% of all the revenue Meta earns from the continent.

Fortune asked two experts in social media advertising to check ORL’s data and estimates. Both found them plausible.

The ORL data focuses only on the U.K. and Europe because the Digital Services Act, which covers those countries, requires large platform providers to publish a public library of all the ads they run, allowing researchers to identify trends or problematic commercial behavior. No such database exists for the U.S.

“Among the top 100 influencers collaborating with Temu, there are certainly some real people among them. In my estimate, however, they account for no more than 15–20%. By contrast, I have serious doubts about the authenticity of accounts that, according to the platform, are based in Russia, China, or Iran,” ORL’s manager, Vendula Prokůpková, told Fortune.

Legally questionable 

That could raise legal issues, according to Stuart Lester, the partner who leads law firm Mishcon de Reya’s advertising and marketing group, in London.

“By all means utilize Facebook and Instagram and influencers, but know that there are rules that you have to abide by, and you certainly can’t be creating fake accounts and portraying these as real individuals. That’s misleading on many levels.”

There are several laws and regulatory agencies that might see the use of fake accounts—by scammers, advertisers, or even their mere existence on Meta’s platforms—as potentially infringing the law. In the U.K., the Digital Markets, Competition &  Consumers Act of 2024 bans advertising that is misleading. The European Parliament recently passed the new E.U. AI Act, which requires publishers of AI-generated content to label it transparently.

Creating a fake account to carry advertising is likely to be a breach of regulations in multiple European jurisdictions, Lester told Fortune. “That will be misleading. I would say, certainly, in the U.K. and I strongly suspect in the E.U. as well.”

A $10 billion business for Meta

Partnership ads like Ya Lily’s work by allowing creators to enable their posts to be used as ads if they recommend a brand and include a link where viewers can buy the product. If a marketer like Temu approves of the post, it can spend money on it as if it were an ad, and boost it in Meta’s algorithms. The creator gets paid either via a contractual agreement or a cut of sales generated.

Meta booked $10 billion in revenues from partnership ads in Q1 2026. Its total revenue was $56.3 billion—meaning that partnership ads were about 17% of its entire revenue. The company did not give an update on the partnership numbers for Q2, in which Meta disclosed $60.8 billion in revenue overall.

Within that $10 billion-plus partnership ad segment, Temu’s presence is huge.

Between January 2025 and April 2026, there were 31 official Temu pages on Meta and they ran over 9 million ads across the E.U. and the U.K., ORL’s data says. The cumulative reach of those ads was 134 billion views.

Of those ads, partnership ads (the ones using creator content) were 65.8% of all Temu’s ads and 49.6% of their reach. That is unusual, ORL believes, because among other advertisers, typically only 2% of their ads are partnership ads.

Burner accounts from China and Russia

And those partnership ads are channeled through a narrow group of creators, ORL says. In the four months between January and April 2026, 100 creators accounted for 74% of all Temu’s partnership ads in the region. Those creators ran more than 1.4 million individual ads that got a cumulative reach of 16.9 billion eyeballs.

Most of those accounts—which initially appear to be run by real people—are dedicated to pushing Temu products. But there is reason to believe they are “burner creators”—fake accounts that don’t represent real people that have been created specifically to take advantage of Temu’s ad budget. Seventy-three percent of the top 100 accounts had changed their handle at least once in the 16 months covered by ORL’s study. One of them changed its name 15 times—something real influencers never do.

Only eight of them were verified with a real identity. And, “despite targeting E.U. audiences, 28 of the top 100 accounts were based in Russia and 19 in China, while only six were based in the E.U. or the U.K.,” ORL said in a data pack it prepared for Fortune. Of the remainder, 34 were U.S.-based, and the others were from Bangladesh, Canada, and Ghana.

Meta has struggled to prevent fraudulent advertising activity on its platforms in the past, especially as it relates to China. Last year, Reuters reported, 19% of Meta’s entire revenue from China—about $3 billion—came from ads for scams or other banned content. The company initially investigated the problem and cracked down on it but later inexplicably reversed course and abandoned its efforts against Chinese scams, Reuters reported.

ORL has filed a report with the EU’s DSA regulator asking it to make a risk assessment of Temu’s army of low-grade influencer accounts.

“Lazy, not sketchy”

“If they [Temu] think this is a way to get slightly cheaper/more performant ads, they might be doing this intentionally to game the delivery system,” said Rob Leathern, a former senior director of product management at Facebook until 2020. He is now the founder and CEO of InfoHawk, a company that protects users from scams and deception.

Luke Stillman, managing director of Madison & Wall, a technology and media consultancy, told Fortune that Chinese marketers often behave very differently online than European or U.S. ones.  “It also wouldn’t be surprising for Temu to allocate a meaningful share of its Meta budget this way given how differently social commerce and the creator economy have developed in China versus Europe. In Europe and North America, this type of model is still relatively unusual. In China, social commerce accounts for a much larger share of activity, and creator partnerships and live social streams are much more integrated into how products are marketed and sold,” he said.

