When Jim Dausch joined restaurant operator Yum Brands in late 2024, one of his first orders of business was to find a way to ensure Pizza Hut’s food was delivered as hot as possible.

Previously, the software that connected the chain’s kitchen and fleet systems processed orders in a rudimentary “first in, first out” flow. An order would come into the restaurant and a ticket would immediately be generated to tell the kitchen to put the pizza in the oven. But there were plenty of times where the the order would sit idle waiting for an available driver.

Dausch and his team created a data-forward automation layer that changed the workflow, telling cooks not to make the pizza until the system knew with greater certainty that further down the chain, a driver would be available for pickup. The change led to hotter food deliveries and a “meaningful” increase in customer satisfaction scores, according to Dausch.

“We are sort of step-by-step going through what it takes to run our restaurant and finding every way we possibly can to automate those things,” says Dausch.

Dausch joined Yum Brands, which recently saw sales take a hit from a cyclospora outbreak, in December 2024 as global chief digital and technology officer of Pizza Hut. He was promoted 11 months later to hold that same title across the entire enterprise, which includes the Taco Bell and KFC brands. He oversees Yum’s websites and apps, digital order platforms, corporate systems, AI and data, and restaurant technology across 63,000 global locations that are operated by around 1,500 franchisees.

Yum is Dausch’s first foray into the restaurant sector, but he says that his thinking around technology closely mirrors his 20 years of experience at hospitality giant Marriott. All of his technology investments focus on customers, workers, and the franchisees. For the franchisees, food and labor have traditionally been their largest expenses, but increasingly, they’ve had to increase their investments in technology. Still, Dausch says they have little appetite to just accept every new tool without a clear return on investment.

“When we talk to our franchisees, they’re very worried about that,” says Dausch. “If we can’t prove that what we’re putting in place is either going to meaningfully improve the customer experience in a way that drives higher same-store sales growth, or meaningfully reduce food waste in a way that is going to ultimately pay for itself, obviously the franchisee is going to kind of resist.” 

With ROI in mind, there are times when Dausch has to say no. Robotics have generated buzz among franchisees, but no single prominent use case has emerged that Yum deems worthy of chasing. He’s also cautious when buying the AI capabilities pitched by software-as-a-service vendors, saying higher chip costs and other infrastructure expenses have made pricing for these features too frothy.

Some bigger technology bets Dausch has placed include digital kiosks, which have been rolled out to around two-thirds of restaurant locations globally and consistently produce higher check averages than in-person orders. An automated, voice AI ordering system has been rolled out to more than 900 Taco Bell U.S. restaurants, also with the intention of boosting order sizes, while also improving accuracy and increasing customer satisfaction.  

Dausch acknowledges that the voice AI system has been a “learning journey,” requiring Yum to make tweaks to the system so that the handoff between the AI and human workers is smoother.  

Across the quick-service restaurant industry, kiosks have been one of the biggest tech hits with diners, but even there, Dausch sees an opportunity for improvement. He’s added a step where consumers can enter their loyalty program information so that kiosks can make more personalized offers based on the data the restaurant has from past orders.

At Yum, which ranks #474 on the Fortune 500, Dausch also oversees Byte, a proprietary SaaS restaurant technology platform that was designed to consolidate online and mobile app ordering, point of sale, kitchen and delivery, menu management, inventory, and labor management tools and systems. For now, Byte is completely an internal platform, though the intent is that it will have an external customer in Pizza Hut, which Yum agreed to sell for $2.7 billion in June.

Dausch credits former CEO David Gibbs, who retired earlier this year, for setting the vision that Yum would need to prioritize technology and AI to compete aggressively in the restaurant sector. A typical restaurant location was managing up to 30 software vendors, and it could take a day or longer to pull insights from some of those systems. Byte operates as a single platform with just one data source.

“The challenge was that the restaurants often did not have a common line of sight across all of those systems to how their business was doing in real time,” says Dausch.

One example of how Byte has helped improve restaurant operations has been in inventory ordering. Yum’s automation system has led to an 85% reduction in “stockouts,” which is when a restaurant would run out of ingredients, resulting in lost sales when menu items aren’t available.

Yum has authorized enterprise licenses for OpenAI’s ChatGPT for district managers and franchise leaders, who are also mandated to take courses through an “AI Academy” that teaches them how to understand prompting, create digital executive assistants, and encourage the development of more than 400 AI agents.

All that said, Dausch knows he’s only one piece of the pie. “I don’t know anyone that has ever just selected a restaurant to go to based on the technology,” says Dausch, who adds that craveable food at a fair price is what wins diners. “It’s important for us that we don’t lose the plot.”

John Kell

This story was originally featured on Fortune.com

Treasury data confirmed last night that U.S. national debt now stands at $40 trillion, with the government now expected to spend more than $1 trillion in interest on the debt in the fiscal year of 2026.

Debt hawks have been warning policymakers for some time that the nation’s fiscal path is unsustainable, and the issue is increasingly rising up voters’ agendas in the run-up to midterms later this year.

A new report from The Conference Board throws the issue into a new light for consumers: The potential impact on their personal finances if policymakers continue borrowing at the current pace.

The Conference Board modeled a series of scenarios: Baseline (using Congressional Budget Office data based on current trends), a good-case (in which federal deficits are cut roughly in half, in line with current targeting proposals), and a bad-case (in which deficit levels grow to 9% of GDP rather than the current 6% to 7%).

The Conference Board also modeled two financial crisis scenarios—a default and an interest rate shock—which economists like Bridgewater Associates founder Ray Dalio have long been concerned about.

Even dismissing the most extreme negative outcomes, consumers still stand to lose thousands if policymakers don’t act to reduce spending.

For example, the report models a family saving to buy a $600,000 house in either 5 or 10 years, with a 20% down payment and a 30-year fixed mortgage. The report does not provide a methodology for calculating rates offered in 2031 and 2036, but concludes that total payments over three decades for a home bought in 2031 come to $2.89m, and $2.8m in 2036.

These are the payments in the baseline scenario. However, under the good-case scenario, in which the government cuts its borrowing and interest is lower, this figure is reduced by $53,000 for buyers in 2031, or by more than $100,000 for buyers in 2036.

Consumers’ spending is closely linked to the debt picture, Michael Peterson of the think tank the Peterson Institute said in a conversation with Fortune this week: “When the U.S. borrows this much … that drives up interest rates, which then increases household expenses because your mortgage goes up, your car loan, your credit card bills, and inflation more generally. So [we] may not get a bill at the end of the month for national debt, but [we] are paying that bill both in the form of taxes as well as an inflated level of expenses.”

Peterson also said programs like Social Security and Medicare are running out of cash, placing further onus on government budgets in the near future. The trust fund for Social Security is due to run dry in a little under eight years, and Medicare in a little under seven years, according to estimates by the Committee for a Responsible Federal Budget.

The Conference Board added that when those coffers run dry, the Treasury will need to decide whether to backfill the expenditure from its general fund by $2.7 trillion, per the CBO—a further burden on its budget.

In the event payouts from these trusts are cut, workers approaching retirement face a hole in their expected earnings. The Conference Board reports the reduction in monthly benefits in 2032 would be $173. However, by 2033, when the trust runs dry, this increases to $705 a month.

In 2034, it represents a $721 hole, and by 2036, a $754 shortfall compared to current expectations.

The worst-case scenarios

The above scenarios are not based on a more pessimistic scenario in which the U.S. government defaults or a financial crisis ensues. Skeptics of this outcome have some grounds: The U.S. economy has the means to lower the value of its debt thanks to the Federal Reserve. Quantitative easing, although inflationary, would avoid the extreme fallout of a default.

Likewise, while Treasury yields are elevated at present, this is only in part due to concerns over fiscal trajectories. They also reflect long-term inflation expectations and traders’ guesses on whether the Federal Reserve will increase rates.

However, should either of these realities come to pass, the threat to households is severe. Total payments for the aforementioned home bought in 2031 rocket to more than $3 million in the case of a default, and over $3.6 million in the case of an extreme interest rate shock, per the report.

The report concludes: “Neglecting the problem will not make it better and worsening our deficits will only increase the negative impacts of the debt on the rest of the economy … Addressing the national debt deserves to be a high priority for both voters and lawmakers, to benefit all Americans.”

This story was originally featured on Fortune.com

In July, for the first time, Chinese developed models took all five top positions on OpenRouter, the neutral routing platform that has become the closest thing the AI industry has to a Nielsen rating. Xiaomi’s MiMo V2.5 ranked first by token volume, followed by models from DeepSeek, MiniMax, Alibaba’s Qwen family and Moonshot’s Kimi. Chinese models now carry more than 60% of the platform’s traffic, which exceeds 20 trillion tokens a week.

That is not a benchmark result but a usage curve.

A year ago, US models carried roughly 70% of OpenRouter’s traffic. Today they carry about 30%. Even more striking is that by mid-July, Chinese models accounted for a record 58% of tokens processed by American firms on the platform. US companies are not being forced into Chinese AI. They are choosing it, workload by workload, because the price/performance math is impossible to ignore.

The race split in two

Here is the paradox that should be on every board agenda this fall. American labs still hold the absolute frontier. GPT 5.5, Claude Fable 5, and Gemini 3.x lead on the hardest reasoning, long-horizon agents, and the most demanding enterprise work. The frontier gap is real and measured in months.

But the race split into two contests: capability and distribution. America is winning the first and losing the second. DeepSeek’s V4-Pro is priced at roughly one-twelfth the cost of GPT-5.5 at comparable benchmark performance. DeepSeek V4 Flash costs $0.14 per million input tokens, compared with $5.00 for GPT-5.5. OpenRouter’s own analysts report that Chinese open models run 60% to 90% cheaper than the leading American offerings. For high-volume production workloads, coding agents, document processing and customer operations that differential decides the purchase order.

Distribution is where ecosystems lock in. Alibaba’s Qwen family has passed one billion cumulative downloads and replaced Meta’s Llama as the most downloaded open model family in the world. Llama, which defined open weight AI in 2023 and 2024 has fallen below 1% of routed volume. Developers optimize what they can download. They build tooling around what they deploy. This is how Linux won servers and Android won phones, and it is happening again in plain sight.

Welcome to the death zone

Between the frontier and the commodity floor sits a death zone: any model, product or corporate AI strategy that is neither clearly the best nor clearly the cheapest. It is being crushed from both directions at once.

The market data shows exactly how this bifurcation works. According to analysis of OpenRouter’s usage data, Anthropic holds only about 12% of the platform’s token share yet captures roughly half of total spending. That is the premium lane with fewer tokens, priced for the work that justifies them. The commodity lane belongs to efficient open models moving trillions of cheap tokens. The middle, closed models without a decisive capability edge and enterprise deployments paying frontier prices for commodity work has no lane at all.

Most Fortune 500 AI strategic plans are standing in that middle right now. The typical enterprise signed one frontier API contract in 2024  routed everything through it, and never looked back. In 2026, that is the equivalent of running your entire logistics operation by overnight air freight.

China built this on purpose

None of this happened by accident. Export controls denied Chinese labs the largest GPU clusters, so they engineered around scarcity with token efficiency, novel attention mechanisms, efficient mixture of expert designs, higher quality data over raw volume and inference-aware architecture from day one. State support lowered the effective cost base further. Xiaomi cut MiMo API prices by as much as 99% in May.

Constraint now became strategy. American labs that prioritize efficiency as a secondary concern risk maintaining their technological edge while losing market volume, developer interest, and ultimately the whole AI ecosystem.

The builder’s playbook for 2026

For the executives and founders actually building on AI, four moves matter now more than anything else.

1. Make hybrid routing your default architecture.

Route the hardest, most regulated, highest stakes work to frontier models. Route high volume, cost sensitive tasks to efficient open models. Companies doing this are cutting inference costs 60% to 90% on the majority of their workloads without touching quality where it counts. If your AI budget runs through a single closed API, you are overpaying for most of what you do.

2. Treat efficiency as a first-class weapon.

Inference optimization, quantization, speculative decoding, and model hardware co-design are now standard practices rather than mere research curiosities. Study how the constrained labs built, and then apply those lessons with American compute behind them.

3. Differentiate above the model layer.

Proprietary data, application layer, domain fine tuning, agent frameworks and rigorous evaluation harnesses outlast any base model advantage. Base models are converging into infrastructure. Your moat was never going to be someone else’s model.

4. Get out of the middle.

If your product depends on a model that is neither the best nor the cheapest then pick a direction this year. Move up the capability curve with real differentiation, or compete hard on cost and openness. The middle does not survive 2027.

America needs an open weight answer now

My point of view is that Washington is preparing to fight the wrong battle. The instinct in Congress is to restrict Chinese models on security grounds, and for sensitive government and defense workloads, that caution is warranted. Data sovereignty concerns already limit Chinese hosted adoption across Western regulated sectors, though self-hosted open weights blunt much of that argument.

A ban is not a strategy, it’s a tariff on your own developers. Chinese open weights succeed not due to deception, but because they are high-quality, affordable, accessible, and no American lab currently releases frontier-class open-weight models on a regular schedule. Meta’s retreat left the field open and China took over quickly.

The answer is to compete with credible US and allied open weight models, released regularly and backed by procurement incentives or direct lab commitments. Open weights are how you export your ecosystem, your safety norms and your standards to the rest of the world. America understood this with the internet stack. America needs to remember it now.

The frontier still matters and the US should defend it. But the practical race in 2026 is won by mastering both contests at once with absolute capability and radical efficiency, closed excellence and open diffusion, the biggest reliable compute and the smartest use of it. Innovation under constraint should no longer be a consolation prize.

The question for the American C-suite, boardrooms, and Washington is the same one. When the next generation of global software is built, whose models will it be built on? Right now, the download numbers are answering. It is not the one America wants to hear.

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

Investors are wondering what Scott Bessent is up to. At the start of the month, the Treasury Secretary’s ‘to-do’ list included purchasing $5 to $10 billion in Japanese yen.

This week, he announced the Treasury would increase “by at least double” the size of buybacks for longer-dated securities—and is prepared to expand the “fiscal consolidation” of purchasing back the costlier debt.

Analysts are generally nonplussed. The Japanese yen—while stronger against the dollar than in its June slump—has unwound to roughly the level it started the year at. The drift back to market-perceived fair value is “hardly surprising,” quipped UBS’s Paul Donovan.

Likewise, analysts fear Bessent’s bond battle this week will amount to very little: “Despite a series of efforts to thwart bond vigilantes, we believe these measures will struggle to offset either declining Fed credibility or rising rate expectations,” the BNP Paribas Markets 360 team wrote Wednesday.

Long-dated Treasuries drifted down since the announcement, but they remain relatively elevated. As such, Daniel Casali, chief investment strategist at wealth management firm Evelyn Partners, suggested: “If policymakers are serious about capping long-end yields, more intervention may be required. Indeed, to borrow from the movie Jaws: “We’re gonna need a bigger boat …” Investors may conclude that this week’s buyback announcement is not the last one needed to stop yields rising higher.”

Bessent is apparently unimpressed by the lack of confidence. Speaking on CNBC, he suggested the Treasury is working beyond the market’s perception.

“People have bad information. I have asymmetric information,” Bessent said. “So I think that the market should think: ‘Why would we have joined the Japanese in the intervention at this time? Do we know something the market doesn’t know … in terms of being willing to do … what I would call a Treasury twist here, in terms of the bond market? What do I know that the market doesn’t know?’”

“So I think the market’s probably gotten a little ahead of itself, a lot of people have not much to do in August.”

Bessent added that further action on bonds will hinge on market reaction, maintaining that what the Treasury is “trying to do is get people to focus on the fundamentals and not trade the headlines during a quiet period in a thin market. So, we are trying to keep the market in equilibrium.”

Warsh and Bessent

On the surface, Bessent’s action seems at odds with the Federal Reserve’s strategy. New chairman Kevin Warsh has suggested tightening at the long end of the yield curve—the very thing Bessent is now trying to loosen—helps the Fed read market signals.

He said in July: “We’re seeing a tightening both in nominals and in reals, even while, at some level, we haven’t done much in 42 days, the markets have done quite a bit.”

Warsh has long been a proponent of central banks reducing distortion in markets, and has also signaled he would like to reduce the Fed’s balance sheet—potentially pushing up borrowing costs as a result.

Bessent responded: “The Treasury and the Fed would work together if there was any change in the balance sheet, and we would adjust to any kind of run-off that they’re doing.”

It might be tempting, on the surface, to see the Fed and the Treasury at cross-currents. However, Bessent and Warsh’s working relationship is clear: the pair continue the long-standing tradition of meeting for breakfast or lunch every week.

Moreover, while Warsh has been clear he wants to pull the central bank back to what he sees as its “lane,” he has been explicit that it is not the role of the politically independent Fed to stray any further, for example, into the work of the Treasury.

He told Congress last month: “The way we erode [Fed] credibility are two things: We wander outside of our lane into your lane, or into the lane of another executive branch, or we don’t deliver on our promises. The first thing we can do is to deliver on our promises, and the second thing is … stick in our lane. That’s what we’re going to do.”

This story was originally featured on Fortune.com

The longtime publisher of Stars and Stripes has announced his retirement, ending a decades-long career with the military newspaper as Defense Secretary Pete Hegseth’s Pentagon moves to exert editorial control and eliminate what it asserts are “woke distractions.”

Max D. Lederer Jr. announced his retirement, effective at the end of September, in a memo to staff on Tuesday, as well as in an interview with Stars and Stripes. In the memo, seen by The Associated Press, Lederer, the second full-time civilian in the position, wrote that it had “become clear that my philosophy of leadership, and my understanding of the value and mission of Stars and Stripes, differ in fundamental ways from the direction the leadership of the Department of Defense has for the organization.”

The announcement comes less than four months after the Pentagon fired Jacqueline Smith, ombudsman for the newspaper, whose job was to safeguard editorial independence. The news outlet is partly funded by the Defense Department, but has a legacy of independence from military and government leadership. “No one should be surprised that they’re kicking out the one person charged by Congress with protecting Stars and Stripes’ editorial independence,” Smith wrote in a column.

Stars and Stripes has recently been at the forefront of some of the reporting on conditions for sailors on the USS Abraham Lincoln and concerns for their mental health during the aircraft carrier’s long deployment — concerns that President Donald Trump has downplayed.

It’s part of larger, government-driven changes to the military publication

The Pentagon did not immediately respond to a request for comment on Lederer’s retirement, which comes seven months after the department announced in a social media post that it would essentially overhaul the newspaper to align with its current messaging.

Hegseth’s spokesman, Sean Parnell, wrote on on X in January that the Pentagon “is returning Stars and Stripes to its original mission: reporting for our warfighters.” He said the department will “refocus its content away from woke distractions.”

“Stars and Stripes will be custom tailored to our warfighters,” Parnell wrote. “It will focus on warfighting, weapons systems, fitness, lethality, survivability and ALL THINGS MILITARY. No more repurposed DC gossip columns; no more Associated Press reprints.”

More broadly, the developments at Stars and Stripes come against a backdrop of increased efforts by Hegseth to control media coverage of the department.

Last year, officials attempted to impose restrictions on journalists working inside the Pentagon, which, in turn, led most news outlets to turn in their access badges and walk out.

Then this summer, the Pentagon took another step and declared its press office a classified space — instituting a policy that journalists must be accompanied by an escort on Pentagon grounds. That policy is being challenged in court. In mid-July, a panel on the U.S. Court of Appeals for the D.C. Circuit said the policy could stand, overturning a lower court decision.

In her April column titled “The Pentagon is trying to silence me,” Smith, the former ombudsman, expressed concern over increasing restrictions on the media.

“For nearly a year, Pentagon leadership has placed more and more restrictions on the mainstream media,” she wrote. “The New York Times sued and when the Defense/War Department lost in court, instead of following the judge’s ruling Secretary Hegseth and company pivoted, finding another way to restrict journalists.”

Of Stars and Stripes, Smith wrote: “This newspaper has a long history of commitment to the military community and to journalistic values. Please don’t let it be controlled by Pentagon brass.”

The departing publisher objected to certain changes

Lederer was not made available for an interview with the AP. In the interview with Stars and Stripes, he said he did not feel the modernization efforts the Pentagon announced in March were appropriate, in particular a transitioning of print products to digital — a move he feared would limit accessibility for troops.

“I don’t feel that there’s a full recognition of the value of multiple platforms versus only a digital platform,” Lederer was quoted as saying.

In another development, Stars and Stripes said it could not confirm an account that, in recent weeks, a new deputy had been installed by the Pentagon under Lederer without the publisher’s prior knowledge.

In a letter seen by the AP, Rufus Friday, the chair of Stars and Stripes’ advisory board of publishers, wrote to congressional leaders about the arrival of the new deputy, a development he said “threatens Stripes’s editorial independence.”

Roughly half of Stars and Stripes’ budget comes from the Pentagon, and its staff members are considered Defense Department employees.

The outlet’s mission statement emphasizes that it is “editorially independent of interference from outside its own editorial chain-of-command” and that it is unique among news organizations tied to the Defense Department in being “governed by the principles of the First Amendment.”

The newspaper has been reporting about the military steadily since World War II, to an audience mainly of service members stationed overseas.

Lederer told Stars and Stripes that while his retirement is effective Sept. 30, he does not know whether the Pentagon will keep him as publisher until then or name an acting publisher.

___

Jocelyn Noveck writes about the intersection of media and entertainment for the AP.

This story was originally featured on Fortune.com

At 8 a.m. Eastern Time today, oil was priced at $95.29 per barrel with Brent serving as the benchmark (we’ll explain different benchmarks later in this article). That’s a loss of 11 cents compared with yesterday morning and more than $27 higher than the price one year ago.

Will oil prices go up?

It’s impossible to forecast oil prices with detailed precision. Many different elements affect the market, but ultimately it boils down to supply and demand. When worries about economic recession, war, and other large-scale disruptions increase, oil’s path can shift fast.

How oil prices translate to gas pump prices

Gas prices at the pump don’t only track crude oil. They also include what it takes to refine and move that fuel, the taxes layered on top, and the extra markup your local station adds to stay in business.

Since crude oil generally makes up a majority of the per-gallon cost, changes in its price have an outsized impact. When oil surges, gas prices typically rise in tandem. But when oil retreats, gas prices often lag on the way down, a trend sometimes described 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 and is more meant to provide temporary relief, assisting consumers and keeping critical parts of the economy running, like key industries, emergency services, public transportation, etc.

How oil and natural gas prices are linked

Both oil and natural gas are key sources of the energy we use every day. Because of this, a big change in oil prices can affect natural gas. 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

To gauge oil’s performance, we often turn to two benchmarks:

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

Between these 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:

• Venezuela oil minister to U.S. companies: ‘It’s an entire world waiting to be discovered’
• UAE halts all trade with Iran after new missile fire in Gulf
• Strait of Hormuz chaos has incentivized China to use an alternative trade route

Frequently asked questions

This story was originally featured on Fortune.com

Target reported a second straight quarter of comparable sales gains on Wednesday, saying that a merchandising overhaul under the retailer’s new CEO attracted more customers and boosted sales both in stores and online.

The mass-market discount retailer also benefited from a tariff refund of $994 million after the U.S. Supreme Court ruled this year that President Donald Trump overstepped his authority when he imposed double-digit import taxes on goods from most other countries.

There is a lot of interest in how tariff refunds from the U.S. government will impact retailers and whether those refunds will be used to lower prices for customers.

Chief Financial Officer Jim Lee, when asked about tariff refunds this week, said the company continues to invest in lowering prices. Target reduced the prices of more than 10,000 items over the past year and “there’s more to come even as we’re facing headwinds overall,” Lee said.

Comparable sales, those coming from stores and digital channels operating for at least 12 months, rose 3.8% in the second quarter. The company also upgraded its annual profit and sales outlook, citing the solid performance during the first half of the year.

Target is emerging from more than a year of weak comparable sales. It started off 2025 with a 3.8% decline, but recorded a 5.6% jump in the first quarter of this year. The second-quarter gain offset a 1.9% drop during the same three months last year.

Target CEO Michael Fiddelke, a 20-year company veteran who became chief executive in February, said the latest quarter was “an important step forward in the plan we laid out earlier this year to open a new chapter of growth for Target.”

Target also reported an increase in the number of customers going to its stores and shopping on its website from May through July.

“We’re encouraged by the progress made so far, and we’re also clear-eyed about the important work still ahead,” Fiddelke said.

In March, Fiddelke unveiled a $6 billion plan to reverse Target’s sales slump and to reclaim the retailer’s reputation as a place to go for affordable yet stylish apparel and home goods.

More than half of Target’s back-to-school merchandise is new, the company said.

That includes a limited-time collection of teen and tween clothes, school supplies and accessories in pastel colors and floral prints from the women’s lifestyle brand LoveShack Fancy. Target also collaborated with Hollister on a dorm decor collection.

Target recruited fashion designer and TV personality Isaac Mizrahi this summer to fill the newly created role of creative director at large. Mizrahi has been brought in to mentor Target designers, advise on product design and innovation, and forge new partnerships.

It is Mizrahi’s second partnership with Target. He became the first major fashion designer to collaborate with the retailer in 2003 for a successful run.

Fiddelke is also remodeling Target stores and improving staffing. The company has more than 100 full-scale remodels underway, with a goal of reaching 130 this year, Fiddelke said Tuesday.

During the second quarter, comparable store sales — sales from established physical stores — increased 2.7%, while increased same-day deliveries pushed digital comparable sales up 8.7%.

Target is one of the first big retailers to report second-quarter financial results, which could give industry analysts and economists another read on whether ongoing price pressures from the conflict in Iran impacted consumer behavior.

The Commerce Department released a report Friday showing weak retail sales in July. The University of Michigan’s consumer sentiment index, also released Friday, showed greater pessimism about the economy this month, likely driven by stubbornly high prices.

Target’s overall sales increased in all six of its main merchandising categories, led by double-digit growth in what the company calls Fun 101 — a division that includes consumer electronics, toys, trading cards, sports paraphernalia, books and gaming items.

Target’s beauty and food and beverage sales were also standouts. Target plans to roll out a new Target Beauty Studio concept next month in more than 600 locations. The new area, which will offer upscale beauty products and enhanced product expertise from staff, will partly replace its in-store shops with Ulta, which ended its partnership with Target this month, the company said.

Target executives said the company still was working to improve the assortment in its clothing and home goods departments, where sales hardly grew during the latest quarter.

Net income was $1.87 billion, or $4.11 per share, for the three months ended Aug. 1, easily beating the $2.34 per share that Wall Street had expected, according to a survey by FactSet. It also outpaced last year’s $935 million, or $2.05 per share. Yet this year also included millions from tariff refunds, which amounted to $1.65 in earnings per share, Target said.

Net sales rose 5.3% to $26.54 billion for the period.

Target now expects sales to increase 5%, up from its earlier predictions for a 4% gain. It also expects earnings per share for the full year to be in the range of $9.90 to $10.90. Analysts expect $8.52 per share for the year, according to FactSet.

In May, Target reiterated its guidance from March for earnings per share to be near the high end of $7.50 to $8.50.

This story was originally featured on Fortune.com

Walmart experienced the slowest growth in U.S. comparable sales in six years during its most recent quarter and it offered a cautious outlook for the year, sending company shares down 6% before the opening bell Thursday.

Comparable sales in U.S. stores, which measure sales at stores open at least a year along with online sales tied to those locations, rose 2.6% in the second quarter. They rose 4.1% in the previous quarter.

Excluding the wellness category that includes Walmart’s pharmacies, comparable sales increased 3.4% in the second quarter. Those sales were hit by federal legislation that requires pharmacies to dispense some high-cost Medicare drugs at capped prices, the retailer said. That was still below analysts’ projections of a 3.8% increase, according to FactSet.

Walmart’s U.S. e-commerce business, which has become an engine of growth for the retailer, rose 24%, trailing the first-quarter pace of 26%.

Walmart is among the first batch of major retailers to report second-quarter results, which could offer industry analysts and economists another read on whether ongoing price pressures from the conflict in Iran have impacted consumer behavior.

Walmart is considered a barometer of consumer spending given its vast customer base. More than 150 million customers are on its website or in its stores every week, according to Walmart.

That may draw even more attention this quarter after U.S. data released Friday showed that retail sales were surprisingly weak in July and a new read on consumers from the University of Michigan revealed growing pessimism about the economy, with so many Americans struggling with higher costs for gas, groceries and just about everything else.

The new figures from Walmart revealed the smallest gain in comparable store sales since a 1.9% gain for the quarter ended Jan. 31, 2020, according to FactSet.

That has broadened Walmart’s customer base and the retailer has begun capturing a larger share of wealthier Americans. The biggest gains in market share for Walmart are coming from households with annual incomes over $100,000.

Walmart’s quarterly net income was $6.37 billion, or 80 cents per share, in the three-month period ended July 31. Adjusted per-share results were 81 cents, easily topping the 74 cents Wall Street had expected, according to FactSet.

Sales rose 5.9% to $187.94 billion. Analysts were predicting $186.62 billion, according to FactSet.

For the third quarter, Walmart expects earnings per share of 62 cents to 64 cents. It projects sales to be up 3% to 3.5%. That would put sales in a range of $184.88 billion to $186.23 billion. The forecasts are below analysts’ expectations of 68 cents per share and sales of $188.19 billion, according to FactSet.

For the full year, Walmart now expects earnings per share to be in the range of $2.80 to $2.87 while sales should be up anywhere from 4% to 5%. That would mean a forecast for sales in the range of $741.7 billion to $748.8 billion, according to FactSet.

Analysts expected $2.90 per share and sales of $752.06 billion for the year, according to FactSet.

This story was originally featured on Fortune.com

Further escalating its battle with the Trump administration, ABC on Tuesday took the rare step of suing the Federal Communications Commission on First Amendment grounds, saying the agency’s demand for early review of its broadcast licenses posed an “existential threat” to the network.

In the lawsuit, ABC, its parent company Disney and the eight ABC-owned stations whose licenses are affected asked a federal court to stop the early renewal proceedings.

The FCC under Chairman Brendan Carr, an ally of President Donald Trump, had ordered the early review in April — itself a rare move — of all eight licenses owned by ABC, years before they are set to expire. Licenses are issued for eight-year periods. The FCC cited the network’s diversity and inclusion practices as a reason, but the development came shortly after a joke by ABC’s late-night host Jimmy Kimmel — an outspoken Trump critic — had infuriated the president, who has called for Kimmel’s firing.

“Again and again, the Administration has attacked ABC’s speech — the stories its journalists report and the viewpoints its network programs air,” the network alleged in its lawsuit Tuesday. “Over time, those attacks have escalated into express demands that ABC be stripped of its broadcast licenses because of its speech.”

“Facing this existential threat,” it added, “Plaintiffs have no choice but to seek redress from the judicial branch for the Administration’s blatant retaliation for their First Amendment speech.”

It said ABC had “no alternative means to eliminate these ongoing and immediate threats other than total capitulation to the Administration’s demands.” It asked the court to “immediately enjoin Defendants from taking or threatening to take any action against Plaintiffs in relation to the early license renewal applications.”

There are already clashes between ABC and the administration

The early license review is only one part of a long-simmering confrontation between ABC and the FCC.

The network has also been fighting Carr’s efforts to make the morning talk show “The View,” whose hosts and guests are often critical of Trump, subject to equal-time rules. That’s a question ABC says the agency itself decided — in the network’s favor — more than two decades ago. ABC argues “The View” is a bona fide news program, meaning it’s exempt from equal time rules, which require granting equal airtime to competing candidates for office.

The ABC lawsuit Tuesday also spoke of broad ramifications that go well beyond one network.

“The consequences of the Administration’s campaign against free speech reach well beyond ABC,” it said. “If the Administration gets its way, the message to every media company in the country will be unmistakable: tell only the stories the Administration deems favorable, or face the coercive machinery of the federal government. In such a world, the press could in no way be described as free.”

It added: “The FCC Chairman has left little doubt that this is his goal.”

The lawsuit also quoted Carr as saying, when Kimmel made comments that angered the administration, “We can do this the easy way or the hard way. These companies can find ways . . . to take action . . . on Kimmel, or there is going to be additional work for the FCC ahead.”

Carr, a longtime FCC commissioner, was named chairman by Trump in November 2025. He has indeed made it clear that he is considering revoking ABC’s licenses, or trying to, in what would surely be a drawn-out legal process.

Longtime free-speech attorney Floyd Abrams noted that tensions between the press and the commission are not new — but have never been this pronounced.

“There has long been a level of tension between the broadcast media, which seeks full First Amendment protection, and the FCC,” Abrams wrote in an email to The Associated Press. “But not until the Trump Administration has the government so directly, so deliberately and so dangerously sought to limit the freedom of the broadcast press to cover and discuss the news.”

The lone Democrat on the FCC praises the lawsuit

The sole Democrat on the commission, Anna Gomez, commended ABC and Disney for pushing back against the FCC’s actions.

“For months, the FCC has waged a campaign of censorship and control against Disney’s ABC stations,” she said in a statement, “using the threat of broadcast license revocations to punish a company for speech this administration doesn’t like.”

“I have long called on companies to push back against this kind of government intimidation, and I’m glad Disney has shown courage and stepped up,” she said. “This should be a welcome sign for every broadcaster who has felt the weight of this overreaching government pressure in silence.”

In late July, Carr defended his agency’s actions against ABC, saying broadcasters have a duty to “operate in the public interest.” The FCC, he said, was merely trying to restore that standard. Besides investigating ABC, Carr has also opened separate investigations into CBS News and NBC News.

Broadcasters like ABC, Carr said in an interview on the Fox Business Network, “struck a deal with the American people. You broadcasters get subsidized access, free access to a valuable public resource, the airwaves, worth billions of dollars. In exchange, you have to operate in the public interest.”

“Look, as a country, we should have a trusted, respected news media, and we’re not there,” Carr said. “So I hope more broadcasters return to their public interest obligations.”

On Tuesday, an FCC spokesperson reiterated that position.

“All broadcasters have a legal obligation to operate in the public interest — even Disney,” said a statement from the commission, responding to the lawsuit.

“The FCC has been examining claims that Disney engaged in illegal DEI discrimination for over a year. Disney is obviously very concerned about the FCC’s proceeding, as evidenced by their ongoing campaign of disinformation as well as their decision to ask a court to stop the FCC from further pursuing matters. The FCC will continue to follow the facts and law wherever they lead.”

___

Associated Press journalist Mike Catalini contributed to this report.

This story was originally featured on Fortune.com

Venezuela’s new petroleum minister sees her South American home not as a dilapidated former oil giant, but as an emerging energy economy ripe for U.S. and foreign investments in new oil and gas exploration, both onshore and offshore.

Paula Henao, who took over as the hydrocarbons minister in March after the forced U.S. removal of former leader Nicolás Maduro, told an overflowing Houston energy audience on Wednesday that Venezuela is much more than just its famed heavy-grade crude oil. There are more than 916 exploration opportunities awaiting foreign investment, she said, including natural gas and other untapped oil basins. She cited an estimated 192 trillion cubic feet of natural gas reserves, as well as the country’s world-leading proven oil reserves of more than 300 billion barrels.

“It’s an entire world waiting to be discovered, just waiting for us to reach these agreements so we can develop these new areas,” Henao said in Spanish to the crowd at the posh Post Oak Hotel in Houston.

Henao and leaders of the Venezuelan state oil company, PDVSA, were in Houston this week for meetings and a showcase event in advance of a bigger Venezuela Energy Week in February in Caracas.

“Go to Venezuela to invest, go to Venezuela to develop businesses there,” said PDVSA Vice President Jovanny Martinez, also speaking in Spanish. “We are at the right place at this historical moment. We have the energy that the world requires.”

After decades of cycling between energy reform and renationalization, including the most recent 2007 appropriation of assets from ExxonMobil, ConocoPhillips, and others, there’s still a lot of hesitancy to invest in Venezuela as it again changes its hydrocarbon laws in the aftermath of Maduro’s ouster. There’s a recognition that this could be the last great chance for the Venezuelan energy sector to thrive.

President Donald Trump has repeatedly insisted U.S. oil companies will spend more than $100 billion in Venezuela to dramatically rebuild its failing infrastructure but, apart from Chevron which never left, large U.S. energy companies are mostly taking a wait-and-see approach, despite Exxon expressing optimism. Others, such as BP and Shell, plan to invest in offshore Venezuelan gas fields near Trinidad and Tobago.

Otherwise, it’s a bevy of smaller, private U.S. oil producers jumping in first. A day prior to the Houston event, Venezuela signed new oil production agreements with the Dallas-based, private producer Hunt Oil and the major oilfield services firm SLB, which already works with PDVSA and Chevron in Venezuela. Hunt CEO Hunter Hunt said in a statement that the company is “proud to be one of the first American companies to sign an agreement with PDVSA to help expand Venezuela’s oil and gas production, and we are looking forward to expanding our presence in the country.”

Crossing continents

One of the next deals signed is expected to be with Denver-based Crossover Energy, which sees more upside in Venezuelan oil—both mature and exploratory oil fields—than in pricier shale oil and gas acreage in the U.S.

“Hopefully we can jump the line by taking a little more risk,” Crossover CEO Eric McCrady told Fortune at the Houston event. “We think that’ll open up more opportunities on the back end with more fields, and growth beyond what we have today.”

Crossover already has acquired a local Venezuelan operator to develop an on-the-ground presence and workforce and expects to sign new productive participation contracts (CPPs) with a “few days or a few weeks,” McCrady said.

The plan is to begin operating Venezuelan wells in January, he said, delayed a few months because of the devastating and fatal earthquakes that rocked the country in June.

“In the oil industry you’re always managing risks,” McCrady said. “I think the risks here are more above-ground—the labor force, equipment availability, the political situation—versus below-ground geologic risk, well failure risk, things like that. We’re comfortable taking risks. I think by being one of the leading companies to get in, it gives us an opportunity to hire the right team and hopefully get moving first so we have access to services and equipment.”

He said more work is needed within the country to build up its power grid, develop infrastructure to transport and process natural gas, and further tweak the laws for regulatory and contract certainty.

Since last year, Venezuela’s oil production has risen from just under 1 million barrels per day to more than 1.2 million barrels daily, an increase of almost 250,000 barrels each day. Largely led by Chevron, that increase primarily relied on optimizing existing oil wells, and not by bringing in new drilling rigs and teams.

Venezuela’s oil industry last churned out more than 3 million barrels daily at the beginning of this century and was still above 2 million barrels a day a decade ago.

Simon Sjøthun, a partner with the Rystad Energy research firm, said the world will need Venezuelan oil over time as existing resources run dry—especially with global oil demand projected to remain stubbornly high for decades—and that Venezuela could again exceed 3 million barrels daily by 2040.

McCrady is more optimistic, he said. He believes Venezuela can grow to 3.5 million barrels a day within five to 10 years, citing how quickly West Texas’ Permian Basin boomed to new heights in the last decade. Modern U.S. drilling techniques could do wonders in Venezuela, he said. “Venezuela has been isolated from the world stage for almost 25 years,” he said.

“With the right legal framework and bringing U.S. investment in, I think 3.5 million [barrels daily] will be reached a lot faster than 15 years. We see tremendous opportunity.”

This story was originally featured on Fortune.com

It’s no secret Meta is one of the wealthiest companies in the world. Last year, it brought in nearly $201 billion in revenue, and at the end of this June, it was sitting on more than $90 billion in cash and marketable securities. 

Now, a federal trial underway in Oakland, Calif., is testing what it would actually take to financially hurt a company that big. California, Colorado, Kentucky, and New Jersey have accused Meta of misleading the public about the risks its platforms pose to young users and of designing features on Instagram and Facebook that keep children and teenagers hooked. The four states are going first in a case brought by a coalition of 29 state attorneys general that sued the company in 2023.

Meta is already fighting child-safety lawsuits across the country, but this case carries an added threat because of who is bringing it. State attorneys general can bring claims that private plaintiffs cannot, including claims under the Children’s Online Privacy Protection Act, or COPPA. They can also seek remedies to address alleged harms affecting potentially millions of people.

“The stakes might be higher in this case because the damages awards are going to measure potentially many millions of people’s harms,” Eric Goldman, co-director of Santa Clara University School of Law’s High Tech Law Institute, told Fortune. “And there might be extra remedies because of the specific claims that the attorney general can bring.”

That helps explain the almost incomprehensible number hanging over the trial: $1.4 trillion. 

That’s how high Meta says potential penalties could climb under the states’ theory of the case, putting the theoretical maximum in the neighborhood of the value of the company itself. 

“It’s a number that boggles the mind, frankly,” Goldman said.

At its most extreme, Goldman said, the potential damages Meta has described could effectively transfer the value held by Meta’s stockholders to the public.

“Essentially, it’s asking Meta to turn in the keys and walk away,” he said.

Actually getting anywhere near that $1.4 trillion is another matter. The eight-person jury hearing the case is advisory, leaving U.S. District Judge Yvonne Gonzalez Rogers with the ultimate decision on liability and remedies.

James Grimmelmann, a professor of digital and information law at Cornell University, told Fortune he does not expect the bellwether trial to end with a penalty that bankrupts Meta.

“It’s always hard to guess with damage awards,” Grimmelmann said. “The jury is purely advisory, so whatever it concludes won’t be binding on the court, and even if it comes in with an extremely high number, the judge could revise it and so could other courts on appeal.”

New Mexico may offer a glimpse of what a major state-level financial hit could look like. A jury there found Meta liable for 75,000 violations of the state’s consumer protection law earlier this year, resulting in $375 million in civil penalties. A judge later found Meta’s platforms constituted a public nuisance and ordered the company to pay another $567 million toward addressing youth mental-health harms, bringing its total financial liability in the case to $942 million. Meta is appealing.

But the Oakland case is about more than how many zeroes Meta could be ordered to put on a check.

What Meta says the states get wrong

“The State AGs may call this a landmark case, but their limited claims are unsubstantiated and their financial demands are vastly disproportionate,” Meta spokesperson Stephanie Otway told Fortune in an emailed statement.

Meta argues the states have not shown anyone in their states was misled or harmed by the features at issue, and that the AGs are attempting to penalize the company for what it calls “industry-wide challenges like age verification,” Otway said.

“Rather than sticking to the facts or the law, the states have instead decided to chase an outlandish payout,” Otway said. “We stand by our record of creating strong protections for teens, and look forward to making our case in court.”

Less than 1% of Meta’s revenue comes from teens on Instagram, but Goldman emphasized the share of Meta’s business directly tied to those users doesn’t answer the central question in the case.

“The relevant question is how much harm is Meta causing in society,” Goldman said.

Goldman said millions of young people still use Meta’s services. If the states convince the court those users were harmed, the potential liability is not necessarily limited by how much revenue Meta directly makes from teens on Instagram.

And money is only one way Meta could lose.

The fight over how social media works

The attorneys general are challenging choices Meta made about how its platforms are designed and how content is presented to users. That distinction is central to how the case got this far.

Section 230 generally protects internet companies from being held liable for content posted by their users. The states argue they aren’t suing Meta over what users post; they are challenging Meta’s own decisions about how that content is presented to users.

Goldman doesn’t think those two things can be separated so cleanly.

“To me, that distinction is illusory. That makes no sense,” Goldman said. “You can’t separate out the editorial function and say we’re going to extinguish the content and the way it’s presented. Those are the same thing in my mind, but Judge Rogers disagreed, and that’s why this case has gotten to trial.”

Goldman also raised a First Amendment concern. He compared Meta’s decisions about how it presents users’ posts to the editorial choices a publication makes about which stories receive more prominence, like how large a headline appears or whether a story includes photographs. In his view, those decisions are themselves expressive choices protected by the First Amendment. Those arguments have not stopped the case from reaching trial.

The result of that fight could matter well beyond whether Meta pays hundreds of millions, billions, or anything approaching $1.4 trillion.

TikTok, YouTube, and Snapchat face similar litigation over alleged harms to young users. Goldman said a victory for the states in Oakland could provide a playbook for challenging how other social media platforms are designed.

And it may not stop at social media. Goldman pointed to lawsuits already testing similar theories against generative AI, video games, and social gaming.

That makes the potentially enormous penalty only one part of what is being decided in Oakland. Meta can challenge a damages award on appeal. A legal theory that survives the case can be picked up and used again.

“That’s why I say that the internet is on trial in Oakland right now, because it’s not just Meta and it’s not just social media,” Goldman said.

This story was originally featured on Fortune.com

CEO Agenda provides unique insights into how leaders think and lead and what keeps them busy in a world of constant change. We look into the lives, minds and agendas of CEOs at the world’s most iconic companies.

Ana Botín doesn’t have “normal” days.

As the decade-long executive chair of Santander, one of Europe’s largest banks by total assets, she’s always flying from one part of the world to another (when she chatted with Fortune, she was back in Madrid from a visit to Mexico, gearing up to head to Doha and New York in subsequent weeks). 

In 2014, when Botín succeeded her father, Emilio, the longtime chairman of Santander credited with making the bank a household name globally, she had big shoes to fill. Today, she’s one of the few women to lead a major bank, counting over 170 million customers. Yet she doesn’t let that burden get in the way of her jubilance. Santander’s blockbuster profits of €12.6 billion last year, up 14% from 2023, followed by plans to return €10 billion to investors through share buybacks over the next two years, may be part of the reason for this.

Santander spent the better part of the 21st century expanding its business and navigating Europe’s regulatory thicket. However, some of its business bets, such as its focus on digital banking, have started to pay off. Santander has recently benefited from a confluence of other factors as well, including higher interest rates following the COVID-19 pandemic, robust retail spending, and a strong performance from its investment banking business in the U.S. 

The Spanish lender has continued to multiply its customers, eclipsing the same figures at behemoths like JPMorgan Chase and Bank of America (as Botín pointed out to President Donald Trump at a World Economic Forum panel in January). 

FABRICE COFFRINI/AFP via Getty Images

Santander’s shares have more than tripled in value since the fall of 2020. On Tuesday, Santander’s market cap surpassed €100 billion, making it the first bank in the European Union to cross that threshold in the last decade, ahead of rivals like BNP Paribas and Intesa Sanpaolo.

Botín says experiencing Santander’s boom has left her feeling “like Jeff Bezos,” a fellow leader
who weathered a decade of stock market under-performance in the 2000s before investors globally recognized the tremendous value he’d created in Amazon.

Today, a key piece of Botín’s job goes beyond the bread and butter of banking; she needs to stay abreast of trends like AI and regulation. How she does that is with a trait she looks for in those she hires at Santander: a sense of urgency.

“When you want to change an organization of 200,000 people, it’s almost like changing a government,” Botín told Fortune. 

Botín may be the most celebrated female leader in Europe’s financial sector, but there’s more to her than a clear passion for the bank she leads: She’s a keen golfer and wearer of Zara jackets who maintains a policy of no email after 7:30 p.m. 

Her typical routine consists of wellness rituals that keep her active. She has also accepted that her lifestyle isn’t without its sacrifices, given her high-stakes gig: “I’m a very happy person, but I cannot enjoy my hobbies as much as I would like, because I need to be—literally—like an Olympic athlete,” she said. “If you have a mission that matters … it’s worth it.”

Following the Santander board’s €6.3 billion dividend announcement, we sat down with Botín
to discuss business, life, and more. 

This interview has been edited for brevity.

Down to business

Fortune: Which long-term trend are you most bullish about for society and the economy at large? 

The rise of AI and automation is going to be the defining trend of the 21st century and will be a disruptive force in society. It will bring increased prosperity in the long run but requires [us] to rethink how government builds new frameworks together with the private sector and academia that rewrite the rules of competition, taxes, education, and pensions, amongst others, to address disparities that exist already and will widen between different sectors of the economy and people. 

How can European leaders address the productivity gap with the U.S.? 

We are in an era of disruption, and we have to be honest with society about the scale of the challenge and the urgency of the need for change. To do that there are some quick wins, like focusing on reducing regulatory and supervisory complexity. But longer term, we must do much more to embrace innovation and enterprise, creating a business environment and culture that rewards smart risk-taking. And a new “social compact” is essential. 

Being productive

What time do you get up, and what part of your morning routine prepares you for the day? 

My usual morning routine involves waking up at 6 a.m., spending 10 to 20 minutes relaxing and drinking warm water with fresh organic lemons—very alkaline—followed by Americano coffee with cashew milk, and doing e-mails. 

I then do 45 minutes to one hour of cardio and weights, followed by breakfast: homemade gluten-free bread; avocado with a spoon of apple cider vinegar, which regulates insulin and glucose levels; olive oil; and protein—two eggs and/or turkey or sardines.

How late do you work? Do you continue sending emails during the night and on weekends? 

I try to stop sending emails after 7 or 7:30 p.m., both during the week and [on] weekends. I also try to not send emails on Saturdays, so we all can take a day off (not always possible). I generally use Sunday mornings to work on reading and writing, and Sunday afternoons (not always) to catch up on one-to-ones with the team. I follow the same routine every day and skip dinner at least five days a week.

What apps or methods do you use to be more productive?

My Olympic gold medalist nephew, Diego, introduced me to the Oura Ring four or five years ago, and I also use Fitbit to track cardio health [and] sleep. I measure daily what helps me sleep better so I am able to have the energy for sports and work and to enjoy my free time (there’s not much).

Oisin Keniry/R&A/R&A via Getty Images

Who is on your “personal board”—that is, who inspires and motivates you?

My family is the biggest inspiration for me: My mother, Paloma O’Shea, has given me a great education and has been an example of hard work and aiming high. She literally founded the Escuela Reina Sofía in a garage 25 years ago. Today it ranks alongside Curtis Institute and Juilliard as a top musical education school. The school’s orchestra will perform at Carnegie Hall in November.

My husband, Guillermo Morenés, has also been an amazing partner. We have three sons together and made a deal at the start that we would share responsibilities for the family 50/50. This support has been essential, and I could not have been able to balance career and family without him.

Finally, Seve Ballesteros, a two-time Masters tournament champion golf player, taught me golf and also how to trust yourself for those impossible Seve shots.

Getting personal

As a consumer, what is your favorite company and why?

Inditex, owner of Zara. I have a Zara €50 jacket that I got eight years ago, which I still use and gets mistaken for Chanel. This is my aim for [Santander’s digital] Openbank: a bank that works for everyone, from the young 20-year-old to the investment manager to the retired pensioner, that is also an “aspirational” brand, affordable but “cool” and fashionable. Zara’s ability to constantly innovate while maintaining its core values is truly inspiring.

And to end on a lighter note: What’s your favorite cuisine to cook and eat?

I love to cook a tortilla de patatas [a Spanish omelet]. The ingredients are simple: just eggs, potatoes, onion, and a little oil. Delicious.

CEO Agenda provides unique insights into how leaders think and lead, and what keeps them busy in a world of constant change. We look into the lives, minds and agendas of CEOs at the world’s most iconic companies. Dive into our other CEO Agenda profiles.

This article appears in the April/May 2025 issue of Fortune with the headline ‘CEO Agenda: A Q&A With Ana Botín’.

This story was originally featured on Fortune.com

A former engineering director at Meta who has testified before Congress about child safety on Instagram told jurors Wednesday at a landmark trial that the company took a “don’t ask, don’t tell” approach on kids under 13 on its platforms.

During his second day of testimony, Arturo Béjar said Meta consistently prioritized profits over safety in designing its products, focusing on how often and for how long people used them, even if it was detrimental to their mental well-being.

“If you step away from the product, they are not going to make any money,” he said.

The trial that began Tuesday in federal court in Oakland, California, pits Meta against the states of California, Colorado, Kentucky and New Jersey and is expected to last about six weeks. The four states were among 29 that sued the tech giant in 2023 over child safety and privacy — the other 25 will go to trial later. The company also faces lawsuits in state courts, including one underway in Tennessee.

The lawsuit accuses Meta of contributing to the youth mental health crisis by knowingly and deliberately designing features that addict children to its platforms and hide these harms from the public. It also argues that Meta routinely collects data on children under 13 without their parents’ consent, in violation of federal law.

Meta says users must be at least 13 years old to create an account, in line with the law, called the Children’s Online Privacy Protection Act, or COPPA.

“The attitude in particular on Instagram was ‘Don’t ask, don’t tell,’” Béjar said in response to a question about his perception of the company’s attitude towards users under 13.

The company has rejected the claims and said evidence at the trial will show its commitment to safety.

“You will hear over the course of this case a lot of important issues, issues like teen mental health, issues like social media, issues like how teens use social media,” Meta lawyer Paul Schmidt said Tuesday. “Those are important issues, and they’re issues where Meta believes that it has a responsibility. It has a responsibility to act on its own. It has a responsibility to try to work with teens and parents in partnership to try to address those questions.”

Plaintiffs’ witness says safety was an ‘afterthought’

Béjar worked at Facebook from 2009 to 2015, attracting wide attention for his work to combat cyberbullying. He returned from 2019 to 2021 as a contractor to work on safety issues. He testified before Congress in 2023 about social media and the teen mental health crisis, saying that Meta executives, including CEO Mark Zuckerberg, knew about harms Instagram was causing but chose not to make meaningful changes to address them.

In his testimony Wednesday, Béjar said employee performance reviews and compensation for those who worked on user-facing products were mostly focused on user numbers and how long they spend with those products.

“In that context, safety was an afterthought,” he said.

The states are seeking changes to user experiences for Facebook and Instagram as well as financial damages, which could include billions of dollars in penalties. In a written statement, the office of California’s attorney general said if Meta loses, the amount of any damages would be set by the court.

“This case is about stopping Meta from offering a dangerous product to teens, and from lying to teens, families, and the public about the dangerousness of their platforms. The primary remedy under our state consumer protection law is an injunction,” the statement said.

Béjar walked through Meta features he said were designed for adults and are “inherently unsafe for teenagers.”

This includes video autoplay, which can mean teens see videos that may cause them harm even if they don’t click on them; as well as various counters that track how many people liked, viewed or commented on your content or how many followers you have.

Child development experts have noted teenagers are more susceptible to social comparison than adults, so products that reward popularity can be more harmful to their mental health.

Age verification has been criticized for not going far enough

Despite Meta’s statements that it works to find kids under 13 on its platforms and ban them, Béjar testified that he found “tens of thousands” of kids under 13 on Instagram through his research. He said it was “common knowledge” at the company that such young children were on Instagram.

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

Over the years, Meta has introduced features it says are designed to make the experience safer for young people. But Béjar said these did not work.

For instance, a tool called “Take a Break,” introduced in 2021, is a feature that is “designed to fail,” he said. First, it is a setting that people have to turn on if they want to use it. Béjar said that in his experience building settings, very few users actually go through the trouble of turning them on. He compared it to an airbag that drivers have to turn on every time they get in a car.

“A safety tool has to be on by default,” he said.

The feature can also be dismissed with a tap of a finger. If Meta was serious about wanting users to take a break, Béjar said, it would not be so easily swiped away.

___

AP Technology Writer Kaitlyn Huamani contributed to this story from Los Angeles.

This story was originally featured on Fortune.com

Consumers are getting fed up with AI slop found on online marketplaces, and companies are beginning to take note.

About a year ago, online 3D model marketplace CGTrader introduced the ability for designers to upload AI-generated assets for purchase on the platform, in addition to the digital models they rendered themselves. The site has more than two million 3D models for sale, which serve as the foundational component for architects, video producers, game designers, and other creatives to build their product around.

But CGTrader may be a case study for how having more AI-generated products for sale does not guarantee the technology’s popularity, and why buyers still favor human-made goods. A recent report from the company found that despite one in six models uploaded to its platform being AI-generated, those assets accounted for just $1 out of every $90 in generated revenue, and just 2.6% of sales.

“AI is entering the catalog rapidly, but buyers aren’t yet opening their wallets for it,” the report said.

The report, which drew data from marketplace sales between June 2025 and May 2026, noted that only 5% of CGTrader’s customers tried an AI model and found it worked well, as compared to the 20% who tried it and found the assets inadequate. 

CGTrader CEO Dalia Lasaite pinpointed the reason why the company’s customers turned away from its AI offerings: It’s not that they hate AI; it’s just that they valued what humans had to offer more—not least of all because humans simply make better stuff.

“Buyers are looking for really high quality when they are shopping at the marketplace,” Lasaite told Fortune. “And as a result, they tend to prefer human-created 3D models, at least at this point.”

As AI adoption increases, consumers’ feelings toward the technology, particularly its application for creative uses, has become tangled. A 2025 Stanford University study found that when participants were given access to an online marketplace with both AI-generated and human-produced art, they gravitated toward AI-generated pieces, with the number of generative AI images on the platform rapidly increasing. However, a Pew Research Center poll last year found half of Americans said they liked a painting less after learning it was made by AI. In a report published on Tuesday, Pew found 52% of American adults were “more concerned than excited” about greater AI use in data life, as compared to 38% who said the same thing in 2022.

But Dennis Zhang, a professor of marketing and supply Chain, operations, and technology at Washington University in St. Louis’s Olin Business School, said more AI-generated products in marketplaces reflects more than just how people feel about AI right now; it also hints at the role AI could play in the economy more broadly.

“One side of economists always tells you, ‘Don’t worry about AI. For every technology revolution in human history, people re-pivot to something else to do,” Zhang told Fortune. “What we’re saying is something else: It’s not only people as workers will re-pivot to something else to do, it’s also people as consumers will re-pivot to the dimension that humans will matter more.”

The rise of AI in the marketplace

In his recent working research, Zhang measured the proliferation of smartphone app launches after the wide release of coding agents Claude Code and Codex. He initially found that compared to 2023 and 2024, the number of apps launched steadily increased, a trend that continued through 2026. But additional analyses controlling for other variables found that the impact of coding agents on app production was about a 160% increase in apps by April 2026 compared to the period two years prior.

Then Zhang looked at how people were engaging with this influx of apps on the marketplace. The number of apps with more than 10 reviews dropped significantly after the AI launches, suggesting people engaged less with AI-generated apps than human-made ones. These results were not causal.

“There is some slight evidence showing that the products that are helped by AI in production are less attractive than the products where we had observed before, where it’s mostly human-crafted on the coding side,” he said. “However, it’s not like the AI products are unloved by everyone, right? It’s still creating utilities for the market.”

Looking more deeply, Zhang hypothesized that for apps where humans still had a larger hand in the concept and development for the app, increased unpopularity could be simply because the apps aren’t as soon as the fully human-generated ones, which were likely developed by programmers who have been in the industry longer, and are therefore more sensitive to factors like user interface. In other words, AI has enabled more vibe coders to design more apps, but lack of experience means those apps just aren’t as good; it’s a labor issue.

On the other hand, for apps that are obviously completely AI-generated, consumers may have snubbed them because they value product scarcity and are seeking out tools with human-added value; it’s a consumer psychology issue.

Put together, Zhang posited, these attitudes toward AI-generated products can begin to paint a picture about the future of how automation is integrated into work and society: “I would actually think people’s affection or judgments of products is going to shift from the parts which are created by AI to the parts which are less likely to be created by AI,” he said.

Zhang sees evidence that AI will transform labor, not largely displace jobs. While how AI is being deployed in the workplace informed this view, he likewise believes that how consumers respond to AI in the marketplace—not completely eschewing the technology, but rather valuing human touches—affirms humans’ place in the economy.

CGTrader CEO Lasaite came to a similar conclusion. When AI was first introduced in 3D modeling, creators were apprehensive, she said, but that sentiment has slowly changed as AI-generated models became faster and cheaper to produce.

“Over time, we all realized that AI will be some kind of part of our life, and we adapt,” she said. “Maybe we can be more productive and just keep the best parts of our job to ourselves, and use the AI to help with the rest.”

This story was originally featured on Fortune.com

Before Boeing named Kelly Ortberg as CEO in August of 2024, the airplane-maker was an enterprise in crisis, and faith was fading that arguably the most iconic of American manufacturers would ever regain its lost luster. 

Just as Boeing was slowly recovering from the Lion Air and Ethiopian Airlines 737 Max crashes in 2018 and 2019 that killed 346 passengers and crew, a Max door-plug blowout over Portland, Ore. in January of 2024 trained the spotlight on its manufacturing practices, which had increasingly put profits over quality. Federal regulators cracked down, freezing Max production at a third below its prior peak. Boeing’s defense and space division, meanwhile, was booking multi-billion losses on federal contracts spouting big cost overruns. 

The laundry list of problems got longer: The challenge of integrating stricken Spirit AeroSystems, the fuselage supplier Boeing had sold two decades and just agreed to re-acquire, greatly upped its risk profile going forward. To make matters worse, Boeing was facing a potentially crippling strike from its powerful union of 33,000 machinists in the Puget Sound area, whose leaders claimed that since management couldn’t do it, the rank-in-file needed to “save Boeing from itself.”

The job looked so tough that Boeing struggled to find a taker. Among the marquee names the airplane colossus reportedly courted, sans sale, were CEOs Larry Culp of GE Aerospace, Dave Gitlin of Carrier Global, and its own chairman, Steve Mollenkopf, the former Qualcomm chief.

Ortberg was a total dark horse. He’d served successfully as CEO of aerospace and defense manufacturer Rockwell Collins for five years. United Technologies acquired Rockwell in 2018, and then sold to RTX less than two years later. Soon thereafter, Ortberg retired. By the time he took the Boeing job, Ortberg hadn’t filled an operating role for over four years. 

Ortberg’s demeanor is so understated, and he keeps such a low public profile, that the scale of his achievement since then hasn’t gotten the kudos it deserves—but that’s beginning to change. Put simply, Boeing’s en route to one of the most dramatic, and quickest, comebacks on record for a formerly ailing corporate giant. “Boeing found its change-agent in Ortberg,” says Scott Mikus, an analyst at Melius Research. “Thanks to Ortberg, the dream of a great industrial company is still alive.” 

A seasoned engineer

Ortberg brings the right stuff as a seasoned engineer who features a history of promoting smooth labor relations. He’s reinstated “an engineering-first” culture at Boeing, a sharp departure from the falling interest and lack of investment in innovation, and focus on share buybacks, that reigned in the pre-Max-crash period from 2014 to 2018. Ortberg’s following a nothing-flashy, back-to-basics, step-by-step approach that targets big improvements in quality, reliability and on-time delivery. He’s also going for win-win agreements with suppliers, in contrast to Boeing’s former penchant for antagonizing partners by severely gouging them on pricing.

As a young engineer, Richard Safran––an analyst at Seaport Securities––saw Ortberg in action at Rockwell Collins. “That’s why unlike most people, I wasn’t surprised by his coup at Boeing,” says Safran. “He’s a Midwesterner who takes the ‘no decision before its time’ approach. He’s methodical about checking all the boxes one after another. He’s such a good engineer that he knows just enough about everyone’s job to be dangerous. And he knows how to make money.”

A person who’s seen regimes come and go, and worked alongside Ortberg at Boeing, marvels at the shift in culture. “He’s set a new tone in the place,” says this observer. “He measures people not just on what they do, but how they do it. You need to reach out and get feedback from colleagues. His approach calls for tying pay and promotions to how people treat and respect one another, in addition assessing their work. Are there still people in senior places who don’t treat people well? Yes, but it’s a good start.” 

This individual also stresses that Ortberg’s own people skills set the template: “He’s a good listener with high EQ. His theme is getting Boeing back to what it needs to be.”  Ortberg’s also renowned for high expectations that colleagues are always well prepared when he quizzes them about their businesses. 

What’s particularly remarkable about Ortberg’s turnaround is that it faced an almost instant hurdle: Within a month of his arrival, the mechanics strike sent production of its best-selling 737 MAX fleet from the already FAA-reduced cadence to virtually zero. Ortberg took a typically conservative stance, raising over $24.3 billion in new capital to cover the coming losses and bolster Boeing’s balance sheet. 

He resolved the stoppage in a relatively fast 53 days, and a string of victories quickly followed. In March of last year, Boeing won the Air Force’s Sixth Generation fighter program in a stunning upset over the competitor that previously cornered the market, Lockheed Martin. The contract opens the way for a new era of profitability in the defense and space sector: After booking an operating loss of over $5.4 billion in 2024, the legacy of grossly underbidding on military aircraft initiatives, the division turned slightly profitable last year, and in Q1 of 2026, earned $233 million for a resurgent operating margin of 3.1%.

In commercial aircraft, Boeing’s largest franchise by far, the campaign to revamp manufacturing safety measures began in the post-crash period, under the close supervision of the FAA. But Ortberg’s relentlessly systematic approach hastened the progress, and the results are now showing in a big way. 

He’s managed to get the FAA cap on the Max, Boeing’s workhorse aircraft, lifted from 38 to 42 a month, and expects to exit 2026 sending 52 off the assembly line, around the peak number eight years ago. Due largely to the jump in Max output and deliveries, Ortberg predicts that Boeing’s heading to $10 billion in free cash flow. Though he doesn’t provide a date, both Mikus and Safran believe Boeing will hit that milestone in the 2028 timeframe. And on the Q1earnings call, CFO Jesus Malave stated that Boeing’s aiming higher. “I think the potential for our cash flow supports being above $10 billion,” said Malave. Getting beyond that figure would take Boeing back to near its top numbers ever in 2017 and 2018—but back then profits roared largely via curbs in R&D and workforce as a share of sales, strategies that robbed from the future.

Ortberg wins high praise from airline customers. “Boeing’s doing a pretty miraculous job of turning around,” United Airline CFO Michael Leskinen said recently. “Our confidence that our Max aircraft will be delivered on time has never been greater during my [over eight year] tenure at United.” 

Still, Captain Kelly faces big challenges in getting Boeing’s wings full level for maximum speed of ascent. Boeing still suffers from ongoing supply chain, quality and certification issues, though they’ve declined. For example, wiring problems on the Max have pushed deliveries scheduled for Q1 into Q2, and a shortage of business class seats is delaying output on its widebody stalwart, the 787. 

Up ahead: Labor challenges, and a new plane

A crucial test looms in October: Boeing’s contract with its 16,000 engineers, which predates Ortberg, is expiring. It’s essential that Ortberg, the engineer’s engineer, secure an agreement that satisfies all parties, and avoids an extremely lengthy strike, as he did with the machinists. “That would send a message that Boeing’s cultural transformation is real,” says Mikus.

Indeed, Boeing will need the world’s best engineering talent to develop an all-new plane that will match if not beat Airbus in narrow-bodies where the Max and A220s and A320s play, and that comprise the biggest airplane class. Since 2010, its archrival has captured around 60% of that market, chiefly due to the superior range of its A320neo and A320XLR families. Ortberg has stated that Boeing must wait until the technology’s right before committing to the crucial new design that will largely chart its future. A major part of that process will involve choosing a highly advanced engine from GE, RTX, or Rolls Royce that delivers both big fuel savings of around 20% and greater longevity that will curb the high repair costs on the current versions. 

The big question: Will Ortberg, whose caution has so far worked well, move fast enough? “By concentrating on getting cash flow up, are they crowding out next-gen aircraft development?” queries one industry veteran. By contrast, Airbus has been highly aggressive in collaborating alongside GE Aerospace in testing the so-called RISE engine—which, in part by removing the nacelles that enclose the fan blades, fashioning the blades from super-strong, lightweight carbon fiber and making them longer, could achieve new frontiers in energy efficiency. 

But Boeing also harbors an ace: the new chief of commercial aircraft development Brian Yutko. The appointment of Yutko, an MIT PhD in aeronautics who at around age 40 stands among the world’s top experts in revolutionary airplane design, signals that Boeing will be carefully weighing all of the most-avant garde options on the market, and decreases the risk the rebounding giant will move too late.

 By Wall Street’s best estimates, the earliest Boeing could commit to a new greenfield plane is 2029 or 2030, with production coming around 2037. Keep in mind that Ortberg just turned 66. “He took the job at an age when most top executives at retiring,” says the aerospace insider. Indeed, Ortberg could stay at the controls for several more years, and even make the call on the all-new plane. 

But for the Boeing board, job one is setting a succession plan, and it will have a jumbo-sized presence to replace. Fortunately, the directors will hold a far stronger hand than when it recruited Ortberg. Then, things were so bleak that even Boeing’s immense size and vaunted legend wasn’t enough to lure the top brand, practicing CEOs. This time, the job’s going to be a lot more attractive. Credit the unlikely pick who fit the times: Kelly Ortberg.

This story ran in the June/July 2026 issue of Fortune as part of a feature called “Innovation Giants on the Rebound.” For more Fortune 500 innovation stories, click here.

This story was originally featured on Fortune.com

Everyone agrees that someone is calling the shots on a corporate AI strategy. It’s just that a lot of executives aren’t clear on who that is. 

Only 34% of C-suite executives in a new Pearl Meyer survey said it’s consistently clear which executive or team makes calls about AI, which was the lowest of all the cohorts polled. Among corporate board members, the figure rose to 53%. But when the question was asked of senior managers and professionals below the C-level, the figure rose even higher to 57%. 

Essentially, the group of executives likely to be blocking and tackling on-the-ground AI implementation are the least convinced that anyone is clearly owning the decisions and results. Those who are furthest away from owning the messiness of implementation, are more likely to feel the matter is settled. The survey found 78% of executives below the C-suite report that their companies have the senior talent required to effectively implement and oversee AI across the whole company. 

Significant gaps between the uppermost rungs at companies and other executives run through much of Pearl Meyer’s Q2 2026 Market Intelligence Survey. The poll of 116 board members, CEOs, C-suite execs, and senior managers below them was conducted in May and June and shared exclusively with Fortune ahead of its release on Thursday. 

The results, which show that AI deployment isn’t going as smoothly as some CEOs had hoped, comes at a high-stakes and expensive moment. Total AI spending, including capital expenditures on AI infrastructure, is poised to reach $2.5 trillion this year, a 44% increase in spending over last year, according to research and advisory firm Gartner. Next year spending is projected to rise to $3.3 trillion, the firm found. At that level of investment, CEOs know their heads could be on the chopping block if they fall behind relative to competitors and if they fail to deliver at their own companies. A survey of 900 CEOs published in May revealed that 80% of U.S. CEOs think their job is at risk if their AI projects wither on the vine, while 81% believe a fellow CEO will be ousted due to an AI failure or crisis. 

“Ambition for AI outcomes is currently outpacing the leadership structure needed to deliver on them,” the Pearl Meyer study, published on Thursday, states. “Additional investment without clear ownership will only widen that gap.”

But at this point, expectations for how much of an impact AI will have on individual companies doesn’t seem rooted in how much progress has been made in implementing it. Brad Jayne, a principal at Pearl Meyer and co-author of the study, said confidence that AI will deliver significant gains within 18 months holds at about 50% among leaders of companies at every stage of maturity including the pilot phase, experimentation, enterprise-level deployment, and companies that haven’t started anything yet. 

“There’s an impact-versus-speed tension,” said Jayne. Handing out licenses for ChatGPT or Copilot is quick and easy, he said. “Building big systems around that and pushing them through the organization and making sure it’s not making errors, that takes a lot longer.”

CEOs may also be overly optimistic when it comes to how close to burnout their employees are. When asked if employees could tackle additional organizational change without feeling stretched too thin, with AI implementation as an example, 63% of CEOs responded affirmatively, with only 33% of the C-suite and 40% of non-C-suite executives in agreement. 

Boards, for their part, may be in the dark as to how much more change is coming. Asked whether achieving strategic goals will require significant changes to how the organization operates within three years, 88% of CEOs and 79% of C-suite executives said yes. Only 42% of directors agreed. 

Coupled with the change-fatigue response, said Jayne, “that’s an alarm bell for me.”

“The board is basically saying, ‘We’re good. We’ve made investments, we’re structured right, go make changes,’” said Jayne. “And the management team is saying, ‘Whoa, whoa, whoa. In order to be effective here and get our strategy done, we’re going to have to make big changes in how we operate together.’”

Jayne’s worry is about what happens when spending has to be justified in a year. If boards and management teams can’t successfully connect AI spending to outcomes investors can recognize and appreciate, there could be problems. 

“I worry about finger pointing,” said Jayne. It could be culture, learning agility, or insufficient experimentation with AI tools. 

“It might come to some turnover,” said Jayne. “I think we’re in for a bumpy ride.”

This story was originally featured on Fortune.com

Meta is once again on trial over dangers its platforms may pose to children. It is once again arguing that it works tirelessly to keep them safe.

A pivotal trial for the parent company of Instagram and Facebook kicked off in a California federal court Tuesday, with four states seeking billions of dollars in damages as well as fundamental changes to how Meta runs its platforms.

A jury will decide whether the states’ attorneys general have made their case that the tech giant designed its apps to “hook the users, hold them for as long as they can, harvest their data and hide the truth from the public,” as Megan O’Neill, a deputy attorney general for the California Department of Justice, put it in her opening statement.

California, Colorado, Kentucky and New Jersey were among the 29 states that sued the tech giant in 2023 over child safety and privacy — the other 25 will go to trial later. The company also faces lawsuits in state courts, including one underway in Tennessee.

The lawsuit accuses Meta of contributing to the youth mental health crisis by knowingly and deliberately designing features that addict children to its platforms and hides these harms from the public. It also argues that Meta routinely collects data on children under 13 without their parents’ consent, in violation of federal law.

The company has defended its safety record and said it has a strong case.

States argue Meta exploited research on child development

O’Neill laid out the states’ case for the eight jurors during opening statements in federal court in Oakland, California.

“You’re going to hear that Meta knew a lot about kids’ brains,” O’Neill said. This includes how they are constantly seeking rewards, how they are sensitive to social feedback and “how they are still developing their ability to control impulses the way adults do.”

Meta, she said, researched these vulnerabilities and talked about how it could change Instagram to respond to them.

“ ‘The young ones are the best ones’ is a title of a Meta study we are going to show you,” she said, telling jurors that for Meta, “kids are the product.”

Meta, O’Neill said, also knew that kids under 13 were using its products even though they were banned from it, and “failed to take the simplest most obvious steps to keep them off.” For instance, she said, when Meta found that someone on Facebook was under 13, it would disable the person’s Facebook account but not their connected account on Instagram.

The states called as their first witness former Meta executive Arturo Béjar, who worked as an engineering director at Facebook from 2009 to 2015, attracting wide attention for his work to combat cyberbullying. He returned from 2019 to 2021 as a contractor to work on safety issues.

Béjar said that, contrary to Meta’s statements, the company’s research on safety issues was not being used to improve products.

For example, regarding eating disorder content, Meta’s engineers “had very good ideas on how to make it better” so that users, especially young ones, would not be exposed to it.

“But once it was reviewed, it got whittled down to a little pebble that didn’t make a difference,” he said.

Meta attorney highlights company work on safety

Meta lawyer Paul Schmidt laid out his case beginning with what is not disputed in the trial — that even though kids under 13 are not supposed to be on its apps, some lie about their ages; that some teens struggle to manage their time; and that some people post “negative content” on Facebook and Instagram.

But he said he will focus on the work Meta does to make its platforms safer and share information with the public.

“Much of this lawsuit is about the government attorneys and their witnesses saying in trying to improve, we’d do it a little differently,” Schmidt said. “In talking about how to improve, we disagree with how you talk about it. It’s meaningful, and the evidence will be meaningful, these efforts Meta has taken to improve.”

U.S. District Judge Yvonne Gonzalez Rogers in Oakland is overseeing the proceedings, which are expected to last six weeks with testimony from Meta CEO Mark Zuckerberg and other executives and former employees. Gonzalez Rogers, appointed to the bench by President Barack Obama in 2011, has overseen a bevy of complex, high-profile cases involving Big Tech. These include Elon Musk’s lawsuit against OpenAI and its founders as well as Epic Games’ lawsuit against Apple over its app store.

If Meta loses the trial, the court would have wide discretion over the size of any financial penalty. Meta has said if it loses, the case could leave it liable for damages amounting to $1.4 trillion, but legal experts say anything close to that amount would be unlikely.

Meta faces thousands of lawsuits, angry parents

The trial is the latest in an avalanche of lawsuits against Meta Platforms and other social media companies including Google’s YouTube, TikTok and Snap, over arguments that their platforms harm young people, illegally collect their data and are deliberately designed to addict them.

As the trial began, child safety advocates and parents who trace their children’s deaths to social media harms gathered outside of the courthouse Tuesday. Many parents held photos of their late children as they spoke to reporters outside, at times speaking through tears.

Several parents held a banner with the names and ages of children who died from social media-related harms. It was several feet long.

Mary Rodee, whose son Riley Basford died by suicide at age 15 after being sextorted on Facebook Messenger, was among the parents and said she had helped write the names. They have had to add 39 new ones to the banner since March, she said.

“It’s not just a banner,” Rodee said. “Every name that I wrote on there is a promise that these children will never be forgotten and that their stories will force accountability where silence once reigned.” ___

Huamani reported from Los Angeles.

This story was originally featured on Fortune.com

Two US senators have sent a letter to TikTok executives demanding answers about an experiment the company ran that withheld a safety feature from millions of users, including a teenager who died by suicide. The senators, co-sponsors of an online child safety bill, called TikTok’s decision to conduct the test “depraved.”

The existence of the experiment, discussed in detail in a confidential 2023 company document, was reported by Bloomberg Businessweek this month. It prompted the letter sent Wednesday by Republican Senator Marsha Blackburn of Tennessee and Connecticut Democrat Richard Blumenthal to TikTok Inc. Chief Executive Officer Shou Chew and Adam Presser, CEO of the company’s US spinoff.

“We write regarding disgusting new reports that TikTok knowingly withheld a critical safety measure for millions of American users — including children — in order to determine whether protecting users would impact its financial bottom-line,” the four-page letter began. The Businessweek story’s “appalling” revelations, the senators wrote, “raise serious questions about TikTok’s repeated assurances to Congress, parents and the American public that it prioritizes the safety and well-being of young people over profit.”

Read More: TikTok Kept a Safety Feature From Millions. One Died by Suicide

The TikTok document, handed over in litigation against the world’s biggest social media companies and placed under a court-ordered seal, shows the company intentionally switched off an algorithmic safeguard from 10% of US users, turning them into a control group. The safeguard was designed to break up online echo chambers of harmful content. At the time, that control group would have been approximately 15 million people. One was 16-year-old Chase Nasca, of Bayport, New York. His account, the document said, was fed thousands of videos about sadness, hopelessness, loneliness and suicide right up until he killed himself. Chase was randomly selected for the algorithm experiment on Jan. 25, 2022. Within a month, he was dead.

The document explains why Chase’s account received what it called an “onslaught” of dark content: “TikTok’s filter bubble prevention strategies did not take effect on this user by design.” And it states why the company turned off the safety setting for some: “The user impact decision was a delicate balance across safety and the ability to measure impact on DAU (daily active users) and core metrics.” 

TikTok didn’t respond to a request for comment about the letter. But in a statement for the Businessweek story, a TikTok spokesperson said the company was “deeply committed to the safety and well-being of users,” especially teens. “Our hearts break for any family that experienced a tragic loss,” the spokesperson said. “To help protect our community, we continue to invest significantly in Trust & Safety, including robust detection systems and dedicated enforcement teams that proactively remove content that violates our Community Guidelines.”

In their letter, Blumenthal and Blackburn pointed out that Congress had raised concerns about TikTok’s algorithm driving young users toward harmful content since October 2021 — before the company rolled out the experiment. The revelations, the senators wrote, “are made even more sinister because TikTok was on notice about the effects of its recommendation algorithms on children.” 

The senators demanded answers to 13 questions, including the names of every employee informed of the experiment; an explanation for why the company permitted minors to be included in the test; details on when executives learned about the experiment; and a “complete, unredacted version” of the document reported on by Businessweek. They also asked for a list of every algorithmic experiment in the US where TikTok has “withheld, disabled, delayed, or reduced a safety feature,” the number of users involved and how many were minors. The senators gave the company until Sept. 1 to respond.

“The fact that they did this knowingly and that they used their users as an experiment is something that just seems inconceivable,” Blackburn said in an interview after the Businessweek story was published but before the letter was sent. “It shows you how when our children are on these social media platforms — they are the products.”

To Blackburn, the experiment is a clear example of a company prioritizing “making money and capturing eyeballs” over safety. “Look at the fact that this was a happy, 16-year-old boy with no mental health issues and then, over this short window of time, the impact of feeding video after video after video and post after post after post about depression and suicide and what it did to this child. It is frightening — and this is something that these platforms need to be held to account for.”

Florida Republican Representative Gus Bilirakis echoed Blackburn’s concerns in a written statement, saying the report was deeply troubling and demonstrated the “devastating consequences that can occur when engagement metrics and corporate profits are prioritized over the safety and well-being of our children.” During a hearing in March 2023, weeks after the confidential document was created, Bilirakis had questioned TikTok CEO Chew about Chase Nasca and said, “Your technology is literally leading to death.” Chew responded saying the company takes these issues “very seriously” and provides mental health resources to users who search for suicide content.

Bilirakis and Blackburn both said American families have waited too long for protections for children online and called for child safety bills to be fast-tracked. Blackburn pushed for the Kids Online Safety Act (KOSA), a Senate bill she co-authored with Blumenthal to force tech platforms to prioritize child safety above profits. The bill died in the House in 2024 but was reintroduced this legislative session. Bilirakis championed the Kids Internet and Digital Safety (KIDS) Act, which includes most of the Senate bill, as well as laws for AI chatbots and video games. It passed in the House in June. 

The House bill stripped a so-called duty-of-care provision from the Senate’s version, which would force tech companies to exercise reasonable care to prevent mental health harms to minors, including anxiety, depression and compulsive usage of the social media products. This provision was removed over First Amendment concerns. 

On Aug. 5, the day after the Businessweek story was published, the Senate Committee on Commerce, Science and Transportation voted to advance the original, revived KOSA, with the duty-of-care provision. That provision, Blackburn said in the interview, would help to prevent future experiments like this because social media platforms would be required to prioritize safety in their products’ design, especially for teens.

This story was originally featured on Fortune.com

Some of America’s biggest companies are receiving hundreds of millions of dollars in tariff refunds, or booking even larger financial benefits. However, many consumers are wondering if those refunds will find their way back into their wallets. 

After the Supreme Court ruled that the International Emergency Economic Powers Act did not give the president authority to impose tariffs, major Fortune 500 companies, including Amazon and Target, have received hundreds of millions of dollars in tariff refunds. Some have pledged to issue refunds to consumers who bore increased costs thanks to the tariffs, while others have stayed mum on the subject. 

The Trump administration said as of July 31, it certified $100 billion in tariff refunds, including interest, out of the $166 billion it collected. 

Companies that received cash

Amazon, ranking No.1 on the Fortune 500, stands out as one of the largest companies to have already collected refunds. The e-commerce giant said “we received approximately $640 million of tariff refunds under the International Emergency Economic Powers Act (“IEEPA”),” during the second quarter of 2026, according to its SEC filing. The amount represented the “significant majority of refunds” it expects to receive. Amazon has said it may offer refunds to only a limited number of customers impacted by the tariffs.

Target received almost a billion dollars in refunds during the second quarter, it said on Wednesday. The department store received $994 million in tariff refunds, adding $752 million to net earnings for a total of $1.88 billion and $1.65 to earnings per share. 

Target CFO Jim Lee confirmed the company will not issue refunds as a result of the company’s IEEPA refunds, but will use the money towards bringing lower prices. “We have, and we will continue to, invest in price to ensure our guests are getting tremendous value each and every time they visit us at Target,” Lee told Modern Retail.

Nike has also recovered most of what it was owed. The sportswear company said it expected to recover $986 million. According to its filing, Nike had received $302 million as of May 31, and recorded another $684 million as “outstanding IEEPA tariff receivable.” Nike has remained quiet on whether consumers will see any refunds, even as consumers sue the company for not refunding tariff-related costs.

FedEx is a different case—the company and its competitor UPS have begun returning refunds to consumers earlier this month. The delivery company said its reported cash balance included approximately $800 million in IEEPA tariff refunds, but that money was being held for refunds to customers, according to its filing. FedEx previously sued the federal government seeking a full refund of tariffs it had paid.

Received refunds, but unclear how much

The results are mixed for automakers. Ford reported a $1.3 billion one-time tariff benefit reflecting tariffs it paid between March 2025 and February 2026, per its filing, even as the company sued the Trump administration over refunds. Similarly, General Motors separately recorded a $500 million favorable adjustment tied to previously charged tariffs, which GM said it believed were refundable in its filing. Neither disclosure, however, confirms that the full amount had already been received in cash. Stellantis, the maker of Jeep and Ram, received a tariff refund of €400 million (about $467 million).

Other companies have reported large financial benefits without making clear how much has actually been received.

Apple reported a boost from tariff refunds, disclosing that the refunds added approximately two percentage points to its fiscal third-quarter gross margin and contributed 11 cents to diluted earnings per share. Apple said it will invest its tariff refund into domestic manufacturing. 

For other major companies, the tariff refund situation is unclear after they sued the Trump administration for refunds. 

Costco said it would issue tariff refunds to consumers after being hit with four class action lawsuits alleging the company passed on the tariffs costs and raised prices. Kohl’s, which paid about $190 million in tariffs, applied for roughly $140 million in refunds but said in its latest quarterly filing that it had not received any payments. Home Depot said in its May quarterly filing that it received an “immaterial amount” after the quarter ended, and its Aug. 18 earnings release said its guidance “includes IEEPA tariff refunds, which are expected to partially offset unplanned fuel, energy, and other product input costs.” Other companies that also sued the Trump administration include Revlon, J. Crew, and Bumble Bee Foods. 

Walmart said in a May disclosure that its financial guidance did not assume any impact from tariff refunds, saying it won’t offer refunds to consumers but that it will put that money toward lowering prices. Tesla was similarly cautious in its latest quarterly filing, stating that it may be eligible for refunds of previously paid tariffs, but that the recoverability and timing remained uncertain. The company previously sued the Trump administration over its China tariffs in 2020. 

This story was originally featured on Fortune.com

Rebel Creamery’s founders say they designed the company’s logo and ice cream pint packaging themselves in Adobe Illustrator. But that DIY branding design landed them in a nearly $24 million legal dispute.

The Utah-based, low-carb ice cream maker, whose products are sold at Target, Kroger, and Walmart nationwide, filed for bankruptcy on Aug. 14, days after appealing a federal judge’s order to pay rival Van Leeuwen Ice Cream millions over a trade-dress dispute.

According to a Chapter 11 filing in the U.S. Bankruptcy Court for the District of Utah, the maker of Rebel ice cream listed $13.78 million in assets and $23.85 million in liabilities. The bankruptcy follows a July 16 ruling over allegations Rebel deliberately copied Van Leeuwen’s design. 

The dispute stems from Rebel’s packaging: solid-colored pint containers with a minimalist design and prominent black cursive lettering. Rebel’s founders told the court they designed the company’s logo and trade dress themselves in late 2017 using Adobe Illustrator, and claim not to have retained any drafts or initial records of the design.

In a July 16 memorandum and order, U.S. District Judge Eric Komitee found Rebel had intentionally infringed and diluted Van Leeuwen’s trade-dress, a legal term for the distinctive visual appearance of a product or its packaging.

“The evidence at that trial left no doubt that Rebel infringed and diluted Van Leeuwen’s trade dress and did so intentionally,” Komitee wrote.

Komitee issued Rebel to redesign its pints, writing the brand used “a near-identical color scheme and script on their packaging, with slight design differences to convey dietary information.”

“Van Leeuwen and Rebel are distributed at the same grocery stores often on the same shelf and are frequently intermingled,” Komitee wrote.

On Aug. 12, Rebel appealed Komitee’s ruling, and in the company’s bankruptcy filing two days later, Rebel listed the $24 million claim from Van Leeuwen “disputed” and “under appeal.”

“We are appealing the decision, and our products will continue to be widely available,” a spokesperson for Rebel told Fortune. 

The similar trade-dresses led to customer confusion

Van Leeuwen’s founders first noticed Rebel after an employee sent them a social media post of the company’s similar pint cup design in late 2018 or early 2019, according to the memorandum. 

They were “shocked,” telling the court  “it looked almost exactly like our packaging.” 

Van Leeuwen eventually sued Rebel in 2021, alleging Rebel’s packaging copied the look of its ice cream pints, and sought $36.4 million from its competitor’s profits. Rebel appealed the ruling, and reduced the final award to just under $24 million, allowing Rebel to claim one-third of sales for customers specifically seeking keto-friendly ice cream.

Van Leeuwen’s trade dress was created by the design studio Pentagram, which kept record of every iteration of the ice cream pint and logo design, and became key evidence used in court. 

Rebel formally stated they were unaware of Van Leeuwen’s existence when designing their trade dress in 2017, and were only made aware of the company a year later in a meeting with grocery store chain Wegmans. 

The ruling cited evidence from customer mix-ups in stores, including a 2024 complaint from a shopper who said her husband returned from the grocery store with a pint of Rebel instead of Van Leeuwen. 

“Your product was placed right next to Van Leeuwen and looked the same,” the customer wrote in a message to Rebel, according to the memorandum. “I nearly did the same thing when I shopped! Later, my friend shared the same experience on the other side of the country!”

Grocery store employees also reportedly confused the two brands when stocking the pints on shelves, and often accidentally assigned the wrong price stickers. At Walmart, Van Leeuwen had its own designated shelfspace, but some sections unintentionally housed many wrongly placed Rebel pints, according to the lawsuit.

Rebel and Van Leeuwen’s origin stories

Rebel was founded in late 2017 by married couple Austin and Courtney Archibald and initially raised money through a Kickstarter campaign, hitting their goal in only three hours and ultimately raising $80,000. The founders marketed their brand as a keto-friendly ice cream and claimed it had the lowest glycemic index on the market.

Van Leeuwen, meanwhile, was founded in New York City in 2008 by brothers Ben and Pete Van Leeuwen and Laura O’Neill, a friend of the Van Leeuwens, and Ben Van Leeuwen’s future wife. The trio opened their first bright yellow ice cream truck using $60,000 raised through crowdsourcing from 15 friends and family members. The brand focused on formulating ice cream with simple ingredients by limiting dyes and preservatives to offer dairy ice cream and vegan alternatives.

“We wanted every single guest who came into the store to feel like they were getting just as good of an experience as any other guests, regardless of their dietary restrictions,” Ben Van Leeuwen told Fortune in 2024.

This story was originally featured on Fortune.com

Dr. Heidi Overton, a top White House aide, has been picked to lead the Food and Drug Administration, President Donald Trump announced Wednesday.

Overton is a medical doctor and deputy director of the White House Domestic Policy Council who has worked on several of Trump’s second-term health initiatives. She has become a trusted administration figure and a champion of the Republican president’s goals.

If confirmed by the Senate, she would have to balance a raft of competing priorities, including Trump’s fixations, the anti-regulatory interests of traditional Republicans and the anti-corporate posture of Health Secretary Robert F. Kennedy Jr.

Those challenges dogged the tenure of the previous FDA head, Dr. Marty Makary. He resigned in May, leaving behind unfinished projects such as work on ultraprocessed foods, antidepressants and COVID-19 shots.

Referring to her as “Dr. Heidi” in a Truth Social post, Trump said Overton was a smart and respected “rockstar” who would deliver on his priorities of faster cures, innovation, lower drug prices and more wins for Kennedy’s “Make America Healthy Again” movement.

Kennedy posted on X that Overton has “exceptional judgment, professionalism, discipline and an unwavering commitment to the American people.”

Overton has promoted Trump’s health priorities

Overton attended medical school at the University of New Mexico and has a doctoral degree in clinical investigation from Johns Hopkins University, according to her LinkedIn profile. Before joining Trump’s second administration, she was the chief policy officer at the America First Policy Institute, a conservative think tank.

In recent months, she has appeared with the president to announce major projects, including some of his “most favored nation” deals with drug companies to lower prices to those of other developed countries and his recent vaccine order that sought to split the combined measles, mumps and rubella (MMR) vaccine into three separate immunizations — against the advice of medical groups.

At an Oval Office event to promote that order, she stood by Trump as he falsely suggested the number or timing of vaccines could play a role in rising rates of autism spectrum disorder. Scientific consensus and decades of studies have firmly concluded there is no link.

Overton has been critical of abortion pills and, if confirmed, would be positioned to roll back FDA rules that made them more accessible. U.S. abortion opponents have expressed frustration that the administration has not acted to stem the flow of such pills prescribed online, a situation they view as undermining state abortion bans.

Mifepristone is typically used with misoprostol in medication abortions that make up close to two-thirds of abortions in the U.S. Medical professionals call it “among the safest medications” ever approved by the FDA.

To be confirmed, Overton must seek approval from a narrowly Republican-led Senate and face questions from the Senate Health, Education, Labor and Pensions Committee. Its chair, Republican Sen. Bill Cassidy, a physician from Louisiana, has been a vocal critic of some of Trump’s actions to sow doubt in vaccinations, including the MMR order.

Cassidy posted Wednesday that while he respects Overton’s medical background, he has “strong concerns” about the nomination, mentioning her lack of managerial experience and her participation in the recent vaccine order, which he called “almost disqualifying.”

Democratic senators slammed the nomination.

“Heidi Overton is a far-right, anti-abortion extremist who has no business leading the FDA,” posted Washington Sen. Patty Murray, a committee member.

Makary’s replacement will face challenges he left behind

Before he left the FDA, Makary had drawn complaints from health industry executives and anti-abortion activists. The frustration with him came to a head over the agency’s lack of movement on flavored e-cigarettes. Companies such as R.J. Reynolds are seeking to market to adult smokers, but sweet-flavored vapes have long been blamed for the trend of teenage vaping. FDA scientists have hesitated to endorse those products to curb adult smoking.

Days before Makary’s departure, the agency opened the door to allowing more unauthorized electronic cigarettes and nicotine pouches onto the U.S. market, in a policy change that essentially bypassed FDA experts.

Pharmaceutical companies will be looking for changes at the agency, too, after rejections or reversals of biotech drugs intended for rare diseases. Complaints from biotech companies, investors and patient groups have grown and been taken up by conservative lawmakers.

Makary tried to assuage industry grievances, in part by announcing new programs designed to streamline or accelerate drug approvals. Some of those initiatives, however, have meant new headaches for the agency.

Several of Makary’s deputies also exited the agency.

Dr. Vinay Prasad, the vaccine and biotech chief, stepped down in April following intense criticism from drugmakers, patients and investors. Dr. Tracy Beth Hoeg, who was involved in scrutinizing the safety of antidepressants, COVID-19 vaccines and other widely used therapies, was replaced as FDA’s acting drug center director.

MAHA has high expectations of FDA’s next head

Overton would also need to tend to the priorities of Kennedy’s health movement, some members of which are already skeptical of her.

Kelly Ryerson, an activist critical of pesticides who is known to her supporters as “Glyphosate Girl,” told The Associated Press that Overton parroted pesticide manufacturer Bayer’s talking points in a MAHA roundtable at the White House in April.

“I was stunned by the blatant, undisguised corruption,” Ryerson said.

One major decision for the next commissioner is what to do about peptide regulation. A panel of federal health advisers selected by Kennedy recently recommended easing access to several peptides that are popular with wellness influencers and celebrities, despite warnings from FDA scientists that the chemicals have not been shown to be safe or effective.

The agency recently proposed a rule change that would require food manufacturers to notify regulators before introducing new ingredients or additives into processed or packaged foods. But it is still working with the White House to finalize a first-of-its-kind definition of ultraprocessed foods, which Kennedy blames for elevated rates of diabetes, obesity and other chronic conditions.

Under Kennedy, the FDA has set up extra hurdles for vaccine testing and blocked the publication of research into the effectiveness of COVID-19 shots. Kennedy stood by Trump recently as the president signed the executive order aimed at upending childhood vaccinations in the U.S. despite unprecedented financial and logistical challenges.

The FDA also has an ongoing review of widely used antidepressants, including whether they may increase the risk of autism and other disorders when used during pregnancy. The medicines have long been a target of Kennedy, who was a leader in the anti-vaccine movement before joining the government.

In recent weeks, the agency has been investigating two summertime foodborne illness outbreaks: a cyclospora outbreak linked to tainted iceberg lettuce that has sickened thousands of people and a multistate salmonella outbreak linked to jalapeño peppers.

___

Swenson reported from New York. Associated Press journalist Laura Ungar contributed from Louisville, Kentucky.

___

The Associated Press Health and Science Department receives support from the Howard Hughes Medical Institute’s Department of Science Education and the Robert Wood Johnson Foundation. The AP is solely responsible for all content.

This story was originally featured on Fortune.com

It’s hard to get a splashy sound bite out of Michael Dell, even if you tee him up for one. When asked how big a growth opportunity the AI wave could be for his namesake company, Dell Technologies, the founder and longtime chief executive doesn’t offer up any pithy one-liners but instead ruminates in real time.  

“It feels every bit as big as previous waves, but probably bigger,” he says, pondering the question, and then adds, “You know, maybe quite a bit bigger.” He takes another brief pause, reconsiders his own words, and delivers a most inconclusive conclusion: “I don’t know for sure. Nobody knows.”

We’re seated in a conference room at Dell Technologies’ headquarters just outside Austin, where the temperature has hit 88° F in early March. Dressed in dark slacks and a navy blue denim button-down (Texan for business casual, no matter the season), Dell has just emerged from a photo shoot that he tolerated but clearly didn’t relish. It’s not that he isn’t on board with being the name and face of his company. That’s been true for a while—40 years, to be exact. He remains Dell Technologies’ biggest believer—and biggest shareholder, with 53% of the $79 billion company’s stock under his or his wife Susan’s name. But he’s not a natural-born showman. Never was. In fact, he seems to go out of his way to not put on a performance—even as he’s embarking on what could be his greatest act yet.

Unlike some other tech CEOs, Dell doesn’t do bombastic declarations or colorful antics; he doesn’t have a side hustle that involves blasting himself into outer space. Despite having spent his entire adult life in the public eye, he is measured, analytical, and almost intentionally unexciting. So his reluctance to put a ceiling, or even a floor, on what generative AI could mean for his company is not surprising. 

But while Dell may prefer to hedge, the market isn’t hiding its exuberance. Just a few days before our interview, on March 1, Dell Technologies’ share price leaped 38%, hitting an all-time high above $131 after the company reported earnings that beat analyst expectations. The announcement generated plenty of excitement about demand for Dell’s growing portfolio of back-end tech products, the kind required for storing and managing the massive datasets needed to run—you guessed it—generative AI applications. Orders for AI-optimized servers were up 40% in the most recent quarter. As chief operating officer Jeff Clarke said in the company’s earnings release, “We’ve just started to touch the AI opportunities ahead of us.” 

It’s not just Dell’s company that’s been buoyed by the buzz. As a result of the massive rise in the stock, Michael Dell’s personal net worth reportedly hit the $100 billion mark in early March—a notable milestone even for a man who became a billionaire at the tender age of 30.

But none of this seems to rock Dell’s world. Over the decades, he’s maintained the same steady demeanor through exhilarating highs and harrowing lows. Along the way, he’s steered his company through multiple major pivots. And he’s showed an uncanny ability to read his customers’ needs and make the right strategic change at the right time, whether de-emphasizing PCs in favor of servers, sensors, and storage, or taking the company private—over the heated opposition of Carl Icahn—in a mammoth buyout.

Mario Tama—Getty Images

That privatization maneuver is precisely what positioned the company to capitalize on the current AI boom. Over the five years that it was privately held, Dell was able to truly diversify from selling laptops and desktops. Away from the market’s obsession with quarterly earnings, Dell consolidated and expanded his company, creating a behemoth provider of infrastructure tools for corporate customers. Along the way, he engineered what was then the biggest tech deal in history, the $67 billion acquisition of data storage provider EMC. 

If Dell isn’t a dynamic, headline-making speaker, it may be because he’s built this four-decade run on listening—deploying his analytical skills and deep curiosity to recognize what his customers need and to navigate his industry’s twists and turns. “I love spending time on the technology, and I love spending time with our customers,” he tells me. And at least where business is concerned, he adds, “I don’t really love anything else.”

Dell Technologies still sells Dell PCs; in fact, computers make up the majority of its revenue. But today it’s a company vastly different from what it was five or 10 years ago—let alone 40. The one constant? Dell himself. “This is probably the longest-sitting CEO in the tech industry,” says Marc Benioff, cofounder and CEO of enterprise-software maker Salesforce and a longtime friend. “He’s six months younger than I am, but I view him as an older brother,” Benioff says of Dell. “He’s a phenomenon in every possible way.”

Sitting across from Dell at his HQ in Round Rock, a corporate campus that’s forgettable except for its sheer size, “phenomenon” isn’t the first word that comes to mind. But Dell has built—and hung on to—an empire that now provides the technological building blocks for 99% of Fortune 500 companies, most of which will have new needs in this new era of AI. If he plays his cards right, the next chapter of the story could make both the CEO and his once-flailing PC maker more relevant than ever, all but ensuring he’ll stay at the helm for years to come. 

The morning after our interview, Dell is speaking on a panel at a health care innovation summit at the University of Texas at Austin, his alma mater. (Dell finished two semesters before dropping out to devote himself to selling PCs full-time.) Investor Jim Breyer, who relocated to Austin from Silicon Valley in 2019 at the Dells’ suggestion, introduces the CEO with glowing superlatives. “Michael Dell is the most courageous entrepreneur I’ve ever worked with,” he gushes. 

Dell’s performance is … just fine. (It’s clear that public speaking is not his happy place.) Still, he comes across as confident and purposeful. At 59, Dell retains a youthful bearing, his curly hair only tinged by gray. And from the audience reaction, it’s clear Dell’s the big man on campus, even if he never graduated.

In his well-documented early days, the nerdy but gutsy founder could seemingly do no wrong. In 1984, as a premed freshman, he started tinkering with computers in his UT dorm room. By age 19, he had left school and turned all of his attention to his business. He faced other, much bigger competitors, including IBM and Apple. But Dell pioneered a new way of doing business: His computers were built to order, and he sold them directly to consumers, cutting out the middleman. In 1988 he took Dell Computer public, raising $30 million and using the capital to expand globally. At age 27, he became the youngest CEO on the Fortune 500. And the company just kept growing—as long as demand for PCs was on the rise. 

But PCs would prove to be the company’s Achilles’ heel. In 2001, Dell became the world’s leading computer maker, surpassing the once-mighty Compaq. But sales soon began to decline. Asian manufacturers had entered the fray, offering cheaper products to American consumers. And by the late 2000s, smartphones and tablets had swarmed the market, slowing demand for desktops and laptops even more. The company tried to jump on the mobile bandwagon, but its efforts were ill-received: Dell’s “phablet,” a product that sat in the unnecessary purgatory between a phone and a tablet, was discontinued after just one year. 

By then, Dell had been trying for years to diversify. In 1995 he entered the server market with the PowerEdge, a product line that still exists—designed for enterprises that were amassing far more data than they could manage with their existing equipment. In 2006, the company launched a business unit to support cloud computing, including tools to power “hybrid clouds”—private clouds (which keep data on a customer’s premises) that can integrate with public ones (where data is hosted by a third party). 

But this expansion wasn’t happening fast enough to offset declines in PC sales, and investors hammered Dell’s shares. In 2013, after more than two years of falling PC revenue (and after the stock price bottomed at under $11), Dell decided to take his baby private—hypothesizing that shielding the company from Wall Street’s short-term focus on profitability was the best way to reset for the long term. 

Benioff refers to the deal as Dell’s “magic trick.” But the maneuver was anything but slick and graceful. “I had no idea how difficult it was going to be,” Dell recalls. “When it started, [I thought], ‘Is this like a one-week thing or two-week thing?’ I didn’t know it was going to be an eight-month thing.”

Courtesy of Dell

Dell wasn’t in it alone. Egon Durban, co-CEO of private equity firm Silver Lake, was his partner from the get-go. The two presented Dell shareholders with what they thought was a good offer, a $24.4 billion deal financed by a mix of equity and debt—the largest leveraged buyout in tech-industry history. But then corporate raider Carl Icahn entered the picture, snapping up a sizable chunk of the company’s shares and agitating for a more generous offer. Before they knew it, Dell and Durban were going to war, fighting Icahn as he made a counteroffer that involved buying the company himself—and ousting Dell as CEO. 

Eventually, Icahn got concessions, and Dell got his deal. Dell and Durban increased their offer by 10 cents a share and threw in a special dividend for some shareholders. And on Oct. 29, 2013, Dell Computer became a privately held company, owned by Michael Dell and Silver Lake. 

During the lengthy feud, the antagonists stayed true to their personalities: Icahn took to CNBC and other outlets to spread his narrative, while Dell lay low. But in recent years, Dell has spoken openly about the clash. His 2021 memoir, Play Nice But Win, opens with a scene in which Dell goes to Icahn’s house for a dinner of mediocre meatloaf, in a (failed) attempt to find common ground. Though Dell says he doesn’t hold grudges, he also says he felt a need to “expose” Icahn’s tactics.

More than 10 years later, it’s clear there’s no love lost between them. “Icahn showing up was the hardest part,” Dell says. “It was a long, painful period where everyone was subjected to this horrible situation.” Dell maintains that Icahn never really planned to buy his company but simply wanted to squeeze more out of the deal. For his part, Icahn, in a phone interview, says that his actions forced a “meaningful improvement” of the buyout. “The shareholders got a lot more money because of me,” says Icahn. 

Tellingly, both men quote World War II–era leaders to describe their conflict. “What’s that Winston Churchill quote?” Dell asks me rhetorically, invoking the former British prime minister: “If you’re going through hell, keep going.” Icahn, meanwhile, puts his own paraphrasing spin on a 1936 campaign speech by Franklin D. Roosevelt: “Dell hates me—and I welcome his hatred.”

Still, the trials arguably made Dell a better leader. Those close to the CEO say that his determination and belief in the deal carried the enterprise through a rough patch. “Relationships are forged on the battlefield,” says Durban, who remains close to Dell and whose firm is one of the company’s largest shareholders. Employees from that era say Dell became more connected than ever to his workforce, and even better about communication with the rank and file.

Just as important, Dell proved himself to a wider swath of the business world as an analytical, decisive chief executive. “He took a large risk, which is easier not to do,” Jamie Dimon, the CEO of JPMorgan Chase, says of Dell’s deal. “But he stuck to his guns.” Describing Dell, Dimon invokes the “OODA loop,” a military acronym for efficient decision-making that he says is a secret sauce for the tech CEO. (OODA stands for “observe, orient, decide, act.”) 

That kind of coolheadedness also characterizes Dell’s very, very few hobbies. Benioff tells me that his friend recently took up hunting with a bow and arrow. (Dell’s company won’t confirm this.) Dell hunts for birds, Benioff says—the kind of elusive target you can hit only when you’re calm, unemotional, and utterly focused. 

Jeff Clarke, Dell Technologies’ COO, is the closest person Michael Dell has to a cofounder, having joined his team in 1987. Speaking to Fortune via videoconference, Clarke—dressed in a red, white, and blue T-shirt that simply says “TEXAS”—refers to the company’s private-company era as one of the most fun periods of his career. “It was liberating,” says Clarke. 

Being out of the public market meant that Dell could make big bets and invest in R&D, even if the payoff wasn’t immediate. The company could rebuild itself around providing all things infrastructure for corporate customers like Home Depot and CVS Health, whose greatest needs increasingly revolved around the growing mountains of data they were accumulating. In 2016, Dell and Durban—with Dimon’s help—orchestrated another financial feat, the $67 billion purchase of EMC and its software subsidiary VMware. The acquisition was “something we had dreamed about doing,” Dell says. 

Still, the deal was an expensive bet that saddled the business with a heavy debt load—and it created hassles down the road. The merged company started trading publicly again under a share class that tracked its ownership interest in VMware; two years later, it bought those shares back and replaced them with a new share class. Along the way, some VMware investors (including, briefly, Dell’s old buddy Icahn) sued, arguing that the complex deal undervalued their shares, and Dell Technologies eventually paid a $1 billion settlement. Still, the acquisition added an even broader data-storage and management portfolio to Dell’s arsenal, making the company indisputably stronger. 

Dell’s company has never been a “market maker,” a company that creates demand for something that didn’t previously exist. But it hasn’t had to be. “What Dell’s been good at is knowing the right time to get into a market,” says Patrick Moorhead, an analyst who has covered Dell and its competitors for years and now runs Moor Insights & Strategy. “They’re so close to their customers that they just know.”

That closeness was embedded at Dell from the earliest days, when Michael Dell himself was building PCs for one customer at a time. In 1988 Dell wrote the company’s first Culture Code, with “Provide high-quality products and excellent customer service” at the top of the list. His focus hasn’t changed much, his allies say, and it’s been central to his ability to keep transforming the company. 

On Dec. 28, 2018, the reorganized, renamed Dell Technologies emerged fully from its cocoon, trading on the NYSE under a new share class. In its metamorphosis, the company had all but shed its image as a lagging PC maker, refashioning itself as an enterprise infrastructure giant. And enterprises, it turned out, were about to need a whole lot more infrastructure—and maybe, just maybe, more PCs.  

Back in Round Rock, Dell is trying to explain what an “AI PC” is, and why anyone would want one. “I have a list,” he says as he gets up to grab his phone from his office. The CEO comes back and proceeds to rattle off a catalog of capabilities.

There’s real-time, AI-powered translation, he explains, and a feature called “circle to search,” which enables PC users to highlight a word or line, which the computer will then provide more context and information for. There’s also “generative AI editing,” which can assist with any kind of writing or content creation. What customers actually end up using these machines for, Dell admits, is beyond his expertise to foresee. “But I believe that people will figure out creative uses and that companies will want to have the capability to make their people more productive.”

In fact, Dell’s lessons from its earliest days of customizing laptops still apply in the AI era: The key is to be flexible enough to meet customers’ demands. “The competitive advantage for Dell today is that it offers services you can tailor to almost every need in AI,” says Orit Gadiesh, the chairman of Bain & Co. and a decades-long consultant and confidante of Dell’s. “It’s not a fixed thing.”

Corporate customers in and outside tech are already clamoring for back-end machines that can both house and make sense of the data that feeds into generative AI applications. At a time when huge platform creators like OpenAI and Google are competing for corporate clients, Dell Technologies doesn’t have to worry about who wins: Its tech “stack” is agnostic to different flavors of generative AI, just as its cloud offerings have always accommodated hybrid, private, and public cloud strategies. And just as with the move to the cloud, Dell is counting on one common denominator with AI: that all companies, regardless of which AI applications they build or deploy, will want control over the hardware where the relevant data is stored.

Courtesy of Dell

To be sure, Dell didn’t know that the generative AI explosion would happen when it did; he credits Jeff Clarke with devising much of the company’s AI road map. But he calculated long ago that going all in on data infrastructure was the best way to position his company for the future. As a result, “he’s not just providing the picks and shovels, but also housing and food and beverages for the AI gold mine,” says Silver Lake’s Durban. 

It’s still early days for Dell Technologies’ AI story. Fast as it’s growing, Dell’s AI server products account for just a tiny fraction of its business. But most financial analysts seem bullish about what’s to come. There’s even hope that the AI craze will jump-start demand for PCs—AI PCs, to be precise. The thinking is that the need for increased processing power won’t just be on the data-center side (where servers and storage systems handle companies’ information), but also on the desktops and laptops that consumers and workers interact with. 

In February, Dell Technologies announced its first line of Latitude AI PCs, which look like normal computers but include a tiny component called a neural processor, the key to enabling generative AI workloads. It’s not the only vendor with high hopes in the category—HP and Lenovo have announced similar products. And it’s not clear when demand will take off. Bloomberg Intelligence analysts wrote that “sales and units shipped may disappoint investors in calendar 2024, having a greater potential impact in 2025.”

Even Dell acknowledges that spurring demand could take a while. That said, “if you’re responsible for the PCs in a company, the last thing you want to do is have a bunch of PCs that don’t do the thing that the users want them to do,” says Dell. “I do think there’s going to be a refresh wave.”

The top floor of the University of Texas’s Innovation Tower, a new high-rise that’s meant to be a startup hub, is still empty. But one only has to look out the windows for inspiration. The 360-degree views of the Austin skyline show a city dotted by cranes and construction in almost all directions. 

Jay Hartzell, the university’s president, is showing me around, pointing out all of the landmarks—including the buildings adorned with the name of the institution’s most famous dropout. Michael Dell never did become a doctor, but his name is on his alma mater’s medical school, the university’s teaching hospital, and its pediatric research center. (Not to mention Austin’s Jewish Community Center.) 

“When we talk about what we want to produce as a university, and why people should come here, he’s sort of Leading Exhibit A,” says Hartzell. He credits Dell not only with being a major employer of UT graduates but also with helping to spur the city’s broader tech ecosystem. Over the years, tech companies from Meta to Apple have set up shop in the Texas capital. Investors, too, from Vista Equity Partners to Pimco to Jim Breyer, have put down roots. UT recently welcomed its first cohort of students in a brand-new AI graduate degree program. 

Dell, who is originally from Houston, never wanted to move his headquarters away from Texas, even when others told him he should relocate to Silicon Valley. “He helped put the place on the map,” Austin Mayor Kirk Watson tells me in a phone interview. “If you took Michael Dell out of the equation, it would be a strikingly different city.” 

Dell has made his mark outside of Austin, too. The Michael & Susan Dell Foundation, which the couple founded in 1999, has 800 active projects around the world at any given time—focusing on education, training, and health innovation to help children living in poverty. (Dell and his wife recently contributed another $3.6 billion to the foundation, bringing its total endowment to $5.2 billion.) Dell says he spends a little more time each year on the foundation. He’s also gotten more hands-on with his family office, which invests in real estate development and hotel companies, among other sectors. 

Could those jobs someday be his life’s work? Dell’s next chapter could be a long one: Even after 40 years leading Dell, he’s still so young, at 59. But the thought of playing any role other than his current one—at the center of the business that he’s synonymous with—seems to stump him. When asked if he could see himself running Dell Technologies in 20 years, Dell says he hasn’t thought that far ahead, but that there’s no other role he craves. Then, at long last, he provides something like a money quote: “I’ve said this before: I’ll still care about Dell when I’m gone.” 

The long and winding road

Michael Dell’s company began life in 1984 as PC’s Limited—selling computers, and that’s it. A few crucial pivots helped the company evolve and stay not just relevant but dominant.

1995 
Dell Computer, by then a Fortune 500 company, releases the first-generation PowerEdge enterprise server—its first attempt to sell data storage to enterprises.

2006
Dell joins the cloud era, announcing a new business unit that provides cloud products and services to customers. Demand is relatively slow to catch on.

Stuart Isett—Fortune Brainstorm Tech

2013
Michael Dell and Egon Durban of PE firm Silver Lake (above, with Dell at left) take the company private in an effort to refocus the company on corporations’ data infrastructure needs.

2016 
Dell acquires EMC and its stake in VMware for $67 billion, at the time the largest tech deal ever—making Dell’s data-storage and management portfolio far larger.

Courtesy of Dell

2023 
Dell Technologies releases a series of infrastructure products, including servers and storage, that are optimized for generative-AI applications.

2024 
In February, the company announces its AI PC, which includes a “neural processor” to handle AI workloads. In early March, Dell Technologies stock hits an all-time high.

This article appears in the April/May 2024 issue of Fortune with the headline, “The [forever] founder.”

This story was originally featured on Fortune.com

U.S. government debt has hit $40 trillion—an alarming milestone for economists who fear the nation’s fiscal policy is spiraling out of control.

It comes after the Congressional Budget Office (CBO) reported earlier this month that deficits are now so large that the U.S. Treasury is paying $3 billion a day in interest, totaling $963 billion between October 2025 (when the 2026 fiscal year begins) and July 2026. 

Treasury data for August 18 shows the closing balance for the day on public debt outstanding totaled $40.04 trillion. 

Budget watchdogs have continually called on policymakers to get America’s fiscal house in order. Proposals range from cutting annual federal deficits in half as a share of GDP, down from the current 6% to 3%, to calls to “cut up the credit cards” entirely. The White House itself has indicated it recognizes a problem, with President Trump suggesting tariffs or visa policy could help plug the budget gap: So far, data suggests it won’t be enough.

With the country’s debt-to-GDP ratio now north of 120% (a metric lenders will watch when analyzing the risk premium on loans to the U.S.), debt hawks are warning the public is already paying, and is only going to start feeling the squeeze more acutely.

Michael Peterson is the chairman and CEO of the Peterson Foundation, a nonpartisan organization dedicated to putting the U.S. on a more sustainable fiscal path. Speaking to Fortune as America hit the $40 trillion benchmark, Peterson explained that even if families don’t receive a “bill in the mail” for national debt, they’re already paying.

He explained: “When the U.S. borrows this much—and continues to borrow more and more—that drives up interest rates, which then increases household expenses because your mortgage goes up, your car loan, your credit card bills, and inflation more generally. So [we] may not get a bill at the end of the month for national debt, but [we] are paying that bill both in the form of taxes as well as an inflated level of expenses.”

While the mechanics of how debt may trickle down to individual households’ finances are complicated, voters are nevertheless expressing concern about the topic as D.C. heads into midterms. In July, a Peterson Foundation study reported 94% of voters are more likely to support a candidate with a plan to address the debt, including 95% of Democrats, 92% of independents, and 94% of Republicans.

“To anyone who cares about America, about democracy and our future, in my view, this is already a crisis,” Peterson said, “because the level of fiscal mismanagement is tragic. It is burdening every household today, it’s laying more and more debt on our children and grandchildren, and that’s not how America got to be the great country that it is.”

What gives?

The bull case for debt is reasonable. Firstly, despite years of warnings, there has yet to be a market meltdown sparked by debt.

Indeed, Treasury yields—the surest sign of confidence in U.S. borrowing and lending—are showing no signs of acute discomfort. At the time of writing, 30-year Treasuries sit above 5%, elevated (in part) by the uncertainty of Federal Reserve policy. 10-year treasuries are sitting above 4.6% for a similar confluence of reasons.

But debt hawks point to other indicators that suggest the budget will have to give in one area or another.

Nancy Vanden Houten, lead U.S. economist Oxford Economics, said in a recent note that “mandatory spending, including Social Security, Medicare, and interest on the debt, continue to see the most growth in spending. Fiscal year-to-date defense spending continues to creep higher as the war with Iran drags on; as of July, defense spending was up 5% y/y.”

A trade-off between two of those outlays seems to be on the books: The trust fund for social security is due to run dry in a little under eight years, and Medicare in a little under seven years, according to estimates by the Committee for a Responsible Federal Budget.

While “it’s hard to pinpoint an exact moment in time or an exact program that will be in jeopardy, if you care about government programs and what the government can do to help society, defend our country, make sure the most vulnerable are protected, and take care of the elderly, the first thing you should do is put us on a more stable fiscal path so that all those programs are less in jeopardy,” Peterson said.

However, any cross-party agreement to examine or target borrowing in order to reduce debt is yet to materialize. Peterson adds: “There’s a lack of urgency that concerns me. Just because the financial markets were OK yesterday doesn’t mean they’re gonna be OK tomorrow. To just continue to cross your fingers and hope that we can get away with a completely irresponsible level of budgeting is not a reasonable way to lead our country.”

This story was originally featured on Fortune.com

In the run-up to the iPhone’s launch in 2007, Apple cofounder Steve Jobs made a fateful decision: Apple would not turn to its partner Intel to make chips for the device, on the grounds that the firm was “really slow…like a steamship,” as Jobs put it. Apple would rely on up-and-comer ARM instead.

Jobs’ decision helped set off a two-decade decline for an iconic Silicon Valley brand. The company was once so celebrated for its chipmaking innovations that hardware makers clamored to attach “Intel inside” stickers to their devices. But Intel went on to miss not only the mobile revolution, but the AI era, as competitors stole its market share.

By early last year, it was unclear if Intel could remain a going concern. But then something remarkable happened. The company brought on CEO Lip-Bu Tan and, in barely a year, became one of the hottest stocks on the market. The story of Intel’s ongoing turnaround could become the rebound story of the decade—one featuring bold leadership, tough decisions, and no small amount of luck.

When Tan took the helm in March 2025, the longtime semiconductor veteran became Intel’s third CEO in six years, and the sixth since legendary cofounder Andy Grove relinquished the post in 1998. By Tan’s arrival, Intel had become defined less by its chips than by the $50 billion in debt it carried. “There was a large recognition that we needed to right the balance sheet,” says CFO David Zinsner, “but not a lot of clarity on how we were going to do that.”

In response, Intel set about selling off noncore parts of the business and raising capital from what Zinsner describes as Tan’s “incredible network.” Soon, Intel had tapped billion-dollar investments from Nvidia and SoftBank; it also grabbed headlines when Tan agreed to let the Trump administration convert a scheduled $8.9 billion grant into an equity stake for the federal government.

“The SoftBank endorsement was good; the U.S. government endorsement was great,” says Zinsner, citing a halo effect that raised Intel’s standing with creditors and investors and shored up its capital structure.

But money alone could not address Intel’s deeper problem of corporate complacency. To combat it, Tan sought to impart a spirit of candor. He cut Intel’s management structure from 12 layers to six and made a point of hearing firsthand about its performance from people at all levels of the company—putting an end to a pervasive practice where managers would filter only good news to the C-suite.

“If there’s a problem and you tell me about it early, it’s our problem, and we’ve got to fix it. If you have a problem, and you don’t tell me, it’s your problem,” Tan told everyone at Intel upon his arrival, Zinsner recalls.

For all the rapid progress it has made under Tan, Intel still needs to show that its chips can compete. “They got fat, dumb, and lazy, and got their ass handed to them,” says Bernstein analyst Stacy Rasgon. Nvidia and TSMC became the leaders of the AI era while Intel mostly watched from the sidelines.

Under Tan, however, Intel has been getting a bigger piece of the AI boom. Part of this has been a matter of luck, Rasgon explains: An insatiable demand for memory to support AI functions has led companies to find more uses for Intel’s traditional CPU chips. This has allowed Intel to sell huge amounts of existing inventory, helping to drive the surge in its share price.

The company’s real challenge is to prove it can still make the cutting-edge chips that once defined it. There are promising signs: Intel’s closely watched efforts to build chips using its next-generation 14A manufacturing process are on track, a technological transition that could help the company woo more big-spending customers. Intel’s design of chips for other companies, the other pillar of its business, got a big boost on recent reports that Apple may turn to the company again as a supplier.

While it’s too soon to say whether Intel will complete its comeback, it clearly has learned from past mistakes. According to Zinsner, Tan has largely departed from his predecessors’ tradition of frequently quoting Andy Grove. He has, however, adopted and repeated one of Grove’s most famous maxims: “Only the paranoid survive.”

This story ran in the June/July 2026 issue of Fortune as part of a feature called ‘Innovation Giants on the Rebound.’ For more Fortune 500 innovation stories, click here.

This story was originally featured on Fortune.com

Internet firm Naver has earned its nickname—Korea’s Google—by pulling off an improbable feat: It dominates South Korea’s search market, having defended its turf from Google, the world’s top search engine, whose revenue is 40 times as large as Naver’s. 

As of September 2023, Naver controlled 59% of Korea’s search market to Google’s 31%. 

Naver is perhaps “the only company in the world that has survived competition against Google and Amazon,” says Choi Soo-yeon, Naver’s CEO since 2022. 

Naver is hardly a household name outside Korea. But it operates a sprawling portfolio that pits the $22-billion-in-market-cap firm against other Big Tech giants on multiple fronts.

It has a controlling stake in both Yahoo Japan—the most popular website in Japan, according to Nielsen—and the Japanese messaging app Line—a WhatsApp rival—through a joint venture with SoftBank. 

It runs Korea’s No. 2 e-commerce service, behind Coupang. (Amazon’s platform ranks fourth.) Naver’s $1.2 billion purchase last year of Poshmark, the U.S.-based clothing-resale site, expanded its retail reach.

In the streaming wars era, Naver has amassed content platforms of its own. It owns Webtoon, which hosts mobile-friendly comic strips that are popular across Asia. (Naver is reportedly planning a U.S. IPO for Webtoon later this year.) Naver bought Wattpad, a Canada-based platform for user-submitted fiction, for $600 million in 2021. 

Naver also has a small but growing cloud-computing business—a category in which Amazon, Google, and Microsoft reign supreme—and it’s launched a series of AI projects to contend with the release of OpenAI’s viral chatbot, ChatGPT.

In recent quarters, Naver’s many business lines have notched record revenues and operating profits, but Choi sees Naver’s mission as extending beyond its own bottom line. 

Despite Naver’s small size, she casts the firm as a counterpoint in a global tech scene in which power is concentrated among a gargantuan few. 

“It’s becoming a world where there are only one or two search companies and one or two commerce companies,” Choi said in a recent wide-ranging interview—her first with the international press—at Naver headquarters in Seongnam, just outside Seoul. (Choi gave her answers in Korean, which were later translated into English.) Naver “is a company that constantly fights against such a world and strives to preserve diversity,” she says. 

Investors question whether Naver can go toe-to-toe with cash-rich rivals. Still, Choi’s goal is ambitious—perhaps even noble—and belies the role she was appointed to fill: that of a caretaker CEO brought on to steady a company in turmoil. 

Choi, 42, was an unconventional pick to run the company. A Harvard-educated M&A lawyer, she joined Naver in 2019 as head of global business support to help lead the firm’s expansion. Three years later, the board named Choi as CEO to show it was prepared to overhaul its culture after a series of crises. 

In May 2021, a senior Naver developer died by suicide after accusing the company of fostering a toxic work culture. A labor union probe found he’d been bullied by executives for years. A later government survey found that over half of employees felt they were bullied at least once in a six-month period. 

In the aftermath, Naver said “there were some acts of workplace harassment by some executives.” Choi’s predecessor as CEO, Han Seong-sook, stepped down, as did Naver’s COO, and Naver tapped Choi. At the time, she told shareholders her most urgent task was “to recover Naver’s corporate culture based on trust and autonomy.”

But a second tragedy struck early in Choi’s tenure. In September 2022, another employee died by suicide while on maternity leave, local media reported. Months later, her family claimed she’d been mistreated at work. A Naver internal investigation did not uncover evidence of harassment, and Korea’s labor ministry could not confirm the family’s claims, a company spokesperson said.

South Korea has the highest rate of suicide of all Organization for Economic Cooperation and Development countries, and South Koreans work 200 hours more per year than the global average.

“There was a lack of trust in the systems, leadership, and board,” Choi says of Naver’s previous culture. One big change Choi made was to reintroduce remote work, a rarity in post-COVID South Korea. The option gives employees “the choice of what kind of working environment they can be most productive and create the most innovation in.” 

The suicides were the biggest scandal to rock Naver in its 25-year history. Naver launched in 1999, when founder Lee Hae-jin turned an internal Samsung project into an independent company. By the mid-2000s, Naver had passed rivals like web portal Daum to dominate Korea’s search market.

Naver and fellow internet firm Kakao (founded by an ex-Naver executive) have made South Korea one of a handful of countries where homegrown search engines outperform U.S. search giants without government intervention.

Naver got a head start on tailoring a search engine that met the tastes of the South Korean market, says Bokyung Suh, a Korea analyst at Bernstein Research. Naver hooked users early with its busy homepage that was heavy on icons, links, and animations. Google launched a bare-bones search engine in Korean in 2000, but it didn’t catch on. Google updated its site to a feature-rich format six years later. 

Naver earned 9.6 trillion won ($7.41 billion) in revenue in 2023, a record. Search and e-commerce generated 37% and 26% of sales, respectively. It made $1.3 billion in operating profit, also a record. 

Yet shares are currently trading about 60% below a COVID-era high. Investors are concerned about slowing revenue from search (up 0.6% in 2023), plummeting revenue from display ads (down 10% last year), and the lack of a “punchy, clear growth strategy,” Suh says. 

Naver’s lack of a punchy, clear growth strategy is especially worrisome in the AI age, when Naver faces U.S. companies that are investing billions in the technology. 

As U.S. giants dominate English-based AI, Naver may be able to establish an edge in systems based in Korean and other languages, Choi says.

Naver has a chatbot, CLOVA X, and an AI-powered search engine, Cue. Both are built on its Korean-language large-language model, HyperCLOVA X, which outperforms OpenAI’s GPT in Korean, recent studies say. Naver has claimed that HyperCLOVA, an earlier version of its model, was trained on 6,500 times as much Korean data as GPT-3.0, which underpins OpenAI’s ChatGPT. Naver is also partnering with Saudi Aramco on an Arabic large-language model.

Choi is especially interested in what she calls “sovereign AI,” or a model that’s tailored to an individual user. “We focus on what companies and governments that want to use AI would want, and what needs Big Tech can’t fulfill,” she says. As AI becomes more common, “each group will need an AI model that best understands” its unique traits. 

Sources: Bloomberg; Naver

Much like her rivals at Google and Microsoft, Choi is also grappling with how to integrate AI—with its penchant to hallucinate false information—into Naver’s search product. “People need accurate information through search,” she says, though she hopes AI’s tendency to make stuff up will become “nearly negligible in the near future.”

At the same time, she also sees room for a traditional search engine—with its list of links to choose from—amid the generative AI revolution. “Not all questions in the world have a single correct answer,” she says. “There is still a need for exploration.” 

Naver’s base in South Korea, a chip powerhouse, is an advantage in the AI race, especially its homegrown manufacturers “that support the Korean language,” Choi says. Naver’s chip partners include Korea’s Samsung and the U.S.’s Intel.

“It’s not healthy to rely on just one company,” Choi says. Is that a coded reference to Nvidia? Yes, she says in English, with a smile.

Just four of South Korea’s top 100 firms by revenue had women CEOs last year, according to global headhunter UnicoSearch. Just 6% of executives at the companies were women.

Choi, whose CEO contract expires in 2025, expresses some unease at often being a “sole woman” in business: “Simply because I am a woman, there are expectations for me to demonstrate skills such as effective communication, adept conflict resolution, and the ability to nurture people.” 

Korea’s internet sector got a little more diverse when Kakao appointed its first female CEO, Shina Chung, in March. 

“I’m not alone anymore!” Choi says. 

Fortune Korea contributed additional reporting and translation assistance.

This article appears in the April/May 2024 issue of Fortune.

This story was originally featured on Fortune.com

U.S. President Donald Trump threatened to bomb Oman because he is unhappy the country is close to a deal with Iran to manage ship traffic through the Strait of Hormuz, two regional officials said on Tuesday, a day after Trump leveled the threat.

The officials said the Trump administration has told Oman it is opposed to parts of the yet-to-be-announced deal, including the joint Iranian and Omani management of the exit route out of the passage that’s critical to global supplies of oil and natural gas. The officials were briefed on the U.S. position and how the administration views Oman’s position.

Meanwhile, a projectile hit a ship as it sailed out of the strait, and a cargo vessel was rendered a “constructive total loss” by multiple projectiles off the coast of Yemen, according to the British military’s maritime monitoring agency. The defense ministry of the United Arab Emirates said two ballistic missiles were launched from Iran toward the UAE.

An Iranian official said the strait would not reopen until the United States meets Iran’s conditions.

US believes Oman has not been tough enough with Iran

The U.S. believes Oman has not been tough enough in its negotiations with Iran and is unhappy with Oman’s agreement to collect voluntary fees from vessels, even if the charges are related to security and maritime environmental protection, according to the officials, who spoke on condition of anonymity because they were not authorized to talk to journalists.

Trump on Monday threatened to bomb Oman as it works with Iran on a deal to open the strait and pave the way for the U.S. and Iran to resume negotiations to end the war.

He posted a map on social media on Tuesday depicting the strait as U.S. territory. The president first mentioned the idea in an offhand comment last week. On Monday, he told reporters in the Oval Office, “I like the idea of declaring it a territory,” without providing details.

The White House on Tuesday referred to Trump’s remarks in the Oval Office and declined to comment further about Oman.

Iran refers to Trump’s ‘delusion’ about the strait

Iranian Deputy Foreign Minister Kazem Gharibabadi appeared to respond to Trump’s post on social media depicting the strait as American territory.

“Just as Trump correctly wrote the name of the eternal Persian Gulf, his delusion regarding the Strait of Hormuz will soon either be corrected, or we will correct this deluded man’s delusions for him,” Gharibabadi wrote on X.

Trump’s comment on Monday was not the first time he has threatened Oman. In May, he told reporters during a Cabinet meeting that Oman “will behave just like everybody else, or we will have to blow them up.”

Iran reiterated Tuesday that Tehran plans to maintain its grip on shipping traffic until Washington meets its conditions.

“Until the United States fulfills its commitments under the agreement, including lifting the blockade, releasing frozen assets, lifting oil sanctions, ending threats and military operations on all fronts, and implementing the other conditions to which it committed, the Strait of Hormuz will not reopen,” said Mohammad Bagher Qalibaf, Iran’s parliamentary speaker and negotiator in previous talks with the U.S.

The Egyptian Foreign Ministry said Tuesday that the Iranian-Omani deal could pave the way for Washington and Tehran to return to negotiations for a “comprehensive and permanent deal that addresses all concerns and enhances regional security and stability.”

The statement came after a meeting between Egyptian Foreign Minister Badr Abdelatty and Omani counterpart Badr al-Busaidi. But it did not address Trump’s latest threat against Oman.

Trump insists the strait is open as more attacks are reported

Trump said on Tuesday the U.S. has no planned talks with Iran but insisted the strait is “open and operating,” despite limited traffic, the reported boat strike and the end on Monday of the 60-day negotiating period between the countries.

Trump wrote on social media that a U.S. blockade of the strait remains “in full force and effect,” adding that all water mines have been removed.

As the talks between Iran and Oman continue, more attacks were reported in the region.

Both missiles fired at the UAE fell into the sea, the UAE Defense Ministry said. No damage or injuries were reported. Iran disputed the UAE’s claim that it launched missiles toward the country.

The ministry later said assessments showed that the missiles targeted maritime traffic. It was not clear whether they directly targeted UAE ships or the country’s territorial waters.

In the weeks after the U.S. and Israel launched a war against Iran on Feb. 28, the UAE was frequently targeted by Iranian missiles and drones, but Tuesday’s reported attack was the first in weeks.

An unidentified projectile hit a ship early Tuesday in the strait off the coast of Oman, damaging the engine room and causing a casualty, according to the U.K. Maritime Trade Operations center.

The monitoring agency did not release any details about the ship or its cargo, and it was unclear whether the crew member was killed or wounded. The agency said the Omani Coast Guard was assisting other crew members and that authorities were investigating.

Elsewhere, the cargo ship that was deemed a loss was struck about 40 nautical miles (74 kilometers) southeast of Mokha, Yemen, the UKMTO center reported.

The center did not identify the projectiles or those responsible.

The Iran-aligned Houthi rebels resumed attacks on commercial shipping in the Red Sea in July and escalated attacks on Yemen’s Saudi-backed government forces. The renewed assaults have threatened shipping through the Bab al-Mandab Strait, another key global trade route.

In other developments, the Houthis claimed they fired drones at an oil refinery in neighboring Saudi Arabia, the latest attack that threatened to reignite Yemen’s civil war and open another front in the Middle East.

The attack targeted a facility run by Saudi Aramco, Saudi Arabia’s state-owned oil company, according to a report by the Houthi-run SABA news agency. There were no immediate reports of damage or comment from Saudi Arabia.

___

Associated Press writer Sally Abou AlJoud in Beirut contributed to this report.

This story was originally featured on Fortune.com

Silicon Valley made a fortune betting on disruption. Now some of the most recognizable names in venture capital and Big Tech are spending millions to disrupt California’s plan to tax them.

Campaign finance records show Palantir cofounder Peter Thiel, crypto billionaire Chris Larsen, Google cofounder Sergey Brin, and longtime venture capitalist John Doerr donated to political action committees opposing Proposition 40, which would levy a one-time tax on billionaires equal to 5% of their wealth if passed. Thiel officially cut ties with California in 2025 ahead of the proposed wealth tax, and Brin has also reduced his official and financial ties to the state.

Larsen gave $5 million to Golden State Promise, a committee opposing Proposition 40, and Ripple Labs, the company he cofounded, has put in another $5 million. 

Another anti-Proposition 40 committee representing teachers, doctors, and small businesses has received $5 million from Building a Better California, whose top donors are Brin and Doerr. Golden State Promise has also received $450,000 from the California Business Roundtable Issues PAC, one of whose top donors is Thiel, who has given $3 million to the PAC itself.  

The stakes are high for the donors. Experts estimate Proposition 40, if passed, will raise $100 billion for California over five years, with 90% earmarked for health care and the rest for food assistance and education. For someone whose net worth is $1.1 billion, the liability is $55 million, according to an analysis from Wealth Management. If the opposition defeats the ballot measure in November, billionaires will avoid that liability. 

Silicon Valley and Washington flashpoint

The multimillion-dollar checks are landing as California’s proposed wealth tax turned into a broader fight about whether taxing billionaire wealth would raise needed funds—or push founders and investors to move out of the state.

Over the weekend, billionaire entrepreneur and investor Mark Cuban publicly sparred over this question with Rep. Ro Khanna (D-Calif.), one of the most prominent defenders of the proposed tax. Cuban argued Prop. 40 misunderstands founders can be billionaires on paper while still being cash-poor and could drive startup talent out of the state entirely. 

“If this passes, only idiot startup founders stay in Cali,” Cuban wrote on X.

Khanna pushed back by arguing truly illiquid “paper billionaires” make up only part of the population the tax would hit, and suggested a workaround in which founders could hand over their shares in the startup to the state in exchange for a loan to pay the tax. 

“The government would still collect from the vast majority of billionaires who are not illiquid,” Khanna wrote. 

Emmanuel Saez, director of UC Berkeley’s James M. and Cathleen D. Stone Center on Wealth and Income Inequality and co-author of an expert report on Prop. 40 arguing the tax asks a fair share from the roughly 250 Californians it would cover—billionaires the report says built their fortunes in the state and can absorb a one-time hit, especially if paid gradually. Saez told Fortune over email founders without the immediate money to pay the tax can “use a deferral option,” paying 5% of “whatever proceeds they take out of their business (as dividends or sales of stock) moving forward.”

“If the business fails, they won’t have to pay anything,” Saez said. “If the business succeeds, they’ll have to pay 5% of that success eventually.”

Khanna has also pushed the fight to tax billionaires beyond California. In March, he and Sen. Bernie Sanders (I-Vt.) introduced federal legislation proposing an annual 5% wealth tax on Americans worth more than $1 billion, with some of the proceeds earmarked for $3,000 payments to lower- and middle-income households.

This story was originally featured on Fortune.com

It’s grape season in Afghanistan’s southern province of Kandahar, and the harvest this year is plentiful. But that’s small comfort for the region’s grape producers and workers, who say fighting between Afghanistan and Pakistan has left them unable to access their primary market.

For months, the two countries have traded fire sporadically across their long, mountainous border, leaving hundreds of people dead. Islamabad accuses Afghanistan’s Taliban government of harboring militants who carry out attacks inside Pakistan — a charge Kabul denies.

With the fighting have come border closures, severing a critical trade route and preventing Afghan producers from reaching what is the primary export market for many goods.

Unable to export their fresh fruit, Afghan grape producers have turned to the domestic market, where the increased supply has sent prices tumbling. To adapt, many producers are now drying their grapes and turning them into raisins — a cheaper product that sells for less.

In a long warehouse in Kandahar’s Zhari district, fans whirl overhead, stirring the hot summer air around bunches of plump, green grapes draped over sticks set up in rows to dry.

But even for raisins, prices have tanked.

Sakhi Jan, who owns a grape orchard in Zhari, says that with 10 people in his household to feed, he’s barely getting by. Seven kilograms (15.5 pounds) of raisins used to sell for around 1,000 to 1,200 afghanis ($13 to $16), he says. Now, the same amount sells for just 400-450 afghanis ($5-$6).

In a country where poverty is rife and malnutrition strikes the most vulnerable, such income losses can be critical.

“We are grateful, but things aren’t like they used to be, the struggle is much harder now,” Jan said. “In the past, work was steady, we used to sell some grapes, and people generally had good work but now the hardship is immense.”

Opening the border crossings and allowing trade to flow once more is critical, he stressed.

“We urge both our government and Pakistan to open these routes and reach an agreement,” he said. “The current situation is causing great hardship.”

Abdul Baqi Bina, the deputy director of the Kandahar Chamber of Commerce and Investment, said the border closure has dealt a severe blow to Afghanistan’s fruit exports — not just grapes, but also pomegranates.

In 2025, five southern Afghan provinces that make up the country’s main grape-producing region exported 44,225 tons — $13.8 million worth of grapes. Nearly all — 43,000 tons — went to Pakistan and the remainder headed to Bangladesh, Iraq and India, Bina said. So far this year, only 256 tons have been exported, at a value of $100,000.

For grape exporter Haji Abdul Hai, it has been a disaster.

“In my 50 years of life, I have never seen these roads closed to the extent they are now,” he said, adding that previous border closures would usually last for a few days, or one crossing would close while another remained open.

“But now, we are facing truly major difficulties,” he said.

Last year, the orchard where Qudratullah Popal worked picking grapes employed about 1,500 workers. This year, Popal said the number has plummeted to around 15.

“There have been good harvests. Grapes turned out well, but the issue is they cannot be exported to other countries. They are being sent to domestic locations … where there is already an abundance of grapes,” he said.

“When routes are blocked and trade halts, everyone’s livelihood is paralyzed, traders, laborers and orchard owners alike,” Popal said.It’s grape season in Afghanistan’s southern province of Kandahar, and the harvest this year is plentiful. But that’s small comfort for the region’s grape producers and workers, who say fighting between Afghanistan and Pakistan has left them unable to access their primary market.

For months, the two countries have traded fire sporadically across their long, mountainous border, leaving hundreds of people dead. Islamabad accuses Afghanistan’s Taliban government of harboring militants who carry out attacks inside Pakistan — a charge Kabul denies.

With the fighting have come border closures, severing a critical trade route and preventing Afghan producers from reaching what is the primary export market for many goods.

Unable to export their fresh fruit, Afghan grape producers have turned to the domestic market, where the increased supply has sent prices tumbling. To adapt, many producers are now drying their grapes and turning them into raisins — a cheaper product that sells for less.

In a long warehouse in Kandahar’s Zhari district, fans whirl overhead, stirring the hot summer air around bunches of plump, green grapes draped over sticks set up in rows to dry.

But even for raisins, prices have tanked.

Sakhi Jan, who owns a grape orchard in Zhari, says that with 10 people in his household to feed, he’s barely getting by. Seven kilograms (15.5 pounds) of raisins used to sell for around 1,000 to 1,200 afghanis ($13 to $16), he says. Now, the same amount sells for just 400-450 afghanis ($5-$6).

In a country where poverty is rife and malnutrition strikes the most vulnerable, such income losses can be critical.

“We are grateful, but things aren’t like they used to be, the struggle is much harder now,” Jan said. “In the past, work was steady, we used to sell some grapes, and people generally had good work but now the hardship is immense.”

Opening the border crossings and allowing trade to flow once more is critical, he stressed.

“We urge both our government and Pakistan to open these routes and reach an agreement,” he said. “The current situation is causing great hardship.”

Abdul Baqi Bina, the deputy director of the Kandahar Chamber of Commerce and Investment, said the border closure has dealt a severe blow to Afghanistan’s fruit exports — not just grapes, but also pomegranates.

In 2025, five southern Afghan provinces that make up the country’s main grape-producing region exported 44,225 tons — $13.8 million worth of grapes. Nearly all — 43,000 tons — went to Pakistan and the remainder headed to Bangladesh, Iraq and India, Bina said. So far this year, only 256 tons have been exported, at a value of $100,000.

For grape exporter Haji Abdul Hai, it has been a disaster.

“In my 50 years of life, I have never seen these roads closed to the extent they are now,” he said, adding that previous border closures would usually last for a few days, or one crossing would close while another remained open.

“But now, we are facing truly major difficulties,” he said.

Last year, the orchard where Qudratullah Popal worked picking grapes employed about 1,500 workers. This year, Popal said the number has plummeted to around 15.

“There have been good harvests. Grapes turned out well, but the issue is they cannot be exported to other countries. They are being sent to domestic locations … where there is already an abundance of grapes,” he said.

“When routes are blocked and trade halts, everyone’s livelihood is paralyzed, traders, laborers and orchard owners alike,” Popal said.

This story was originally featured on Fortune.com

Bitcoin has found some relief after months of selling pressure. On Wednesday, the cryptocurrency jumped nearly 6% to over $69,000, reclaiming a level it had not touched since early June. The jump came right after the Treasury Department announced that it would double purchases of older long-term government bonds.

“The market read this as a quiet form of quantitative easing, a move that weakens the dollar and sends scarce, debasement-hedge assets like Bitcoin higher,” Matt Mena, a senior strategist at crypto research firm 21Shares, told Fortune in a written statement.

Investors quickly piled into those assets, which in turn forced short sellers to cover roughly $1.5 billion in positions by buying Bitcoin in the market. That included purchases of about $700 million in a single minute, an event that 21Shares said may have amounted to the largest short squeeze in Bitcoin’s history.

The rally follows months of weak price action as Bitcoin struggled to recover from a brutal crash last October. Since that rout, which triggered more than $19 billion in liquidations, Bitcoin has fallen about 40% from the $115,000 level where it traded at the time, according to CoinGecko.

Alongside the Treasury announcement, Mena said investors have increasingly priced in a pause in rate hikes over the past two months. U.S. spot Bitcoin ETFs drew roughly $1 billion in inflows during the first two weeks of August, adding another source of demand for the cryptocurrency.

Bitcoin wasn’t the only cryptocurrency to rally following the Treasury announcement. Ethereum and Zcash led major tokens, each rising 9% in the past 24 hours.

A possible bottom

The rally may signal that Bitcoin’s bear market has moved past its worst phase, according to Zach Pandl, Grayscale’s head of research.

“Our best guess is that Bitcoin potentially bottomed at $58,000 earlier this summer… and [that] it’s a compelling time for investors with longer-term horizons to be allocating to Bitcoin and the crypto asset class,” he said.

Pandl said the Treasury’s move highlighted deeper fiscal pressures and could prompt investors to consider alternative stores of value. The national debt is expected to reach $40 trillion before the end of the month, while the U.S. war with Iran has driven inflation higher across the country. Pandl added that recent favorable developments for the crypto industry may have also influenced Bitcoin’s price performance.

On Tuesday, the Securities and Exchange Commission proposed a regulatory framework for crypto assets that could reduce uncertainty as the CLARITY Act remains stalled in Congress. The proposal would exempt eligible crypto firms from certain federal securities rules and make it easier for them to issue tokens and raise capital.

This story was originally featured on Fortune.com

Weight-loss drugs are becoming a popular employee perk, with almost a third of workers saying they’d switch jobs to get GLP-1 coverage. Now, Bank of America is spending $250 million or more yearly on the drugs for its staffers—and CEO Brian Moynihan says the upsides are well-worth the eye-watering cost. 

“What we see is a great impact on the employees,” Moynihan recently said in an interview with CNBC. “We’ve always been about mental wellness, physical wellness.”

It’s part of a wider $2 billion a year wellness package for employee healthcare at the $436 billion bank. Staffers may have to cover the premium or copay for their GLP-1s, but the Wall Street titan is picking up the rest of the bill, amounting to nearly a quarter of a billion dollars annually. And Moynihan says the health investment is worth it to support a healthier workforce. 

“It’s lowering near-term incidents of heart issues for people taking, even if they don’t have all the attributes,” the CEO continued. “That’s the payback.”

The chief executive even acknowledged that Bank of America may not fully realize all the long-term benefits. He noted that some Bank of America staffers on GLP-1s may not see the health upsides until later in life, years after they’ve left the company, but he still believes in the investment.

“It’s the right thing to do for your teammates…We do it because we want to be the great place to work,” Moynihan said. “It’s been fascinating to watch our teammates’ behavior on these adjustments, the loss of weight. We monitor that, we give them coaches and everything, and so it’s a good investment by us.”

Bank of America had no further comment to share with Fortune.

Weight loss drugs are popular—but 60% of firms only offer it for diabetes

In the past couple of years, weight loss drugs like Ozempic, Wegovy, and Zepbound have exploded on the wellness market. 

Now, GLP-1s—originally created to help manage blood sugar levels for people with type 2 diabetes—have become a fixture of millions of Americans’ lives. Around 11% of U.S. adults currently take GLP-1 medications for weight loss purposes, a stark jump from 3% just two years ago, according to a recent Gallup analysis. So companies are steadily expanding their health offerings to meet workers where they are. 

While 60% of employers said they offer GLP-1 coverage for diabetes only, around 36% also cover it for both diabetes and weight loss purposes, according to a recent study from IFEBP.

Earlier this year, consulting giant PwC announced it would no longer cover GLP-1s as an employee benefit for solely weight-loss purposes, blaming “rapidly rising costs.” Instead, the company said it would continue to offer the drug “when prescribed for conditions aligned with established standards of care, such as type 2 diabetes, but [they] will not be included under pharmacy coverage for weight management.” 

Companies are weighing the high costs of GLP-1 offerings for workers

GLP-1s are an increasingly sought-after benefit for talent; around 30% of workers even said they would switch jobs if that got them coverage for the drugs, according to a survey from insurance broker NFP. 

And they’ve gotten cheaper thanks to high demand, manufacturer price cuts, direct-to-consumer options, and new government programs. Now, a starting dose of Wegovy is available for just $149 a month, compared to $1,600 a month when it first launched in the U.S. in 2021. Or in the case of Amazon One Medical’s GLP-1 management program, insured individuals can snag the weight-loss drugs for as low as $25 a month. 

While the drugs have become cheaper, soaring demand and long-term use have put employers in a financial pickle. 

Now, more than a quarter of large corporations are ramping up GLP-1 coverage criteria in 2026 or 2027, according to an analysis from Mercer earlier this year. Around 11% of these big employers have dropped—or are planning to drop—coverage of the drugs for weight-loss purposes this year or next. 

Health services company Cigna stopped covering GLP-1 weight-loss drugs including Wegovy and Zepbound in its employee health plan this July. The company said it made the change “as availability has increased and new options ​have emerged,” but maintained that staffers still have access to weight management programs and resources. 

And HCA Healthcare, which employs hundreds of thousands of workers across its hospitals and medical centers, stopped covering the drugs for weight-loss this January after use of GLP-1s on its employee plan shot up 90% in 2025 alone. It still covers the drug for diabetes. 

This story was originally featured on Fortune.com

Treasury yields are on the march with some analysts suggesting Fed chairman Kevin Warsh is being “tested” by the bond market. But those who know the boomerang central banker well told Fortune that while Warsh will note market “teething” problems, a reaction shouldn’t be expected.

Yields have climbed higher as softer inflation and labor data have dampened the picture for Fed rate hikes, which the market has already priced in. Thirty-year Treasuries sit near 5.3%, heights which haven’t been seen since 2007. The 20-year is around the same mark.

Yields have been elevated since the conclusion of Warsh’s latest press conference following the meeting of the Federal Open Market Committee. In July, markets got the impression that they were perhaps doing some of the legwork for the Fed by tightening financial conditions with higher yields. Warsh also declined, as is his policy, to provide forward guidance, leaving analysts questioning whether the central bank would follow through with hikes.

“It is too early to draw firm conclusions, but the rise in the term premium and bear steepening of the curve following Warsh’s first two [Federal Open Market Committee] FOMC meetings could indicate that the Fed’s credibility is being tested,” said Bassam Nawfal, chief asset allocation strategist at Alpine Macro in a report yesterday.

Warsh’s defenders point out that he has been clear in his intention to bring inflation to heel at 2%. At his first post-FOMC conference in June, Warsh stated: “I’ve said for years inflation is a choice. You bet it is. And today I’m announcing that this Committee, unambiguously and unanimously, have decided we are going to deliver on that.”

The declaration was notable given that President Trump had insisted his nominee would have to be willing to cut the base rate.

Warsh’s credibility at the Fed is clear, Randall Kroszner, a professor of economics at the University of Chicago Booth School of Business, tells Fortune. Professor Kroszner has worked closely with Warsh in the past: He was confirmed to the Fed’s Board of Governors in the same year—at the same hearing—as Warsh, and the pair sat side by side during FOMC meetings until Kroszner left the central bank in 2009.

“There’s a teething process whenever there is a new Fed chair … there were concerns about Jay Powell when he first came in,” Prof. Kroszner told Fortune—speaking last week, ahead of the latest yield jump. “Kevin is very clear that he wants to change the communication strategy, and people … in the press as well as in the markets don’t like change, [they think]: ‘I’m used to this, I know how everything works, and now I don’t know how everything works and I’m frustrated.‘”

“But that’s part of the changeover process … I don’t think Kevin could be clearer about how he really doesn’t want to give forward guidance, he doesn’t want the focus to be on every bump and wiggle in the data. He wants the Fed to think in terms of the bigger picture … and people are finding that frustrating, but I think he’s been very, very clear.”

A market watcher

Prof. Kroszner, like Warsh, worked closely on the Fed’s response to the 2008 financial crisis, chairing the Committee on Supervision and Regulation of Banking Institutions and the Committee on Consumer and Community Affairs. The pair worked closely with private sector stakeholders: Warsh, a former Morgan Stanley executive, with Wall Street, and Prof. Kroszner speaking daily with credit card companies to evaluate the health of consumers.

Wall Street may now be wondering why one of its own is proving so surprisingly unhelpful. Prof. Kroszner suspects—unsurprisingly—that Warsh will still be keeping a watchful eye on markets.

“You certainly don’t want to dismiss what’s happening in the markets, that’s not appropriate,” Prof. Kroszner said. “You want to be aware of what’s happening in markets, but you certainly don’t want to be a slave to what’s happening in the markets … Kevin will be aware of and sensitive to that.”

Economists are divided on Warsh’s approach thus far, with current unease in the bond market just one symptom of that split. Federal Reserve alum Claudia Sahm has suggested Warsh is “long on symptoms and short on solutions.” Jeremy Siegel, emeritus professor of finance at the Wharton School of the University of Pennsylvania, wrote for WisdomTree, where he serves as senior economist, that central bankers have an “obligation to explain the economic framework behind their decisions” and that last month Warsh had fallen short.

Prof. Kroszner suggests that whether or not experts agree or disagree with the approach, Warsh is nevertheless asking “very important questions.” Prof Kroszner added of Warsh’s task forces to examine current practice at the Fed: “Getting outsiders to have input and then have a good discussion at the Fed—as he said, family fights at the table, and he may well get that—because the answers may be controversial, but I think the questions are good ones.”

This story was originally featured on Fortune.com

The United Arab Emirates suspended all trade with Iran on Wednesday after the UAE said it had come under renewed fire from the country — a move that will further isolate the Islamic Republic, which is suffering under U.S. sanctions and a blockade.

Incoming ballistic missile fire triggered nationwide warnings Tuesday night for UAE residents to seek shelter, the first time in weeks such an alarm had sounded.

Early in the conflict, the UAE regularly came under intense fire from Iran, and most trade between the countries, which had once been important trading partners, ground to a halt. In late June, as hostilities eased, some maritime trade resumed, Iran’s state-run IRNA news agency reported.

Beyond the trade in domestically produced goods, “the UAE has been very important for Iran as a re-export hub and has helped the country absorb some of the shocks caused by sanctions,” Mohammad Farzanegan, a professor of Middle Eastern economics at Germany’s University of Marburg, told The Associated Press.

“Iran therefore depends heavily on the UAE, not because the UAE itself produces one-third of Iran’s imports, but because it serves as a major gateway for Iran to access third-country goods and commercial infrastructure.”

The UAE accuses Iran of firing missiles. Tehran denies it

Following the announcement that two ballistic missiles had been fired toward the UAE, both of which splashed down in the Persian Gulf late Tuesday, the Emirati Foreign Ministry said it decided to impose the punitive measures, while also saying it remained committed to “dialogue, cooperation and regional integration.”

The step halted all trade and financial transactions “until further notice,” the ministry said in a statement.

The UAE Defense Ministry said assessments showed that the missiles targeted maritime traffic. It was not clear whether they targeted Emirati ships or the country’s territorial waters.

Iranian Foreign Ministry spokesperson Esmail Baghaei denied that Iran had launched any missiles toward the UAE.

As part of Iran’s efforts to maintain a stranglehold over the Strait of Hormuz, it has regularly attacked ships attempting to use the waterway, including four tankers owned by Abu Dhabi’s state-owned ADNOC oil and gas company over the past two weeks. None of the attacks caused injuries, but they brought harsh condemnation from the UAE and others in the region, including Kuwait and Bahrain.

Since the beginning of the conflict, nearly 20 ADNOC vessels have been attacked by missiles and drones in the Strait of Hormuz, killing one person and wounding another 20.

Iran’s ability to control traffic through the strait, through which a fifth of traded oil and natural gas passed during peacetime, has proved its biggest strategic advantage in the war. While the U.S. and Israel — which launched the war on Feb. 28 — have given various aims, including toppling Tehran’s government and ending its nuclear program, the conflict has devolved into a fight over the strait.

U.S. President Donald Trump insisted Tuesday that the strait was “open and operating” and posted a map depicting it as a U.S. territory. Iranian Deputy Foreign Minister Kazem Gharibabadi called him a “deluded man.”

Ten vessels transited the strait Tuesday, according to the MarineTraffic website, fewer than a tenth the number that typically sailed through before the war began, when there were no restrictions.

During the war, Iran has launched hundreds of ballistic missiles and thousands of drones in strikes that Tehran said were targeting U.S. assets but hit buildings in Dubai and Abu Dhabi, Dubai’s commercial airport, ports and energy infrastructure.

Iran has accused the UAE and other U.S. allies in the Gulf of facilitating American military attacks on Iran, and Iran’s chief of staff, Gen. Ali Abdollahi, issued a new warning Wednesday to “countries on the southern shores of the Persian Gulf.”

“Any assistance or facilitation provided to the aggressor U.S. military amounts to participation alongside U.S. military forces,” he said in a statement distributed by Iran’s semiofficial Fars news agency.

The UAE embargo could put new pressure on Iran

Before the war, the UAE was one of Iran’s biggest trade partners, providing more than 30% of its imports valued at some $21 billion, according to the World Trade Organization’s latest figures from 2024. It was the destination for nearly 13% of its exports worth some $7 billion.

The embargo carries its own risks, however, Farzanegan said.

“As a relatively small country seeking to remain a regional hub for business and finance while attracting tourists and investors, the UAE depends heavily on regional stability,” he said. “Any major conflict with Iran can therefore cause substantial damage to its economy.”

The Emirati announcement comes as the U.S. prepares to apply new economic pressure on Iran. Speaking last week, Treasury Secretary Scott Bessent said the measures would be a combination of economic isolation and the continued blockade of Iranian ports.

The economic pressure follows the intense bombing campaign that targeted industrial and civilian infrastructure in addition to military targets. Already the International Monetary Fund forecasts inflation of nearly 70% this year in Iran and an economic contraction of 5.4%. Meanwhile, its rial currency has hit record lows.

This story was originally featured on Fortune.com

More governors are shifting their stances or taking more steps to squeeze data centers as the midterm elections near and public opinion sours on the energy-hungry behemoths that tech giants and developers are building to fuel artificial intelligence products and cloud computing.

The backlash to the massive server warehouses is enveloping races for governor in some of the nation’s biggest states and presidential battlegrounds.

On Tuesday, Pennsylvania Gov. Josh Shapiro, a Democrat, said his administration would no longer put data center projects at the head of the line when it comes to issuing construction permits or granting developers a lucrative tax exemption if they don’t meet certain standards.

Those include plans to pay the full cost of their electricity and show how they will use advanced technology to limit water use. They also must first win local approval before they can seek state approval. They are, he said, the “strictest guardrails in the nation,” although he stopped short of imposing a moratorium on issuing permits.

Shapiro, considered a potential contender for the White House in 2028, is facing increasing pressure from his GOP opponent, Stacy Garrity, as communities across the state revolt against proposed data centers.

At a news conference, Shapiro slammed what he called “predatory developers” trying to bully local officials and ram through dozens of projects in Pennsylvania — his administration said it counted reports of more than 100 — that likely will never be built because they don’t have the financing, power supply or tech-sector clients to use the space.

“These speculators are nevertheless scaring our communities, being aggressive with township officials, bullying our neighbors, our fellow Pennsylvanians, and refusing to listen to the people,” Shapiro said. “And they are threatening to fundamentally change the character of our communities.”

In particular, he singled out developers aiming to build six campuses of about 50 server warehouses in tiny Archbald Borough that has spawned a community uprising, a lawsuit by one developer and motions by another to force the recusal of six of the town’s seven council members.

In a statement, Garrity said Shapiro “lit the fuse on the chaos we are seeing in community after community.”

Meanwhile, in Texas, Democratic challenger Gina Hinojosa released a TV ad in rural markets Tuesday accusing Republican Gov. Greg Abbott of “selling you out” to data center executives and companies.

The ad airs as Hinojosa, a state lawmaker, has aggressively looked to exploit an undercurrent of discontent in rural, Republican strongholds over data centers’ perceived threat to rural life, ranchland and dwindling water supplies.

Challengers are capitalizing on growing discontent

At one time, both Shapiro and Abbott had been cheerleaders for data centers and actively sought to recruit them, with Shapiro appearing with Amazon officials to announce a $20 billion investment in Pennsylvania, and Abbott, likewise, appearing with Google execs to announce its $40 billion investment in Texas.

But in recent weeks, Abbott ordered regulators to take steps to ensure Texans were not paying higher electricity bills because of data centers, even telling them to hold up data center projects until they complete their work.

He also promised to push a legislative agenda next year to impose regulations on data centers, including taking away the state’s billion-dollar-plus-per-year tax break.

For much of the past year, a growing number of data center projects have met rejection in local zoning or permitting board votes across the U.S., as angry residents pack once-sleepy municipal meetings.

Losing open space, farmland, forest or rural character is a big concern. So is the damage to quality of life, property values or health by on-site diesel generators kicking on or the constant hum of servers. Others worry that wells and aquifers could run dry or electricity bills will skyrocket.

Small, under-the-radar data centers have been around for decades. But the explosion of artificial intelligence chatbots has given rise to data centers that are larger than anything just about any town has ever seen. Some of them dwarf football stadiums and factories and use more energy than small cities.

States trying to tighten the screws on data centers

In some states, governors and lawmakers are trying to force data centers to pay for their own electricity supply, limit their water use, disclose more about their operations and do more to win community support. They are also chafing at the rising tab for the sales tax exemption most states offer data centers.

In Arizona, Democratic Gov. Katie Hobbs, who is seeking reelection, got lawmakers to agree to slap a three-year moratorium on the state’s sales tax exemption for data centers. Hobbs, who had voted to create the tax credits when she was a legislator, called it a “corporate handout.”

New York Gov. Kathy Hochul, a Democrat seeking reelection, ordered a one-year ban on large data centers to give the state time to impose protections for the environment and its energy grid.

Shapiro isn’t the only potential 2028 White House hopeful to step up his criticism of data centers. Illinois Gov. JB Pritzker, a Democrat running for a third term, halted new sales tax exemptions for data centers there until lawmakers impose tougher standards on their operations.

In Ohio, the Democratic and Republican nominees for governor — Dr. Amy Acton and Vivek Ramaswamy — in recent days each unveiled dueling data center policies that called for developers to meet tougher standards before being built.

Data center opposition isn’t necessarily a golden ticket

In Wisconsin, the Democratic nominee for governor, David Crowley, narrowly defeated a challenger who made her call for a one-year moratorium on data center construction a centerpiece of her campaign.

Crowley has taken a more nuanced approach, saying local communities must have veto authority, while also saying data centers are a part of the modern economy and could bring significant economic benefits to the state.

His Republican opponent, U.S. Rep. Tom Tiffany, has attacked Crowley on the issue, including a TV ad released this week where he calls him “Data Center David Crowley.”

___

Associated Press writers Scott Bauer in Madison, Wisconsin, and J.J. Cooper in Phoenix contributed to this report.

This story was originally featured on Fortune.com

Moderna and Merck said Wednesday their personalized mRNA cancer vaccine met its main goals in a Phase 3 trial targeting melanoma, marking the first time a therapy of its kind has succeeded at that stage of testing. The news sent Moderna’s stock up more than 100% and lifted Merck shares as well, as investors bet on a new era for a company long defined by its Covid-19 vaccine.

The companies’ vaccine, called intismeran autogene, is built from a sample of a patient’s own tumor. It is designed to teach the immune system to recognize the specific mutations in that person’s cancer. In the trial, patients with high-risk melanoma, one of the deadliest forms of skin cancer, who had already had their tumors surgically removed received either the vaccine plus Merck’s immunotherapy Keytruda, or Keytruda alone.

The trial included 1,137 patients with high-risk melanoma, cases in which the cancer had grown deep or spread to nearby lymph nodes, or in some cases, to other parts of the body. All of them had already had their tumors removed by surgery before enrolling. Patients who got the vaccine alongside Keytruda went longer without their cancer coming back or spreading than those who got Keytruda alone, meeting the trial’s two main goals.

The companies haven’t released the exact numbers behind that improvement yet, saying only in press releases on Wednesday the results were “statistically significant and clinically meaningful.” They plan to share full data at an upcoming medical conference and bring the results to regulators.

The readout builds on earlier data from the same drug combination, which showed a 49% reduction in the risk of recurrence or death and a 59% reduction in the risk of distant metastasis or death, compared with Keytruda alone. Merck referenced those figures directly in Wednesday’s release. Fortune has previously covered how personalized cancer vaccines like this one work, and Moderna CEO Stéphane Bancel discussed the melanoma data’s earlier stages in an interview with Fortune last year.

Dr. Danish Nagda, an otolaryngologist who has treated head and neck melanoma patients and is founder and CEO of the healthcare startup Rezilient Health, told Fortune just how big this news is. For patients with advanced, stage three or four melanoma, five-year recurrence-free survival today runs around 20 to 35%, he said.

“This potentially doubles it,” Nagda said.

Bancel described the trial as validation of an idea the company has pursued for years.

“For many years, the idea of creating an mRNA treatment designed specifically for an individual patient’s cancer was aspirational. We are now helping turn that vision into a reality,” Bancel said. “Together with Merck, we have started to demonstrate the transformative potential of this technology to address critical unmet needs in the adjuvant melanoma setting.”

Dr. Dean Y. Li, president of Merck Research Laboratories, said the results were evidence for treating cancer earlier.

“By intervening earlier in the course of disease, when many cancers are considered most treatable, the goal of adjuvant therapy given after surgery is to increase the possibility of cure for more patients,” Li said. “We believe individualized neoantigen therapies have the potential to redefine how patients with completely resected stage IIB-IV melanoma are treated.”

“Today’s results represent a landmark moment for adjuvant melanoma treatment,” said Georgina Long, the trial’s principal investigator, medical director of Melanoma Institute Australia, and chair of melanoma medical oncology and translational research at the University of Sydney. “Intismeran in combination with pembrolizumab has the potential to establish a new treatment paradigm in the adjuvant melanoma setting, helping patients remain cancer-free for longer.”

The market reaction

Investors responded immediately to the news as Moderna’s stock more than doubled in early trading Wednesday, while Merck shares climbed as well. Merck is currently valued at roughly $371 billion and Moderna at around $62 billion.

“This makes Moderna a great acquisition target,” said Nagda, who sees the stock move as still behind where the platform’s value should land. “Moderna is still incredibly undervalued. It seems like a large increase, but it’s actually very much underestimating the value of a platform,” Nagda told Forutne. “Now that mRNA has been used in this way to go after melanoma, what stops us from going after other targets? I bet you over the course of the next 12 to 18 months, Moderna will be significantly higher than it is right now.”

Nagda trained at the University of Pennsylvania’s Perelman School of Medicine and completed his ENT residency at Washington University in St. Louis, where he treated patients with head and neck melanoma, often on combination immunotherapy regimens. He said the promise from this study comes from relaxed regulations that have helped move drugs forward, faster.

“This looks good for the Trump administration’s Operation Warp Speed, because this would not have existed without the mRNA vaccine coming out. This accelerated potentially a long-term solution for us to target cancers,” Nagda said, adding he doesn’t expect the treatment to face a difficult path to approval given how strongly oncologists are likely to embrace it for advanced melanoma patients. Fortune has reported on declining public trust in the FDA amid political interference, a backdrop against which any accelerated filing timeline for the vaccine would play out.

The results are promising

Nagda pointed to a factor he said gets little attention in coverage of the trial: rising skin cancer rates tied to climate change.

“Melanoma is not just an American issue. Australia has incredibly high rates of melanoma. It’s a big global issue, and it’s only going to get worse with climate change, as you continue to see more UV radiation and hotter climates,” he said. “Even right now, we’re seeing sunscreen rates going down. Melanoma is going to become more and more prevalent amongst Caucasians, but also amongst other ethnicities.”

Nagda pointed to the safety data as another reason for optimism. In earlier trial data, reactions resolved in about 80% of the 40 patients studied, he said, calling that figure “huge.”

“The side effect profile is minimal compared to a traditional therapeutic for patients with cancer,” Nagda said.

Nagda explained why the vaccine’s side effect profile differs so much from older cancer treatments. Traditional chemotherapy works by exploiting the fact that cancer cells mutate, replicate, and consume energy faster than healthy cells, he said—the goal is to kill the cancer before the drug kills the patient. The mRNA vaccine takes a different approach entirely.

“This is going directly after the cancer cells at a direct level, targeting a unique mutational fingerprint specific to that patient’s own tumor,” Nagda said. “It’s not just personalized across all patients. It’s personalized to the patient’s own tumor.” This is different than traditional chemotherapy, where “our goal is to kill the cancer before the drug kills the human.”

This story was originally featured on Fortune.com

Trying to keep up with AI developments can feel like a losing game, and Jeff Dean, who has spent the last three decades working on the new technology at Google, says Gen Z shouldn’t even try to master it all. Instead, his advice is simple: Skim widely and look for connections others might miss.

“I often tell students it’s better to skim 10 papers than to read one in detail because you then get 10 points in your cloud of what might be possible,” Dean said yesterday at the Asian American Scholar Forum’s 2026 Frontier & Pioneer Symposium in his first public talk since leaving Google. 

“Or, even skim 100 abstracts because what you want to be able to do is connect important ideas that have not yet been connected.”

For young people entering the tech field, Dean’s advice is less about cutting corners than learning how to use time wisely and think broadly. That approach, he said, can identify solutions to problems that previously seemed unsolvable—and help narrow an appropriate timeline. 

A problem that could take 20 years to solve is probably too ambitious if you don’t have a clear idea of how to attack it, he added. But a problem that can be solved in two years may be too obvious to produce a major breakthrough.

“The perfect shape of a problem that you want to work on in a reasonably long-term manner [is] like five years or something,” Dean said. “Try lots of things that might not work. Some of them will.”

AI can put Ph.D.-level expertise in everyone’s hands, according to Dean

Dean stepped away from Google earlier this month after working at the company for 27 years, notably serving as the head of Google AI from 2018 to 2023 and Google’s chief scientist from 2023 to 2026. The 58-year-old is now the cofounder and CEO of DiscoveryLoop, an AI company focused on accelerating scientific and engineering discovery. 

Despite predictions that the technology could lead to massive unemployment and widening wealth inequality, Dean remains bullish on AI’s potential to improve lives.

“The vast majority of the uses of these models is incredibly positive for the world. Like advancing AI in healthcare and AI in education…being able to make people able to solve problems they couldn’t solve on their own will make people able to do more,” Dean said. “And I think that’s super exciting.”

And while Dean acknowledged that even he doesn’t have a “magic answer” and frequently encounters failure, part of his optimism comes from AI’s potential to give people access to expertise that once would have required years of specialized training.

“By building models that are really good at understanding many many different domains of science and engineering you can get Ph.D.-level expertise in a model across many different domains,” said Dean, who graduated with a Ph.D. in computer science from the University of Washington in 1996.

AI leaders are promising a ‘new golden era’—but the hype faces a reality check

Dean isn’t unique in his optimism. Some of the biggest names in tech have made even bolder predictions about what AI could mean for humanity.

Demis Hassabis, Nobel laureate and chairman of Google Deepmind has predicted that AI could radically transform industries like healthcare, energy, and space.

“In 10, 15 years’ time, we’ll be in a kind of new golden era of discovery that [is] a kind of new renaissance,” Hassabis previously told Fortune. In addition to curing diseases, he said he foresees AI unlocking new materials to solve the energy crisis through fusion or solar breakthroughs, eventually allowing humanity to “travel the stars and … explore the galaxy.”

Elon Musk has been even more bullish about AI’s impact. The Tesla and SpaceX CEO believes the advancement will be so great that goods will be abundant and money will not be a major factor.

“Don’t worry about squirreling money away for retirement in 10 or 20 years,” said the world’s richest man on the Moonshots with Peter Diamandis podcast earlier this year. “It won’t matter.”

Challenges, however, persist—especially when it comes to public skepticism. Anthropic CEO Dario Amodei recently acknowledged on X that promises of AI have begun to sound hollow to the public.

“At this point, saying that AI will cure cancer is more a cliche than it is inspiring, and most people think it is deceptive. The thing that will work is actually curing cancer,” Amodei said. “I think by far the most accurate criticism of AI companies including Anthropic is that we haven’t yet delivered on our big promises to benefit the world. That is totally on us.”

This story was originally featured on Fortune.com

What came first—the chicken or the egg? Or in AI’s case, the investment or the revenue?

Nvidia is guaranteeing up to $105 billion for OpenAI’s planned data center campus in Pike County, Ohio, coming in lower than the roughly $250 billion guarantee it was considering, according to reporting by the Wall Street Journal and CNBC. 

The deal moved through at least two known checkpoints before landing at its final size: the Journal reported August 14 that Nvidia had already cut the guarantee to “less than $120 billion,” before the companies settled on $105 billion when the partnership was signed Monday. Nvidia disclosed the final figure—an “aggregate payment obligation” capped at $105 billion—in an SEC filing tied to the announcement. The partnership deal was signed yesterday. The reduction represents a recurring concern among AI investors—the circular cycle of money in the AI ecosystem.

Nvidia and OpenAI did not respond to requests for comment from Fortune.

The Ohio data center is a test of whether the AI boom can generate enough outside revenue to justify the spending being financed from within the AI industry. There is already evidence of market concern from the deal. When reports surfaced in July that Nvidia could guarantee as much as $250 billion, the company’s shares fell about 4.5% intraday from investor reaction to concerns of circular financing.

Reuters also noted anxiety about the sustainability of AI investment remains despite record market performance, with investors increasingly focused on enormous capital expenditures, rising debt and uncertainty over when that spending will generate returns.

Nvidia’s funding is designed to help SB Energy, the SoftBank-backed company developing the campus, secure financing by supporting certain lease and power payments and guaranteeing the value of parts of the completed infrastructure if OpenAI were to default. The structure of the deal substantially reduces Nvidia’s financial exposure to risk.

The rollback comes as Nvidia faces growing questions about a financing model in which the world’s dominant AI-chip maker is increasingly helping finance the infrastructure that ultimately creates the demand for its own chips. The self-funding cycle has been ongoing for years—Nvidia has invested in AI companies and data center operators that purchase its hardware, while also developing financing arrangements intended to make it easier for those customers to acquire more computing capacity.

Last week, Nvidia partnered with six major financial institutions to launch compute-financing platforms targeting more than $500 billion in third-party funding for AI infrastructure—a push that recently got a regulatory tailwind. SEC staff guidance issued in July concluded that certain data-center debt falls outside Dodd-Frank securitization rules requiring sponsors to retain a share of the risk on their own books, making it easier for Nvidia to mobilize outside capital rather than carry the exposure itself.

But Nvidia CEO Jensen Huang disputes the circular financing model. Huang said in a press release the company was “securing long-lived infrastructure for Nvidia compute so OpenAI can deploy the most productive AI factories that can be upgraded repeatedly ⁠with each new generation delivering more intelligence and better economics.”

In the Ohio project, OpenAI will lease the data center from SB Energy for as long as 20 years, while Nvidia will be the exclusive chip provider for the initial phase. The campus is ultimately expected to reach as much as 8 gigawatts of computing capacity, and Nvidia is also investing $1.5 billion in SB Energy.

“The first 800 megawatts are expected to become available in 2028 largely using existing AEP infrastructure,” OpenAI shared in a note. “Further development will require new power plants connected to the grid, including natural gas generation, as well as new transmission lines and associated infrastructure.”

SB Energy plans to build the campus at a former US Department of Energy uranium-enrichment site, with approximately 9.2 gigawatts of natural-gas generation planned to support the broader development. SoftBank and SB Energy are expected to invest billions more in regional power infrastructure.

Nvidia’s graphics processing units, or GPUs, have become the primary computing workhorse for training and running many of the world’s most advanced AI models. Unlike CPUs, GPUs can perform many numbers of calculations simultaneously, making them well-suited to the matrix operations used by machine-learning systems. Nvidia also built a software ecosystem around its chips, including its CUDA programming platform, making its hardware deeply embedded in the development of AI applications.

Nvidia’s investor materials described the OpenAI partnership as an integrated infrastructure offering encompassing architecture, chips, systems, networking, data centers, software, operations and financing. Nvidia said each gigawatt of infrastructure would require roughly $50 billion to $60 billion in total spending, while OpenAI would need to reinvest future revenue to fund its buildout.

The web of deals extends past OpenAI and Nvidia, with partnerships with Microsoft, Oracle, SoftBank, Coreweave and other companies to secure computing capacity to train and operate its models. Many of these arrangements involve companies simultaneously investing, purchasing computing capacity, and building infrastructure from one another.

“We expect to use this capacity to meet growing demand for advanced AI and maintain our lead as the frontier AI research laboratory in pursuit of our mission,” OpenAI said.

This story was originally featured on Fortune.com

Florida Democratic voters delivered another shock upset to the party establishment by nominating state Rep. Angie Nixon, a democratic socialist, over Alex Vindman, a moderate former national security professional who played a central role in President Donald Trump’s first impeachment.

Democrats have hoped to mount a comeback in the diverse, populous and economically dynamic state but have struggled to craft a message that resonates with the state’s electorate. Nixon’s upset sets up a long-shot challenge to U.S. Sen. Ashley Moody, a former state attorney general who Gov. Ron DeSantis selected to fill the seat after it was vacated by Marco Rubio, who Trump chose as secretary of state.

The race has already inflamed tensions within the Democratic Party over how to energize liberal voters eager for unapologetic, combative candidates while not alienating independents and moderates who have been key to winning in battlegrounds.

“If you’re surprised by tonight’s election results, you haven’t been paying enough attention to what’s happening in the South,” Britney Whaley, the southeast regional director of the Working Families Party, which backs populist candidates. “Tonight’s election results must be a wake-up call to a political establishment that believes a populist message can’t win in the South. Angie’s campaign proves voters will respond to a bold economic vision that meets their basic needs.”

A spirited progressive who had the backing of Reps. Rashida Tlaib of Michigan and Ilhan Omar of Minnesota, Nixon recently joined the Democratic Socialists of America. She championed policies like universal healthcare and childcare and has been an outspoken critic of U.S. foreign policy and the war in Gaza.

In May, Nixon protested the Republican-controlled Florida legislature’s redistricting of the state’s congressional maps by shouting through a megaphone during a hearing. She was later reprimanded by an ethics committee but earned plaudits from Democratic allies and voting rights for her demonstration.

Vindman raised about $16 million in his race and had spent more than $9 million by the end of July. Nixon, by contrast, had raised just shy of $1 million. Progressives immediately touted her win as a sign of greater momentum for the region.

Vindman served on the White House’s National Security Council during Trump’s first term. His testimony was central to Trump’s first impeachment over a phone call in which he pressured Ukrainian President Volodymyr Zelenskyy to investigate Joe Biden and his family. Vindman became a national Democratic star and target of Trump’s ire for his actions, a dynamic that garnered him millions in small-dollar donations.

His twin brother Eugene, who also served on the National Security Council, is serving as a Democratic congressman from Virginia.

“Rep. Nixon ran a strong campaign. I will be standing by her side in the fight against Ashley Moody. I hope you’ll join me,” Vindman said in a statement after he conceded the race.

Democratic leaders like Senate Minority Leader Chuck Schumer had hoped Vindman’s reputation and campaign war chest would help turn what election analysts had considered a solidly Republican seat into a more competitive race. But Nixon’s upset victory has now buoyed already high Republican confidence in the state.

Once an archetypal political background, Florida has shifted to the right since 2016. Trump himself moved his residence to his Mar-a-Lago resort after leaving the White House in 2021 following his first term.

This story was originally featured on Fortune.com

Meetings have become a constant annoyance for white-collar professionals: they often drag on, interrupt focused work, and don’t require everyone’s participation. Now, business leaders are protecting their time by setting boundaries around when they join the conversation. Dropbox co-CEO Ashraf Alkarmi follows a simple rule to decide what calls make the cut.

“I pick things that I can make significant progress on in that quarter,” Alkarmi said in a recent interview with Business Insider. “And so it becomes a filter for how I prioritize my time.”

To pick and choose which meetings to attend, Alkarmi sets five goals he needs to achieve every quarter at the $7.3 billion cloud storage company. He uses these important target areas—like people, business, and performance—to shape his weekly schedule and decide where to focus his efforts. 

To make sure everyone is on the same page, the Dropbox co-CEO shares his key priorities with his circle, chief of staff, and administrative team. These objectives “win at all times” when a lot falls on his plate, he explained.

For example, during the first quarter, Alkarmi combed through data and worked toward solutions to reduce customer churn. Making “meaningful progress” and moving that needle became one of his main priorities on Dropbox’s business side. To keep all his ducks in a row, the leader said he uses the daily task-tracking app Trello to stay on track with his quarterly goals. Sticking to his list of five also narrows down what conversations are worth a chunk of time in his busy schedule. 

“Sometimes I get meetings that are not related to these things, and I don’t go,” Alkarmi continued.

Fortune reached out to Dropbox for comment.

CEOs have their own meeting rules: call-free afternoons, fewer one-on-ones, and later start times 

Like Alkarmi, other CEOs are protecting their calendars from unnecessary meetings and treating uninterrupted focus time as a resource worth guarding.

Southwest Airlines CEO Bob Jordan has called out the fact that meetings are crowding out the actual work that really needs to get done. For that reason, he set a 2026 goal to keep his calendar completely clear every Wednesday, Thursday, and Friday afternoon, blocking anyone from booking calls during those hours. 

Jordan acknowledged that approach might sound “crazy” to some executives, but he reasoned that CEOs are hired to do work only they can do—and that rarely happens if they’re trapped in back-to-back meetings.

“When you first start, it’s easy to confuse busyness and going to meetings with leadership,” Jordan said on a panel of CEOs at the New York Times DealBook Summit last year. “Because what we all find, I’m sure, is there’s no time to ‘work,’ and you confuse going to meetings with the work.”

Similarly, Jensen Huang, the cofounder and CEO of $5.3 trillion technology giant Nvidia, has trimmed the fat from his work routine by prioritizing efficiency over regular check-ins. The chips leader doesn’t believe that frequent catch-ups with his 55 direct reports are the best use of his time, given that a continuous stream of meetings would only clog up his work schedule and slow him down. Instead, he frees up space for broader, team-wide collaboration—which Huang said also helps maintain transparency within one of the world’s largest companies.

“I don’t do one-on-ones with any of them, unless they need me; then I’ll drop everything for them,” Huang said at the Stanford Institute for Economic Policy Research summit in 2024. “They never hear me say something to them that is only for them to know. There’s not one piece of information that I somehow secretly tell the staff; I don’t tell the rest of the company.”

Airbnb CEO Brian Chesky has established his own rules around the flow of his day. He believes that no leader should apologize for how they choose to run their businesses—and he’s unabashedly following his own advice. Even though many leaders operate on a rise-and-grind mindset, Chesky hits his creative stride later into the night. For that reason, he’s set boundaries around when he takes meetings, barring any calls before 10 a.m.

“When you’re CEO,” Chesky told The Wall Street Journal last year, “you can decide when the first meeting of the day is.”

This story was originally featured on Fortune.com

Wealth, perhaps second only to loyalty, has been a prized attribute for Donald Trump during his second presidency.

The number of people worth at least $100 million whom the Republican president has appointed to his administration is more than four times the combined total under the three previous presidents, according to a report from the consumer advocacy group Public Citizen.

It’s the numbers for the Trump administration, in the context of previous administrations, that stand out in Public Citizen’s report, published on Monday. In all, 57 Trump officials are worth at least $100 million, including 17 ambassadors and the remaining 40 in senior posts across the executive branch.