He thinks Temu might be being “lazy, not sketchy.”

“On one hand, I suspect there is a better form of an ad campaign that would result in higher [return on ad spend] compared to creating a huge volume of AI-generated creator content to feed through the partnership system,” he said. “On the other hand, dynamic content creation, creators augmented with AI production processes, and performance-optimization algorithms are only going to become more prevalent across the industry, and this seems to fit squarely into that bucket.”

This story was originally featured on Fortune.com

Tinned fish is having a real global moment. The pandemic-era pantry habit has since turned into restaurants nationwide serving tinned fish straight out of the can. Domestically, sales climbed from $2.3 billion in 2018 to more than $2.7 billion in 2024, and Americans’ obsession with sardines specifically is only accelerating: Americans bought $3.52 billion worth of shelf-stable seafood over the past year, and sardines have overtaken salmon as the second-most popular tinned fish behind tuna, as Grubhub reported orders of tinned fish tripled in 2025. The global tinned fish market was worth roughly $10.24 billion in 2024, and is projected to reach $17.97 billion by 2034. Canned sardines are a meaningful part of that worldwide push.

But the supply behind that growing appetite is shrinking fast. Morocco, the world’s top sardine supplier, saw its fish landings fall nearly 46% between 2022 and 2024, and the reasons are feeding into each other. Because sardines are unusually sensitive to temperature, warming water as a result of climate change is pushing the fish into new territories further away from fishermen.

The war in Iran, coupled with the effective shutdown of the Strait of Hormuz (through which one-fifth of the world’s global oil supply normally passes) is driving up fuel costs. Fishermen now need to go further to catch the fish, just as it’s costing them that much more to travel longer distances, and they’re bringing back almost half the quantity as before.

Rashid Sumaila, a professor and Canada Research Chair in interdisciplinary ocean and fisheries economics at the University of British Columbia, has spent his career studying fisheries, and how they navigate geopolitical forces. He said the combination is something he hadn’t seen affect the fishing industry in quite this way before.

“Sardines are very sensitive to climate change, and we know it here on the Pacific coast,” Sumaila told Fortune. “When it’s warm, sardines move from Mexico through the U.S., and Canadians see more sardines. They follow the temperature.”

He said Morocco’s sardines are shifting, too, and he saw the same thing while on a trip to Portugal this summer, when he left his hotel at 2 a.m. for an early flight and the temperature was still 34 degrees Celsius.

“Some of that 45% drop is partly climate change, and partly overfishing, and other things happening to the ocean,” he said. “There’s just so much. You have pollution, plastic. Stresses upon stresses.”

The Iran war adds a second layer. While Iran’s own waters are not a major sardine source, Sumaila said the war still reaches Moroccan boats through the price of fuel.

“The Iran war doesn’t quite impact this directly,” he said, “but indirectly it does, because prices are going up. The price of fishing, the vessels, the energy. It’s crazy.”

Since the conflict began, Brent crude surged 10% to 13% after the Strait of Hormuz effectively closed to shipping this spring, an event the International Energy Agency called the largest oil-supply disruption in the market’s history. Sumaila said he heard the same complaint from fishermen thousands of miles from the Persian Gulf.

“Thai fishers are feeling it,” he said. “You can imagine Moroccan fishers, or fishers in Africa or the Pacific Islands, feeling it too.”

A global supply food chain risk

Warmer water and higher fuel costs interact in a specific way for Moroccan boats, Sumaila said, because the two pressures compound each other. As sardines shift toward cooler water further from shore, fishermen who once worked familiar nearby routes now have to travel farther to find them, exactly as fuel becomes more expensive and the yield becomes less guaranteed.

“It’s more risky,” he said. “Insurance payments, security for people. The combination is really quite scary.”

“It’s the whole region. If the fish are moving up, that’s what we face,” he continued. “It’s like we’re facing a double whammy.”

The same pressures are also reshaping how Morocco sells what it does catch. Frozen sardine exports have fallen from about 70% of the country’s small open-sea fish exports in 2020 to roughly 23% in 2025, even before the war began.

“Climate change is a factor,” he said. “The war is going to affect it too, because all the freezing is energy, and energy prices are going up while the sources are being reduced.”

This all means we’ll be paying more to eat less, he said of his research of Canadian consumers’ grocery spending on seafood.

“This isn’t only the fish we catch in Canada,” he added. “It’s fish coming from all over the world. All of it gets affected. The price dynamics show up in people’s household budgets.”

He said affordability pressure isn’t limited to lower-income countries, even if it hits hardest there: “It’s not only in the developing world. We all feel it. I don’t go to any meeting these days where somebody isn’t talking about the cost of groceries.”

A fish called subsidies

Government fuel subsidies make the underlying problem worse, even if not all subsidies are bad. Public spending on fish stock research and fisheries management is useful, but paying for fuel is a different story.

“If you pay for fuel, that just makes people fish more than they would,” he said. His team’s most recently published estimate puts global fisheries subsidies at $35.4 billion in 2018, with $22.2 billion of that classified as capacity-enhancing subsidies that fuel overfishing, which shrinks fish populations further and forces boats to travel even farther for a shrinking catch.

Sumaila has a running joke about it, borrowed from an old British comedy about a heist gone wrong. “A fish called subsidies,” he said, referencing “A Fish Called Wanda.” “A third of your fish is actually subsidized and illegally caught.”

Outside of Morocco, both Portugal and Spain are trying to mitigate this as well. Portugal’s 2026 sardine season reopened in May with a national catch limit of 33,446 tonnes, nearly 1,000 tonnes fewer than the year before. Portugal closed its purse-seine sardine fishery for the final weeks of 2025 after its fleet worked through roughly two-thirds of the combined Portugal-Spain quota. Spain’s own 2026 quota came in nearly 3% lower than the prior year, yet the fishery still fetched stronger prices, with sardines selling for an average of 1.50 euros a kilogram, 50 cents above 2024 levels.

To combat the less fish but sold for more money scenario playing out in Spain, Morocco put an export ban on frozen sardines, meant to protect domestic supply and stabilize prices at home. Sumailia had little hope in this working, basing his prediction on the ongoing trade fights between the U.S. and Canada.

“When you ban fish, maybe in the short term you have some relief,” he said. “But that triggers a lot of other things you might not even perceive when you do it. You get a counter reaction, and a counter reaction to that.”

Most of all, he said, sardines are one of the primary food sources for lower-income individuals.

“When you shut off these consequences, you’re affecting small-scale people,” he said, referencing the emigration of fishermen from Morocco to Portugal. “People move. And then you have immigration. Where is this coming from? It’s coming from things like the war in Iran, too.”

This story was originally featured on Fortune.com

Three years after Warner Bros. abruptly canned “Coyote vs. Acme,” the Looney Tunes feature film that was given a new lease on life by Ketchup Entertainment debuted with an estimated $15.5 million in ticket sales over the weekend.

For the fifth weekend in a row, “Spider-Man: Brand New Day” topped the North American box office. It grossed $22.2 million in U.S. and Canadian theaters, according to studio estimates Sunday, bringing its worldwide total to $2.33 billion and making it the fourth highest-grossing film of all time.

“Coyote vs. Acme,” which had been relegated to the dust heap, fared best of the newcomers, debuting in second place.

The Ridley Scott action drama “The Dog Stars,” starring Jacob Elordi, flopped. The film, about a pilot and German Shepherd who survive a global pandemic, opened in fifth place with $8 million for 20th Century Studios. It added $11.3 million overseas, a poor result for a film that cost between $70 million and $100 million to make.

“Coyote vs. Acme,” however, scored a decent, hard-earned opening. In 2023, Warner Bros. canceled any release of the completed film, preferring to take a tax write off on a movie originally produced for its streaming service, HBO Max. Warner also shelved “Batgirl” and “Scoob! Holiday Haunt,” both of which remain unreleased.

“Coyote vs. Acme,” based on an Ian Frazier “Shouts and Murmurs” New Yorker column, is about Wile E. Coyote suing the Acme Corporation for its faulty products. It stars Will Forte, John Cena, Lana Condor and Eric Bauza.

The film, a hybrid of live-action and animation directed by Dave Green, cost $70 million to make. But despite good test screening scores, Warner Bros. opted to ax the completed film. After an outcry, the studio put the film up for sale. Early last year, Ketchup Entertainment acquired worldwide distribution rights for about $50 million.

For the makers of the film, just getting released was a victory.

“‘Coyote vs. Acme’ is finally, gloriously, in theaters now,” Green said on social media Friday. “Which means one thing. WE WON.”

Reviews and audience scores were also good. Moviegoers gave it an “A” CinemaScore. Critics were just as enthusiastic. On Rotten Tomatoes, it rated 95% fresh.

While the result might have counted as a disappointment for some $70 million movies, the opening weekend for “Coyote vs. Acme” was the best ever for Ketchup. The indie distributor’s previous high was 2024’s “The Day the Earth Blew Up” (a $3.1 million opening), another Warner-made Looney Tunes castoff.

“The box office numbers are fair, but the reception is excellent and the release should set up good ancillary business for kids’ viewing, where the film should recoup its investment,” said David A. Gross, who publishes the box office newsletter FranchiseRe.

“The Odyssey,” in its seventh weekend, slid to third place with $14.3 million. It remains unusually strong overseas, where the Christopher Nolan movie grossed $48.6 million over the weekend. Altogether, the Universal Pictures release has collected $1.55 billion in global sales.

Top 10 movies by domestic box office

With final domestic figures being released Monday, this list factors in the estimated ticket sales for Friday through Sunday at U.S. and Canadian theaters, according to Rentrak: