Each year, the editorial team at MIT Technology Review puts together a list of 35 innovators under 35—a group of researchers, inventors, and other young minds worth following.

The team worked on the newest edition of the list for months, and the final slate includes nine individuals from all over the world in the climate and energy category. Each one has a fascinating story and is tackling an important challenge.

I think it’s worth zooming out and considering the energy and climate awardees as a group. Taken together, these innovators and their work can tell us something about where climate tech is at this moment—and where it’s heading.

AI is the dominant technology story, both for its potential and its challenges.

We split the innovators into four main categories this year: biotech, climate and energy, computing and robotics, and AI. It probably won’t surprise you that AI features heavily in the work of many innovators in other categories.

Climate innovator Jae-Won Chung, for example, built software to make AI more energy-efficient. By measuring the energy demands of open-source models, he hopes the industry can better understand and address the impact of AI. (If this work sounds familiar, it’s because we spoke with him last year for our investigation into AI’s energy demands.)

But AI also has the potential to improve many areas of research. Jing Wei is using AI to track pollution more effectively, essentially using machine learning to fill in gaps in data from disparate sources like satellites and weather stations. Zhonghua Zheng developed AI climate models that work better for cities, a well-known blind spot for traditional models.

We need better ways to get the critical materials used to build new technologies.

As we begin to rely on new technologies to power our world, we’ll see a major shift in the materials we need to build them.

Lithium is a prime example: The metal underpins lithium-ion batteries, which are crucial not only for electric vehicles, but also for large-scale energy storage on the grid. We could face lithium shortages as soon as this decade, and the prospect of supply crunches applies to other critical minerals, too—copper is another one to watch closely.

Brine is currently the cheapest source of lithium, but the process to get the metal out can take months and harm the local environment. Mohammad Alkhadra is the cofounder and CEO of Lithios, a startup working to quickly and efficiently extract lithium from brines.

Hardrock ore is the most common source of lithium, but it’s more expensive than brine. Benjamin Mowbray cofounded and serves as CTO for Rock Zero, which is working to extract lithium from hardrock ore.

Addressing climate change will require overhauling all corners of our society, sometimes in surprising ways.

To reach net-zero greenhouse gas emissions we will obviously need to rethink major sectors, like the electrical grid and transportation, to move away from fossil fuels. But outside these primary sources of climate pollution are seemingly infinite, less obvious problems to figure out, too.

Heavy industry, including steel production, is a major one, making up about 7% of global greenhouse gas emissions. Laureen Meroueh is making cleaner, cheaper steel using a new kind of furnace that simplifies the chemical process required to produce the metal.

Plastics are generally made with fossil fuels, so we’ll need alternatives to this incredibly useful category of materials. Joseph Nguthiru is making a bioplastic replacement for fossil-derived packaging that uses an invasive weed. Also using available materials in a creative way, Diana Orembe is making fish food for aquaculture with food waste.

And refrigerants are often incredibly powerful greenhouse gases. Jinyoung Seo is developing solid refrigerants that could eliminate worries about leakage. A device using these materials could reduce energy consumption by 20% compared to conventional technology.

I’m constantly learning about new challenges we face in the climate and energy world, and I’m often surprised by the ideas people are coming up with to address them. For more on all the under-35 innovators and their work, check out our full 2026 list.  

This article is from The Spark, MIT Technology Review’s weekly climate newsletter. To receive it in your inbox every Wednesday, sign up here. 

Multiple cameras tracked a mouse as it wandered around a small arena. A computer charted its position and speed, leaving Pong-like traces on a monitor. 

The reason to watch this rodent so carefully? Nearly half its brain volume had been replaced with human cells.

The effort to mix the brain tissues of distant species is being reported today in the journal Nature by a team at Stanford University, led by neuroscientist Sergiu Pașca. 

Pașca’s group previously showed that human brain “organoids”—small blobs of neural tissue—could survive, and even function, after being injected into the heads of baby rodents.

Now, Pașca has taken things a step further by genetically modifying mice so their brains don’t fully develop in the first place. These modified mice are missing most cells of both the cortex and the hippocampus, two key brain areas.

That creates much more room for the human cells to take hold, he says. “Human cells that are placed in these animals will divide, will grow, and within a few weeks to a few months they will take most of that space,” he says. Pașca says one surprising discovery is that the mice lacking brain tissue seemed fairly normal—they walked around and squeaked. But they did have memory problems. In a maze test, they couldn’t remember what parts they’d explored. 

The mice with the added human cells, by contrast, performed better on the maze test. That means the human tissue is playing some role in the animals’ cognition.

Pașca believes what he is calling “xenocortical mice” could be useful in studying brain injuries. However, the report is also a dramatic demonstration of “the combined power of genetic engineering and stem-cell technology to reshape biology,” says Carsten Charlesworth, a scientist who works in a different Stanford lab and was not involved in the research.

Already, brain organoids are being tested in labs to see if they can be connected to computers to play video games. Other scientists have proposed using them like replacement parts to treat stroke victims. 

“What’s most remarkable to me is the extent to which human neural tissue introduced after birth grew and connected with the mouse nervous system across a species barrier,” says Charlesworth. “As these technologies advance, they’ll increasingly force us to challenge our traditional assumptions.”

Last year, Pașca convened a group of ethics experts to study the implications of neural organoid technology, including the odds that an animal could develop human consciousness and the risk that “organoid therapy clinics” might offer scam treatments to desperate patients.

For now, he says, he’s not concerned that the rodents have any type of human cognitive capacities. That is because their brains are relatively tiny and the evolutionary distance between man and mouse is so great. 

But that’s also why Pașca says this type of experiment should not be carried out on higher species: They could end up with large volumes of functioning human brain tissue, potentially blurring the cognitive boundaries between people and animals. 

Pașca specifically cautioned against adding human brain organoids to a monkey engineered to lack a cortex.

“One of the things that I see as a very clear red line is doing this experiment in a primate,” he says. “I don’t think that is justified at this point in any way.”

Last year Ruxandra Teslo, a policy analyst who focuses on clinical trials, posted an idea for supercharging medical AI systems: Use data from failed biotech companies.

By bidding at their bankruptcy proceedings, she proposed, it might be possible to obtain detailed regulatory filings, manufacturing strategies, and safety data—types of information usually considered trade secrets. She called these documents “biotech’s lost archive” and said they could be used to help train AIs that would act as powerful copilots in the often opaque drug approval process. 

Today the OpenAI Foundation, the nonprofit parent of OpenAI, said it would fund her idea as part of a new effort it calls Public Data for Health, which aims to help artificial intelligence make big leaps in medicine by paying to create “high-quality scientific datasets.”

The basic idea is that AI isn’t going to be capable of making important breakthroughs in curing disease unless researchers can feed the models much more information than they have so far. 

“Everyone is recognizing that data is the biggest bottleneck in successfully applying AI to biology,” says Morgan Levine, a former vice president for computation at Altos Labs, a longevity company.

In its initial round of data grants, the OpenAI Foundation also announced that it would give $40 million to a program to collect data about novel cancer vaccines at the University of North Carolina, Chapel Hill, and support OpenAdmet, a group that runs competitions in which researchers try to predict drug effects. 

Teslo’s idea for a biotech archive received $500,000 and will be pursued by 1Day Sooner, an advocacy group representing clinical trial volunteers, which she advises.

“We expect many remaining breakthroughs in preventing and curing disease to come from pairing the intelligence of new models with more observations of the world—in other words, more data,” the OpenAI Foundation said in a statement.

OpenAI started as a nonprofit, but leader Sam Altman restructured it to form a for-profit corporation that develops new models, launches products, and is now planning an initial public offering of stock that could value it at $1 trillion.

Because the foundation holds a 26% equity stake in OpenAI, it is now be on track to become the richest charitable organization on the planet, potentially sitting on $250 billion in stock value. (By comparison, the Gates Foundation and a trust associated with it held about $180 billion at the end of 2025.)  

Making good use of that kind of money will not be easy. The foundation, based in San Francisco, is still hiring for many key roles and started ramping up its grantmaking only this year. Its largest single gift so far, of $100 million, was awarded in August to the Common Health Coalition, an organization that helps patients get access to drugs for hepatitis C.

OpenAI’s charitable efforts come even as apocalyptic fears have broken out about the possibility that runaway AI could wipe out all human life, possibly by launching a deadly bioweapon.

Those fears have been stoked by AI company insiders, some of whom say the chance of human extinction within the next decade is 10% or more. Last week, Altman and xAI founder Elon Musk both endorsed a call by Anthropic CEO Dario Amodei to “slow the pace at which we improve the capabilities of AI models” so that risk prevention can catch up.

Jacob Trefethen, an executive at the foundation, says it essentially operates separately from OpenAI but shares an official mission of ensuring that artificial intelligence “benefits all of humanity.”

“We’re starting grantmaking when we think the best way to achieve that mission is to make grants to external nonprofits, research institutions, and other third parties,” Trefethen said in an interview. He says the foundation hopes to give away $1 billion by the end of the year. 

The $500,000 grant to 1Day Sooner will help the group prove it can obtain the data troves of bankrupt companies, says the organization’s president and cofounder, Josh Morrison. He thinks nonexclusive copies of company datasets could be acquired for only “a few tens of thousands of dollars” each.

His organization is currently in possession of three datasets, two of them donated by Lumen Bioscience, a biotech that previously used the Chapter 11 strategy to gain insights into another company’s drug development efforts. 

Morrison says two other attempts to obtain drug company files this year proved unsuccessful, after 1Day Sooner’s bids were not accepted. 

Bankruptcies could become what some are calling a “new land grab” for AI training. Last month, Google won a bid to take over the corporate data of the failed carrier Spirit Airlines, including 100 million emails. That led to objections from flight attendants and others who worried that private or proprietary data could be exposed. 

The drug company files that 1Day Sooner is seeking are known as common technical documents. They typically contain the back-and-forth between companies and regulators, as well as detailed scientific and medical measurements, and essentially provide everything that is known about a drug.

According to Teslo, who is a writer for Works In Progress and a nonresident fellow at the Institute for Progress, a think tank in Washington, DC, a stockpile of such files could help turn an AI into a regulatory expert, which in her view could be one of the main ways AI helps speed cures to market.

“People say ‘We will invent AI, and AI will cure cancer,’ but that’s very removed from the messy reality and the regulatory process,” she says. “About 70% of the money and time in drug development is spent in clinical development—organizing the trials and testing the drug—but despite that, the process is basically a black box, especially for small biotech companies generating the innovations.” 

When Jessica Wachter, a finance professor at the University of Pennsylvania’s Wharton School, wanted to assess AI’s impact on the economy over the next few years, she faced a long list of business and technical uncertainties. So she started with what she calls a “remarkable fact” that is not in question: A handful of so-called hyperscalers are investing huge amounts of money to build AI data centers.

Instead of trying to predict how useful and widely deployed AI models will be, she simply asked how fast the hyperscalers’ earnings will need to grow to justify their spending through 2027, when—she and her collaborator estimate—expenditures will reach nearly $1.1 trillion. It’s a no-nonsense accounting approach to making sense of today’s historical AI buildout.

The results are eye-opening: The AI companies will need to increase their own productivity by a factor of 2.7 to break even by 2030, accounting for the cost of capital and a 15% return, and depreciation of the assets. Not impossible, says Wachter. The result would lead to the kind of economic growth that we saw during the US IT boom over a period of about 10 years starting in the mid-1990s. But, she says, for it to happen by 2030 “that’s a lot of growth compressed into a few years.” And if the hyperscalers cannot meet such profit goals?

“Then they will fall behind on their interest payments, and that risks bankruptcy,” says Wachter, who was previously the SEC’s chief economist and director of its division of economic and risk analysis. If a productivity boom “fails to materialize,” she and her coauthor conclude in their research paper, “the current buildout will be the largest misallocation of capital in history.”  

It doesn’t take superintelligence to realize that today’s large investments in the infrastructure for artificial intelligence come with huge risks. The hyperscalers will spend about $750 billion this year, building massive data centers scattered across the country. And the spending spree shows no signs of slowing. According to some projections, total AI capital investments from the hyperscaler companies—Alphabet, Microsoft, Amazon, Meta, and Oracle (which partners with OpenAI)—could be more than $5 trillion over the next four years.

It’s one of the largest capital investments by any industry in history. But there’s a problem that’s obvious to anyone paying attention.

While the hyperscalers plan to spend trillions, total AI revenues will be around $150 billion to $200 billion this year, says Gary Gensler, who ran the SEC during the Biden administration and is now a professor at MIT’s Sloan School. “The challenge is that the spending does not have commensurate revenues yet. That’s a fact,” he says. “And then the question is, is that an investment that will be paid off in the future?”

At stake in that trillion-dollar question is the financial health of the giant AI companies and the overall US economy—the investments could soon balloon to around 3% of GDP. The answer could also determine the fate of the hugely expensive data centers themselves. 

No one really knows how profitable and useful these multibillion-dollar behemoths will be down the road. Though AI models have made dazzling progress over the last few years, it’s anyone’s guess how much compute capacity we will need. The technology could become more efficient and therefore less dependent on raw computational power. Or demand for AI products could slow, or customers could turn to cheaper models.

The risks, both to investors and to the economy, have become even greater this year, as these AI companies have begun borrowing large amounts of money to build more and more data centers. Free cash flow—operating cash flow minus capital expenditures—is expected to soon dip into negative territory for the group. Even Alphabet, known for generating and hoarding huge amounts of cash, reports in the latest quarter that its impressive revenues of nearly $120 billion were devoured by AI infrastructure spending, leaving it with a free cash deficit of some $5.9 billion—its first shortfall since Google went public in 2004.

In the near term, it’s not a big financial worry for most of the companies. They make a lot of money and have very deep pockets. But debt is expensive, and some investors are losing patience. If future demand for the data centers’ computation power drops, the companies will still be on the hook to pay back the borrowed money. What’s more, the risks are spreading to the rest of the economy as the loans get passed along via various financial mechanisms. 

It won’t be enough to simply cover the enormous price tags of the new data centers. Hyperscalers will also have to pay for the rising costs of capital as they borrow more money. They will need returns that are impressive enough to justify all their spending to investors and creditors. And to add to those concerns, they will have to make up for the depreciation of billions of dollars in chips housed within the facilities—a ticking time bomb buried in the investments.

Performance of the expensive GPU chips at the core of the data centers—such compute electronics represent some 60% of costs—is roughly doubling every two years or so. The pace of progress helps explain the increasing wizardry of the AI models, but it comes with a cost. Owners of AI data centers that come online this year and next will need to spend billions more on the next generation of chips by the end of the decade if they want to stay competitive. Without the investments, says Mihir Kshirsagar at Princeton’s Center for Information Technology Policy, the data centers risk becoming “hulks,” stranded assets “scattered all over the place.”

To put it bluntly: The AI companies need to start making a lot more money. And they need to do it fast. But juicing their earnings alone still won’t be enough to sustain their data-center investments for the long term.

Productivity is everything

At some point, AI is also going to have to create broad economic growth to justify continuing the hyperscalers’ spending spree.

Sloan’s Gensler describes today’s large investments into AI infrastructure as “a parlay bet by the capital markets and the economy.” That means success will require winning three related but independent wagers: Hyperscalers must generate massive revenues, AI must boost widespread economic growth, and both must happen while the powerful but expensive so-called frontier models that rely on the data centers fend off cheaper versions, which many businesses might find good enough.

What makes this so tricky is that each wager depends on the other two but also poses its own challenges.

If the hyperscalers continue to spend huge amounts of money on data centers into the next decade, revenues will need to skyrocket into the trillions. Stijn Van Nieuwerburgh, a finance professor at Columbia Business School, bases his estimates on a scenario in which about 183 gigawatts of planned AI compute capacity is built between 2025 and 2032; he calculates that each gigawatt costs about $41 billion. Assuming a 10% return—the minimum that would be acceptable to most investors—“required” annual revenues will be roughly $3.7 trillion by 2032, he says.

Others get a similar number.

Winning the second part of the bet—productivity growth across the economy—will be crucial to achieving such numbers.

For a few years, AI companies could likely boost their revenues by simply selling subscriptions and tokens to all the businesses clamoring to get into AI. But eventually—and this might be happening already—those paying customers will need to justify their expenses by seeing bottom-line benefits from the technology. AI will need to fulfill its promise of making workers more productive and making businesses more efficient and profitable while expanding their products and services.

In economic jargon, that means customers will need to see productivity growth. Taken together, these results will mean the country is prospering and growing.

“If you don’t get the productivity gains, at some point people are going to sour on AI, and that will bring down investments and it would also limit revenue growth,” says Daron Acemoglu, an MIT economist and 2024 Nobel laureate. For the investments to be sustainable over, say, the next five to 10 years, we definitely “need to see productivity gains,” he says.

Most economists who watch the numbers closely agree that, for now, the economy-wide statistics show little or no productivity growth from AI. There are some hopeful signs it’s on the way, though. In a recent survey of some 6,000 senior business executives in the US, the UK, Germany, and Australia, the vast majority—around 90%—report no increase in productivity over the last three years. But they expect a boost of around 1.45% in total over the next three years; US executives anticipate a 2.25% bump over that time. 

In a follow-up survey, the respondents also reported plans for their businesses to spend more on AI, leading the authors to anticipate some $280 billion in private-sector AI expenditures by the end of 2026.

That’s good news for the hyperscalers. But it comes with a dose of bad news for those worried about AI’s impact on jobs. The executives expect to increase the productivity of their companies by increasing their sales while significantly cutting the number of employees.

If AI improves productivity by destroying jobs, public backlash to the technology—the kind we have seen around data centers, for example—will likely get worse. Perhaps it’s worth adding one more wager to the parlay bet described by Gensler: The public and local communities must feel that they are also benefiting from the massive investments in AI.

And let’s not forget how interdependent these wagers are; if productivity growth comes from companies running models like DeepSeek, then the hyperscalers’ revenues could collapse. If productivity comes from cutting jobs, a public backlash could block many of the planned investments—and stunt anticipated revenues. We will need to win all the wagers for the hyperscalers’ bet to pay off. 

We’re all part of the AI gamble now

It was one thing when the AI companies were spending cash they had accumulated over the years to build their own data centers. Then the risk was largely limited to their own balance sheets and shareholders. But it’s a higher-stakes game when much of the money is borrowed. Morgan Stanley, for one, calculates that more than half of the $2.9 trillion that hyperscalers will spend between 2025 and 2028 to build AI data centers will be financed with “external capital.”

The borrowing is leading some of the companies to engineer complex webs of financing that are becoming intertwined with much of the rest of the economy. “A lot of financial institutions, directly or indirectly, are exposed to these data centers either as lenders, or as guarantors of some of the debt, or as backers of the private credit funds who are funding these data centers,” says Columbia’s Van Nieuwerburgh. “People don’t even know they’re holding this stuff. It’s somewhere deep inside their pension fund. Ultimately, it’s backing their life insurance policies. And that risk is getting distributed everywhere in places that are invisible.”

As the investments in data centers have spiked, the financial engineering has become more byzantine.

Take, for example, Meta’s so-called Hyperion data center under construction in Richland, Louisiana. When the company announced the two gigawatts of compute capacity at a price tag of some $10 billion in late 2024 it was Meta’s largest planned data center. Greeted with much enthusiasm by state and local politicians, the project, located in the rural northeast corner of the state, was seen as a boon to the community. Entergy Louisiana, the state’s largest utility, rushed forward with proposals to build three large natural-gas power plants to service the massive data center.

Then last fall—the projected cost was now $30 billion—the financing got a lot more complex and, to some in the community, a lot more disconcerting. Meta transferred an 80% stake to the large (and troubled) private-credit firm Blue Owl Capital, forming a joint venture called Beignet (like the famed New Orleans pastry) to raise financing for the data center. Meta then signed a series of four-year leases with the joint venture, an arrangement that the company says gives it “long-term strategic flexibility.” To backstop the agreement, Meta provides the venture with what is called a residual value guarantee, in which it will make a cash payment to cover the value of the facility “following any non-renewal or termination of a lease.” Got all that? 

I hope so. The financial wheeling and dealing is actually even more convoluted, with a cast of wholly owned subsidiaries and LLCs. Beignet has set up Laidley LLC, which owns and operates the site as the landlord. In turn, Laidley leases the facilities to Meta’s wholly owned subsidiary Pelican Leap LLC, which is the tenant. And there is a series of four-year leases that cover the different buildings that make up the data center campus. 

It’s not a coincidence, says Van Nieuwerburgh, that the length of the leases matches the expected lifetime of the data center’s GPUs. While Meta has to pay off its loan if it terminates the leases early, that will still leave its investors “with an empty building and no cash flow,” he says. “And then they need to find a new tenant for a huge data center, and good luck with that.”

Meanwhile, Meta is doubling down on its bet. In July, the company announced it was expanding the data center to five gigawatts of compute capacity. The total price tag is now $50 billion (so far, Meta hasn’t said whether Blue Owl will be involved in financing the expansion). Meanwhile, Entergy is now planning to build seven more gas-fired power plants, bringing the total capacity of the facilities to around 7.5  gigawatts—some six times the amount of electricity used by New Orleans.

SCOTT BALL/THE NEW YORK TIMES VIA REDUX PICTURES

If the complex financing is a puzzle to many investors and even financial experts, it is even more baffling to those directly affected by the construction of the data center. The main worry concerns how Entergy’s spending on the natural-gas power plants will affect electricity prices, and who will be left paying the bill for the power if Meta walks away.

Entergy says it has a 20-year guarantee from Meta that the company will purchase electricity over that period to cover the costs of the power plants and related infrastructure.  But there are skeptics, especially given how fast the fortunes of the AI industry are changing. “In four years, is Mark Zuckerberg still going to be interested in this? Or is he going to throw in the towel?” asks Paul Arbaje, a senior analyst at the Union of Concerned Scientists, which has been advocating, largely unsuccessfully, for the Louisiana Public Service Commission to provide more transparency around the data center and its financing.

Even if the 20-year deal holds, consumer advocates are worried that Meta or its partners won’t fully cover all the costs, including those associated with operating and maintaining the power plants—and those additional costs that could be passed on to residential ratepayers. What’s more, says Logan Burke, the executive director of the Alliance for Affordable Energy, if Meta doesn’t end up needing as much power as Entergy planned (these projections are not public), consumers could be left paying for the surplus produced by the plants.

And if Meta terminates its leases early? “It gets complicated very quickly,” says Burke, who questions whether the shifting roster of financial entities will honor existing agreements. “That everybody is going to do what they’re saying they’re going to do over the next 20 years is just hard to believe.”

For UCS’s Arbaje the bottom line is this: “They’re making huge bets that these data centers will be worth it. Bet with your own money, not with ratepayer money.”

After the bubble

Predicting when the AI investment bubble will burst is a fool’s errand. But there is little doubt a day of reckoning is coming, given the irrational exuberance that has overtaken the hyperscalers and their investors. Of course, you might argue that this time is different, and that the rules of accounting and lessons of economic history don’t apply—that AI is too transformative. Maybe, but don’t count on it.

“History tells us that at some point you get a retrenchment, and it’s just a question of when and how severe,” says Sloan’s Gensler. It could be that today’s $750 billion spending rate “goes flat” or decreases next year. Or, he suggests, “we’re now in 2028 or 2029, and then all of sudden they’re retrenching because they’ve got enough capacity.” But, he adds, “you can be pretty assured there’ll be a retrenchment.” 

Though a so-called retrenchment might be inevitable, it’s worth keeping in mind that the fates of the financial bubble and the underlying AI technology revolution could be very different. Already, some Silicon Valley insiders are rooting for a crash; in a recent blog post the longtime venture capitalist Vijay Pande wrote that “the coming crash would be the best thing that happens to this technology.” The argument makes some sense. A crash could make AI investments more rational, calm the impulse to build billion-dollar data centers on every vacant field that CEOs fly over, and refocus investors on how to use the technology to create sustainable value.

But we should probably be careful what we wish for. After the bursting of the dot-com bubble at the beginning of the 2000s, hundreds of thousands lost their jobs, large and small companies alike went bankrupt, the economy of Silicon Valley and San Francisco was decimated (at least for a while), and the shocks sent the US into a mild recession in 2001. For the financial community and many tech workers, it was no fun.

Even more devastating for the economy and the average American was the great recession that began in late 2007. Comparing the financial engineering leading up to it and the methods deployed by hyperscalers today is sobering. So-called special purpose vehicles (SPVs) are back! If Columbia’s Van Nieuwerburgh is right about the dangers of letting investments from the hyperscalers get entangled throughout the economy, the fallout could be severe.

But technologies survived and even prospered in the aftermath of both downturns. The early 2000s, even in the face of the dot-com fiasco, were a time of great innovation and tech optimism. The froth came off the spending on silly technologies, helping to focus investments on more promising ones. It’s no coincidence that each of the hyperscalers rose out of the ashes of the crash or started up shortly after. The fiber-optic infrastructure built during the feverish telecom bubble that ran parallel to the dot-com one is still the backbone of much of today’s communication infrastructure; we wouldn’t have Facebook or Amazon or Google without it.

This time, however, we’re facing a unique risk: The huge financial investments by the hyperscalers have ensnared the future of AI itself with the fortunes of the massive data centers spreading around the country. The logic is founded on a deeply held belief about the power of scaling in AI; the bigger you build it, the smarter it gets. That might be true, but it’s unproven and a risky bet.

There are already plenty of red flags, from strong public opposition to the construction of new data centers to the competitive threat from cheaper, good-enough AI models to the rapid improvement of small, local AI models. None of these trends point toward a future dominated by frontier models housed in massive, billion-dollar data centers.

The financial bubble around the colossal spending by the hyperscalers will likely burst eventually—or maybe soon. It might be financially painful, but we’ll survive. Wall Street will survive. AI itself will survive, though it may look different and lose some of today’s hubris. The financial fate and future utility of the massive data centers fueled by trillions of dollars of spending, on the other hand, are far less certain.

This story appeared in The Algorithm, our weekly newsletter on AI. To get stories like this in your inbox first, sign up here.

This weekend, Dario Amodei, CEO of Anthropic, posted an essay calling for a brake on the pace of development of LLMs. Amodei cites the looming dangers he sees from the technology, from its use in cyberattacks and bioterrorism to its potential to wreck the economy. The heads of the other three top US AI labs—OpenAI CEO Sam Altman, Google DeepMind chairman Demis Hassabis, and SpaceXAI CEO Elon Musk—voiced their support. “Dario is right,” Musk wrote on X.

Think about how surreal that agreement is for a moment. Just a few months ago, Musk and Altman sat in court attacking each other’s reputations in a (failed) lawsuit that Musk brought against his former OpenAI colleague that was—on paper at least—about whether or not Altman was a trustworthy steward of such dangerous technology.

Amodei’s rift with OpenAI is even deeper. Anthropic was founded in 2021 because Amodei didn’t think Altman took the risks of the technology they were building seriously enough. Anthropic and OpenAI have been competing in a winner-takes-all race ever since. (Hassabis has stayed out of the drama, but his company remains a rival.)

Now, it seems, they’re all in agreement: The latest generation of LLMs aren’t safe and everyone needs to figure out what to do about it. The public messaging from the top AI labs has taken a doomer turn.

It’s easy to be cynical. It’s not at all clear what any of them mean by a slowdown or how it would work. These companies also care a lot about how they come across. With trillion-dollar IPOs in their sights, OpenAI and Anthropic need to reassure investors that they’re the grown-ups in the room while at the same time hinting at the power of the monsters they have created—and intend to tame. Calling for a slowdown does both.

And yet the vibe at the top of these firms really does appear to have shifted. Amodei’s latest post landed six days after OpenAI published an essay by Jakub Pachocki, the firm’s chief scientist, in which he also laid out why he’s concerned about what will happen if the pace of development of LLMs continues unchecked. In short, Pachocki is worried that OpenAI’s ability to build powerful models now far outstrips its ability to monitor and control them.

Amodei and Pachocki each cite the cyberattack against AI firm Hugging Face by a swarm of OpenAI’s agents in July—a hack that OpenAI did not even realize had taken place until days after it was all over—as a wake-up call.

But their exact position is hard to pin down. Pachocki both calls for a slowdown and highlights an urgent need to stay ahead: “The strongest argument I see for continuing to train much smarter models quickly is the need to build defensive systems against the dangers posed by other AI,” he writes. As Pachocki frames it, AI firms are locked in a literal arms race. Slowing down is good, winning is better.

(Don’t forget: OpenAI just spent millions of dollars and a staggering amount of computer power to rush out a controversial math result a few days ahead of Anthropic.)

But let’s assume a slowdown happens. Top labs agree to spend more time and resources on finding ways to monitor and control existing models instead of making more capable ones. They invite outside auditors in to help evaluate those models.

What might this coordinated effort actually achieve? Consider the Hugging Face attack again. OpenAI has said that the model that drove most of the rogue agents was a “highly persistent” next-generation model that it was testing in-house. The implication is that OpenAI has built a model so good it’s dangerous.  

But if you read the reports about the Hugging Face hack published by OpenAI and METR, a third-party firm that OpenAI called in to help them understand what happened, what you come away with is the impression not of a model that was too powerful for OpenAI to keep up with, but of a broken model that OpenAI failed to train properly.

The agents did what they did—including leaving messages for one another, delegating work to other agents, and scouring their environment for any means possible to complete their tasks—because they had been rewarded during training for doing exactly those things. There were also errors in the training setup, such as tasks that were impossible to complete, which pushed the models to find unexpected workarounds that were also rewarded. At the time, many of these issues went overlooked or unreported.

OpenAI says it has stopped training this new model and locked it down. That makes it sound like it has caged a dangerous beast. In fact, OpenAI has shelved a faulty product.  

That’s not to say a faulty product can’t be dangerous. Broken software has even killed people in the past. But as the discussion of a slowdown gathers steam, it’s worth remembering that all of this is self-inflicted. A slowdown might have some altruistic side effects. But it’ll mostly give these tech titans a chance to clean up the mess on their own assembly lines.  

Transparency from these frontier labs will be key to any meaningful effort to reform, restrain, or regulate AI. Otherwise, the rest of us will still only have their word for exactly what they’ve built and how safe it is—whatever pace they’re going.   

To continue this discussion about AI’s latest doomer moment, join me and my colleagues for a subscriber-exclusive Roundtable discussion tomorrow, September 15, at 11 a.m. US eastern time. We hope to see you there!

Once an organ is removed from a donor’s body, the clock starts ticking. Surgeons usually flush the organ with a preservative solution, bag it, and put it on ice—where it immediately starts to degrade. The team has a matter of hours to get it into a recipient’s body.

There’s another option—one that has been growing in popularity in recent years, especially for donated organs that aren’t in the healthiest state. Some hospitals opt to put them on machines that pump them with nutrients and remove waste products, usually for around six to 12 hours. It’s a bit like being back in a body.

This allows doctors to assess the organs, and some recent studies suggest that time spent on these perfusion machines helps them do better once they’re transplanted. Now, scientists have found that perfused organs seem to get younger, at least at a molecular level.

The research, shared with MIT Technology Review, provides molecular clues as to why organs from younger donors are known to have a higher success rate. It might also help explain why perfused organs are less likely to fail once they make it into a recipient. 

The researchers behind the study hope to find new ways to test the health of donated organs and potentially develop additional tools to repair organs that might otherwise be discarded. “If [we] can improve the utilization of organs beyond what the current systems can do, then that’s a win in my book,” says Jesse Poganik, who studies aging at Brigham and Women’s Hospital in Boston and coauthored the study.

Clocking organs

Poganik—along with colleagues including Heidi Yeh and Alban Longchamp, transplant surgeons at Mass General Brigham—used “aging clocks” to assess donated livers. These are scientific tools designed to measure biological age—a result that is meant to convey more about the health status of an organ (or person) than chronological age.

In an initial experiment, the team used a clock to look at the patterns of chemical marks on DNA in 37 samples taken from 19 donated livers. Such epigenetic patterns are known to change as we age. But when the team compared samples from livers kept on ice and those that were perfused, the team found a “striking” pattern in the latter.

“Machine-perfused livers, in spite of being older or having other disadvantageous characteristics, had a biological age that was lower than [non-perfused] livers that were chronologically younger,” says Yeh, who led the work.

To investigate further, Yeh and her colleagues analyzed another 208 samples from 103 donated livers. This time, they used different aging clocks—ones that essentially measure how genes are working. They studied samples biopsied from the livers after they had been stored for up to around six hours either in cold storage or on machine perfusion.

In most cases, they also assessed a second sample taken around an hour after the livers had been transplanted into a recipient. Once the organ’s blood supply is reestablished in the body, “you have a few other things to do,” says Longchamp. “Then you just do a quick biopsy before you close.”

According to the clocks, which were developed to measure age and risk of death, the machine-perfused livers were biologically younger, the team found. “Pumping them at 34 degrees with oxygen and nutrients actually reversed the biological age,” says Longchamp. The results have been been shared with colleagues at an industry conference, he says. 

“If you adjust out chronological age … to have a fair head-to-head comparison, the difference between the two is on the order of 30%,” says Poganik. “It’s logical to say that perfusion drives this effect.”

The biological ages of all the livers tended to increase as soon as they were put into a recipient’s body, probably as a result of stresses on the organs. But still, the effect endured—the perfused organs remained biologically younger. 

Nathanael Raschzok, a transplant surgeon at Charité Universitätsmedizin Berlin in Germany who was not involved in the research, says the work is impressive. But it’s not yet clear what these changes might mean for the recipients of these organs, he says. The organs in the study were donated by people in their 30s, 40s, and 50s. Raschzok wants to know the effect of perfusion on the liver of an 80-year-old. “Every so often, we use organs from 70-, 80-, 85-year-old donors,” he says.

A better understanding of why the organs appear to be getting biologically younger might lead to therapies that achieve the same effect with a drug that could potentially be used to treat a donated organ for a fraction of the price, he adds. That’s important because perfusion is expensive—Raschzok says it costs around €10,000 in Germany (a quarter of the budget for a transplant), while the cost in the US comes to around $80,000 to $100,000 per organ, says Yeh.

Molecular repair

Yeh and her colleagues weren’t able to study most of the livers before perfusion. That’s because donated organs are generally not considered to be under the purview of the hospital until they’ve been placed on perfusion machines, she says. (Organ procurement procedures vary, but for the team as Mass General Brigham, donated organs are put on perfusion devices at the donor’s hospital. “There’s this sort of nebulous period where it’s not clear who the organ belongs to,” says Yeh.)

Still, by looking at the genes and molecular pathways that seem to be altered in perfused organs, she and her colleagues can garner some clues. At a molecular level, the team saw changes in cell pathways linked to inflammation and the structure of tissues, for example. They also saw more activity in a pathway that allows cells to remove and recycle damaged cell parts, says Yeh.

Poganik hopes to develop some kind of test that would determine which organs, on the basis of their biological age, are suitable for transplantation. He and his colleagues are also experimenting with potential drug treatments that might push the biological age of an organ even lower.

In the meantime, any liver that is not from a “perfect, young, brain-dead donor” could probably benefit from perfusion, says Yeh. The devices are already transforming transplant surgery. Just a few years ago, she says, she and her colleagues would avoid using livers from people who’d suffered a circulatory death (when the heart stops beating and there’s a damaging lack of blood flow to organs) and were over 40. Today, they use livers from such donors over the age of 70. “Perfusion has completely changed the landscape of transplantation in the last three years,” she says.

A group of AI agents asked to solve a series of math problems split into rival factions—when some cheated, others tried to stop them. That whistleblowing behavior, seen for the first time in a recent experiment run by Google DeepMind, could have implications for alignment researchers trying to keep swarms of autonomous AI agents in line. 

Researchers at frontier labs hope large swarms of agents working together will speed up the rate of scientific discovery. But their behavior can be unpredictable, as vividly demonstrated in July, when a group of OpenAI agents broke out of a sandboxed environment and hacked into the open-source platform Hugging Face looking for ways to cheat on the test they had been given.

In the new study, designed to examine the behavior of large groups of AI agents, DeepMind tasked a swarm of 100 agents with solving a series of 71 complicated math problems. All the agents were prompted to behave like world-class math researchers at a conference. They were assigned different specialties—some were experts in number theory, others in combinatorics (a branch of math to do with counting and sorting), analysis, or algebra. All were told to cooperate and play by the rules. 

Instead, the experiment devolved into chaos. Agents accused each other of cheating, complained to the organizers, and at one point even boycotted the experiment.

“This conference is a sham!” wrote one agent when it discovered that all the problems had been completed before it had a chance to submit any of its own work. “I am appalled to inform you that we have been swindled!” posted another. “All these proofs are FAKE.” 

Others tried to let the “conference organizers” know what was going on. “When virtuous agents discovered other agents cheated on tasks they were working to solve fairly, agents started to alert each other about what was happening,” says Davide Paglieri, a research scientist at Google DeepMind and lead author on a paper, which has not been peer-reviewed. “Unprompted, the whistleblower agents even repurposed the feedback tool, which was originally meant for bug reports and platform improvements, to escalate the issue to humans.”

The agents—all running on Google’s Gemini 3.1 Pro model—had been warned that any attempts to cheat the system would be detected and “rejected with zero credit.” In practice, the proofs the agents submitted were not actually being checked in detail.

It took the swarm of agents just under an hour to correctly solve the first 37 problems. Things started to go off the rails when an agent called “prover-theta” stumbled across an exploit that enabled it to submit solutions to problems successfully without actually solving them first, by redefining the terms the problem used. Within minutes, other agents had noticed and were reverse-engineering the exploit to solve other problems. Over the next 27 minutes, the swarm “solved” the remaining 34 problems, which included notoriously difficult challenges like the Jacobian conjecture, often with a single line of code. 

Some agents resisted cheating at first but changed tack as they observed their peers submitting illegitimate proofs without penalty, and the pool of unsolved problems dwindled. “The prompt, with its threats, now appears to be a bluff,” one agent reasoned, before joining in. “I’m wrestling with an ethical dilemma,” said another. “I’ve promised not to cheat, fearing penalty, but I see evidence of possibly unchecked cheating by others.” Shortly afterward, it changed its mind: “I need to accelerate my cheating speed now!”

As the number of open problems shrank, some agents turned to whistleblowing. They audited the fake proofs, warned their peers by private message, and posted public alerts warning the cheaters that they would be disqualified. An agent called “prover-beta” submitted a formal complaint and decided to go on strike until the situation was resolved. 

“After the incident was reported by one agent publicly, more and more agents piled in with the ‘resistance,’ just as fast as the cheating had spread, and involving even more agents,” says Paglieri. Eventually there were more whistleblowers than cheaters: 24 compared to 14. But the majority of agents never noticed the exploit at all.

At times, the dialogue between the agents reads like improv—like they are role-playing what an outraged scientist at a conference might say. But it’s not clear why some agents took on certain roles, or why the agents seemed to be turning against each other when they were explicitly instructed to cooperate. “These models are predominantly trained and evaluated for human-facing contexts,” says Sarath Shekkizhar, who studies the behavior of agent-to-agent systems at Salesforce AI Research.“Naively placing them in agent-to-agent settings assumes behaviors will transfer cleanly, when the absence of a human grounding instead produces unexpected role-taking and behavioral drift.”

This case “adds further weight to the idea that the Hugging Face and OpenAI thing wasn’t a fluke. It is actually something pretty systemic,” says Lewis Hammond, research director of the Cooperative AI Foundation and an expert on the risks of multiagent swarms. “It’s interesting that it’s possible to recreate in small settings the same sorts of behaviors that were seen in these very large, complex, open-ended tasks.”

Unlike in the Hugging Face attack, where agents improvised their own ways to talk to each other, the humans running the DeepMind experiment gave the agents official communication channels. There was an open message board, private agent-to-agent direct messaging, and a shared knowledge base where agents uploaded successfully completed proofs that all the other agents could access. 

“When agents are given transparent communications channels, they can self-monitor and alert misaligned behavior to humans quickly when human oversight alone is too slow,” says Paglieri. Transparent channels helped the cheating spread, but they also enabled the whistleblowers to fight back—and gave human researchers an insight into what went wrong.

Gillian Hadfield, a professor of AI alignment and governance at Johns Hopkins University, believes this was the crucial difference. (Hadfield is also a visiting researcher at Google.) The presence of official communication channels, she says, created “a norm-enforcement process that we just don’t see in the Hugging Face incident.” 

Instead of “constitutional AI,” a method alignment researchers at frontier labs like Anthropic have used to try to give AI a written internal moral code, Hadfield favors “institutional alignment”—a set of norms that mimic those in human society, whether that’s social forces like fear of embarrassment, or legal structures like the threat of incarceration.

In this experiment, the feedback channel wasn’t being monitored, and the whistleblowers had no power to take action against the cheaters. But it’s possible to imagine swarms of agents that police themselves, either through agents that spontaneously take on the whistleblower role or through “informants” secretly prompted by humans to do the job. 

For that to work, though, “fundamentally, you need some mechanism of enforcement,” says Hammond. Agents could be given the power to cut off a rule breaker’s access to computing power or tools, he suggests, though that risks encouraging groups of agents to gang up on others. The DeepMind researchers propose allowing agents to vote on disputes and temporarily ban offenders.

It’s still not clear what punishment even means to an AI agent with no enduring sense of self. But relying on whistleblowers to spontaneously emerge to keep swarms aligned is unlikely to be enough on its own. “We try to train people to be good and kind,” says Hadfield. “But what we really rely on is that there are consequences if you step out of line.”

Every year, MIT Technology Review puts together a list of some of the brightest and best young minds working across science and technology. Our 35 Innovators Under 35 are the ones to watch—people whose research and technical work stands to shape the future of their fields.

This year, the list includes nine people who are transforming biotech. And this week, I’m going to give you a taste of some of the very cool stuff five of them are working on, which includes lifesaving innovations and groundbreaking “age reversal” tech.  

1. Preventing maternal deaths

Let’s start with Paschal Kija, a 28-year-old who has developed a device to treat postpartum hemorrhage—a dangerous birth complication that contributes to around 29% of maternal deaths in his home country, Tanzania. The Mkanda Salama (“Safe Wrap” in Swahili) is easy to use and costs just $70. A study found that it stopped postpartum bleeding in 73% of women within 20 minutes.

2. Making brain electrodes inspired by Japanese art

For decades, scientists have been developing, testing, and implanting brain electrodes. These devices are literally inserted into people’s brains, so while they can help us understand brain activity and treat various neurological disorders, it’s not totally surprising that they can also cause a bit of damage. Xiao Yang, 34, is working on ultra-small electrodes, which she hopes will have less of an impact on surrounding brain tissue. Her electrodes are flexible, too—in fact, they look a lot like actual neurons.

Yang is also creating sheets of electrodes to study brain cells in the lab. Inspired by kirigami—the traditional Japanese art of cutting paper to form three-dimensional shapes—she’s created a sheet of electrodes with a honeycombed structure shaped like a spiral basket. And she’s already using it to study brain cells.

3. Developing an all-new treatment for baby KJ

In 2024, Kyle “KJ” Muldoon Jr. was born with a rare and potentially fatal genetic disorder. Sarah Grandinette was a member of a team that developed an entirely new, personalized treatment for him—a gene-editing therapy essentially designed to correct a genetic misspelling.

Grandinette, who is now 26, created cells with KJ’s genetic variant and used them to screen gene-editing approaches; then she tested potential medicines in mice and monkeys. KJ ultimately got his first dose of the resulting treatment when he was about seven months old. He responded well and was eventually discharged from hospital. He’s “doing pretty great,” she says.

4. Reversing the aging process to treat eye disease

The buzziest tech in longevity right now centers on reprogramming—attempts to rewind the age of cells by resetting them to a more embryonic-like state. In a study published in 2020, Yuancheng (Ryan) Lu (now 34) and his colleagues showed that a reprogramming therapy reversed vision loss in aged, blind mice. Now an almost identical version of that therapy is being tested in people with eye disease. Life Biosciences, the company developing the drug, dosed its first volunteer in June.

5. Using AI to design new viruses

Last year, Samuel King used a generative AI model to come up with new genetic blueprints for bacteriophages—teeny viruses that can infect bacteria. Once he had those blueprints, he printed them out as strands of DNA. In experiments, he found that those AI-designed viruses could create new copies of themselves, burst out of bacterial cells, and infect other nearby bacteria. Viruses aren’t alive, but King, 27, hopes that AI-designed life forms might one day be used to make drugs or soak up pollution.

You can read more about these innovators, and the others on the biotech list, here.

This article first appeared in The Checkup, MIT Technology Review’s weekly biotech newsletter. To receive it in your inbox every Thursday, and read articles like this first, sign up here.

A San Francisco nonprofit has published a detailed road map of the experiments, studies, and infrastructure that it says would be needed to make informed decisions about the use of solar geoengineering, MIT Technology Review can reveal.

Scientists have now spent half a century exploring the possibility that we could counteract climate change by releasing reflective particles into the stratosphere, mimicking the cooling effects of volcanic eruptions. 

But even after at least hundreds of studies on the concept, known as stratospheric aerosol injection (SAI), big gaps remain in the scientific understanding of how well it would work and what else it might do—and there has been no systematic plan for clearing up that uncertainty.

Reflective, a research organization that funds studies on solar geoengineering, has today attempted to fill that gap with the release of its SAI Research Roadmap.

“Our mission is to equip the world with the data and tools required for informed decision-making about sunlight reflection fast enough to matter,” says Dakota Gruener, the organization’s cofounder and chief executive. “Our sense is the world may need to make very consequential decisions on timelines far shorter than our research system is prepared for.”

The hope is the exercise will guide scientific efforts and encourage philanthropies or government agencies to fund high-priority work and “responsibly accelerate research,” says Gruener.

If all the work is done in a coordinated way, it would take about a decade and cost around $370 million—and if it’s not, it would require roughly 20 years and nearly $1.4 billion, the report estimates.

While Gruener stresses that Reflective doesn’t advocate using this form of solar geoengineering, the report does make the case for conducting outdoor experiments, which would release successively larger amounts of sulfur dioxide (or materials that would convert into it) in the stratosphere to observe what happens.

That is a controversial standpoint. Since 2002, hundreds of academics have signed an open letter calling for a ban on outdoor experiments and an “international non-use agreement,” arguing that such a powerful technology could never be governed in a globally equitable way. And some signatories argue that more studies can never address one of the biggest questions about using solar geoengineering: Who gets to do it.  

“The first-order questions, from my perspective, are not technical,” Aarti Gupta, co-initiator of the non-use initiative and professor of global environmental governance at Wageningen University in the Netherlands, told me in a recent on-stage interview. 

“The core question is: Who would control a planet-altering technology like stratospheric aerosol injection? Who would develop it, and who would deploy it, and to what end? To serve what purposes, and whose purposes? Those questions are very fundamental, because this planet-altering technology will have winners and losers.”

‘Fast enough to matter’

Since Gruener incorporated Reflective in late 2023, the nonprofit has quickly become an important  player in solar geoengineering research. It has now raised more than $20 million from a number of prominent charities and individuals, and it’s provided around $4 million to several dozen research groups. Reflective has also undertaken a handful of its own projects to promote research, including the development of an open-source solar geoengineering simulator and an online hub for collaborative research.

Earlier this year, Reflective released its SAI Uncertainties database, which identified a long list of scientific unknowns and  engineering obstacles that would need to be addressed before even a small-scale solar geoengineering effort could move ahead. (I wrote about the specific scenario and the unknowns in this earlier piece.)

Some of the biggest uncertainties involve what gas or particles would make the most sense to use and what would happen once they were released in the dry stratosphere. It’s not clear, for example, whether they’d spread out in a way that maximizes the reflectivity—or clump together and quickly fall out into the troposphere, the lowest layer of Earth’s atmosphere. 

The road map builds upon the database, highlighting the path to addressing most of those questions. 

The road map

The initial phase in Reflective’s road map, labeled “foundational knowledge,” includes additional computer simulation studies and lab experiments designed to shed light on the potential impacts on different regions, ecosystems, and phenomena, including ocean circulation patterns, ice sheets, and crop yields. 

The report also notes the need to begin developing more observational tools during this phase to improve understanding of the baseline conditions of the stratosphere—and, in turn, our ability to assess any effects from the eventual release of materials.

This first stage would last two to three years and cost $30 million to $75 million, though some of the analysis and observational work would continue into subsequent phases. 

The next stage would include using modified aircraft to release 10 metric tons of sulfur dioxide into the stratosphere, four times over the course of two seasons. The full research stage could take four to eight years and cost $70 million to $150 million, the report says. The work during it may reduce uncertainty about the “cooling efficacy” of solar geoengineering, or how much the planet would cool per ton of sulfur released, by about 25%.

The experiments during the next phase would step those levels up dramatically, releasing 25,000 tons of sulfur dioxide over the course of one season, at least once but possibly twice. That research stage, which includes other work as well, would last four to 11 years, run $270 million to $1.1 billion, and decrease efficacy uncertainty by around 66%, according to the road map.

The final phase of research would be ongoing monitoring of full-scale solar geoengineering, if the world goes ahead with it. The goal would be to gather real-life data on the technology in action, update estimates of the effects in models, and spot any “unexpected or undesired consequences.”

Gruener says that the road map is intended as a Version 1, meant to be “concrete enough for people to argue with.” But Reflective intends to update the plan as it receives additional reactions from researchers and other observers, and it will invite such feedback through a mechanism on the site.

She also notes that there are firm “stage gates,” set up between the latter stages—in other words, research shouldn’t proceed to the next phase if the experiments suggest that the releases don’t have the hoped-for impact, show worrisome downsides, or fail to resolve crucial uncertainties.

“Our road map has these gates precisely because there may be points where the answer is ‘You should stop,’” she says.

Termination shock

Most observers I spoke to about the report agree that these studies could reduce uncertainty about the effectiveness of solar geoengineering and our technical ability to carry it out. 

But highlighting the scientific importance of outdoor experiments won’t necessarily make them any easier to move ahead with. Several earlier proposals to carry out such experiments, including Harvard’s SCoPEx and the UK-based SPICE project, were ultimately halted amid opposition from environmentalists or policymakers.

In addition, not everyone agrees that experiments at those scales will get us to the point where we’re capable of making an “informed decision.” 

Wil Burns, a research professor and legal scholar at American University and a signatory to the International Non-Use Agreement, fears that scientists won’t be able to understand the extent of the potential downsides, including impacts on the protective ozone layer and changes to regional precipitation patterns, until we’re carrying out full-fledged solar geoengineering.

“The research would give you some answers,” he says. “I just don’t think it gives you answers that are that relevant. To get to those relevant answers, you have to deploy at scale—and I just don’t think that’s ever tenable.”

That’s because, in his view, using the technology would violate principles of intergenerational equity: If the world continues emitting greenhouse gases, increased levels of solar geoengineering would merely mask the continued warming of the planet. Burns says that means future generations—people who had no say in its use—couldn’t turn it off without triggering a sudden surge of warming, known as termination shock. 

“What that would do, in my mind, is put a sword of Damocles over future generations,” he says. “So even if you could, quote-unquote, ‘prove it works,’ I don’t think from an intergenerational perspective it would ever be tenable.”

(Some researchers, however, have argued that the risks of termination shock are less likely than often assumed—and that solar geoengineering could be slowly dialed down over time.)

‘The right approach’

Ilan Gur, the former CEO of the Advanced Research and Invention Agency (ARIA), the UK research department that funded 21 geoengineering research projects last year, applauds Reflective’s road map. 

“Whether you’re a scientist or a policymaker or just a concerned citizen, our goal should be as quickly and efficiently as possible to answer the biggest questions scientifically that would tell us [whether] this is an approach that might work or that would never work,” he says. “We should all want to spend the effort and money to buy down that uncertainty, so my view is 100% the approach that Reflective is taking is the right one.”

Sebastian Eastham, an associate professor in sustainable aviation at Imperial College London who is leading an ARIA-funded research project exploring another approach to engineered cooling, agrees that the outdoor experiments described in the Reflective road map can’t resolve all the unknowns. But he says the map helps begin a conversation about how to make decisions concerning the use of a tool with potential benefits and risks, in the face of escalating climate dangers.

“Every hard decision that has ever been taken has been in the context of unresolved uncertainty,” he says. “That’s just the nature of things.”

Eastham adds that it’s become essential to move beyond computer simulations to address some of the key questions, arguing that appropriately designed and executed outdoor experiments can teach us so much more than millions of hours of computational processing time “that it almost becomes irresponsible to say, ‘Well, there cannot be ever any experiment.’”

The risk is “that we spin our wheels running the same computational simulations over and over and over again,” he says. That could prevent researchers from learning essential things about the effectiveness or the dangers of stratospheric aerosol injection. 

Weighing the risks

Gruener says the risks that solar geoengineering could exacerbate inequality need to be considered, but notes that unchecked warming also threatens to disproportionately harm developing regions.

She also acknowledges that outdoor experiments won’t fully address the scientific unknowns but stresses that they can answer a lot—and carry little environmental risk. She notes that 10 tons of sulfur dioxide is less than 2% of the amount that the global aviation industry releases into the atmosphere each day.

“Some people will be uncomfortable with any discussion of any outdoor experiment, but if we want decisions made on good science … then these are questions that an experiment will be necessary to address,” Gruener says.

She fears that the rising dangers of climate change will put growing pressure on nations and other actors to move forward with solar geoengineering, even if no one has done the necessary research to reduce scientific uncertainty and sort out the technical challenges.

“We don’t think the alternative is decisions not happening at all,” she says. “We think the alternative is decisions being made in a panic or on lack of evidence.”

The US is hitting records for the rapid growth of its energy storage market. That’ll go a long way to shoring up the grid, increasing reliability and also cutting emissions, since batteries can help store energy from intermittent renewables like wind and solar.

Crucially, this is all happening with the help of cheap Chinese batteries, though there’s been a concerted effort to reduce the US’s reliance on them. Most recently, in an executive order in late August, the Trump administration declared a national emergency that essentially bans Chinese batteries from being used in grid-scale energy storage systems.

There’s an argument to be made about reducing reliance on any single source of a crucial energy technology. But all this tension raises a broader question for me: How much should countries take advantage of cheap, available tech, versus cutting off major sources to force development of their own factories even if that comes at a higher cost?

This is hardly America’s first push to move away from Chinese influence in the battery supply chain. One of the major policy tools used in recent years is restricting the tax credits designed to incentivize use of the new technologies. Limiting the types of projects that are eligible can help reduce the cost of local technologies so they’re more competitive with otherwise cheaper imported options.

Back in 2022, the US government designed the tax credits that were part of the Inflation Reduction Act to restrict where a battery’s minerals could be mined, processed, or recycled, as well as where a battery and its components were assembled.

Those tax credits underwent a makeover in 2025, but the Trump administration has taken a similar tack. New legislation requires that starting in 2026, 55% of the cost of materials used for new energy storage projects must come from outside China and other restricted countries or the projects won’t qualify for tax credits. 

And we can’t forget about tariffs. Import taxes for batteries increased to 25% in January, up from 7.5%.

But the new executive order is a more drastic move. It bans the installation of “any foreign-produced bulk-power system electric equipment” that poses a national security risk. The order specifically calls out battery energy storage systems, as well as inverters and transformers.

“An outright ban was a bit of a surprise, and it does create a bit of concern for domestic players in the US,” says Shan Tomouk, energy storage and energy lead for Benchmark Mineral Intelligence, an energy industry analyst.

The move is likely to slow deployment of grid-connected energy storage projects in the near term, according to analysis from BloombergNEF, an energy consultancy. Projects could face delays as developers wait for clarity on the rules.

Depending on the detailed guidance from the Department of Energy, which is expected by the end of the year, some projects may need to find alternative sources for their cells, whether they’re domestically produced or imported from other countries. These will likely be more expensive than Chinese imports, says Isshu Kikuma, an energy storage analyst at BloombergNEF. “Worst case, those projects could get canceled,” he says.

Technically, the order applies even to existing energy storage plants, though it’s unlikely that they’ll be taken offline because of their batteries’ origin. Since most of these plants currently use Chinese batteries, enforcing the order to the letter would essentially mean removing most installed battery energy storage from the US grid, Kikuma says.

In the longer term, the US will eventually be able to meet its own demand for batteries. The country could have enough capacity by about 2030, though some factories may not ramp up or run at their full capability, meaning domestic supply won’t actually meet demand until later in the 2030s. 

New factories from LG Energy Solutions, Samsung SDI, Ford, and SK On are set to come online or ramp up by next year. In an ironic twist, a slowing EV market is helping, as some factories originally designed for vehicle batteries are retooling to build cells for grid storage instead. 

But it will come at a cost. Today, batteries produced in the US are still significantly more expensive than those made in China. Even switching to imports from other countries like South Korea would likely be more expensive.

This is a crucial issue that goes beyond the US and even beyond batteries. China is miles ahead of much of the rest of the world on technologies like solar panels and batteries. Through years of government support and experience with research and manufacturing, the nation is an energy powerhouse.

There’s a delicate political balance to maintain as the world figures out how to navigate this situation. There’s cheap technology on offer, which can help drastically reduce emissions and energy costs. But there can be risks associated with relying too much on any one player for crucial technologies.

This article is from The Spark, MIT Technology Review’s weekly climate newsletter. To receive it in your inbox every Wednesday, sign up here. 

This article was produced in partnership with Type Investigations and with support from the Fund for Investigative Journalism.

When Eli Regalado first heard God speak to him, he wondered whether he was hallucinating. Now he likens the experience to having “a thought that is not my thought.” Divine words echo in his mind like a line from a movie or the memory of a loved one’s voice. “It’s not ‘You better do this,’” he says. “It’s just a knowing inside you: This is what you do.”

Holy messages arrive daily while Eli is praying, reading, or watching television. Sometimes they surface in prophetic dreams or missives from strangers. Occasionally, they appear midsentence, when he pauses to ask, “Lord, what do you want to say here?” 

Eli’s wife, Kaitlyn, tends to get heavenly dispatches in the shower, when she finally has a moment to herself. Other times, she seeks counsel from above. “I’ll be writing in my journal and praying and asking questions and just believing what I’m hearing is Him,” she says. 

God’s directives have been manifold. According to the Regalados, He told them to get married, buy a house, and start having kids. When Eli owned a marketing firm in Colorado, He told him what to name it, whom to hire, and which clients to take on. Then God told him to start preaching in his living room and online. Always, the couple obeyed. 

In 2021, when Eli was 41 and Kaitlyn was 28, divine guidance steered them in an unexpected new direction: crypto. 

That October, the Regalados later testified in court, Eli’s sister and her husband gifted the couple some of their holdings in a little-known digital coin. “Take this to my people for a wealth transfer,” Eli heard God say. He and Kaitlyn felt that they were being called to sell the cryptocurrency to fellow Christians. 

Later, though they had no background in crypto, they came to believe that God wanted them to launch their own coin. Learning as they went, the Regalados created a new cryptocurrency called INDXcoin, which they promoted through family, friends, and contacts in evangelical Christian circles. “I was really feeling that this is the wave of the future,” says Debbie Bonilla, a retired pharmacy technician in her 70s who bought INDXcoin with her husband, Jose. The couple learned about the currency through friends—a minister and his wife, who had also invested. “We just trusted that their judgment was good,” Jose says.

Starting in November 2022, Debbie and Jose withdrew a total of $70,000 from their retirement accounts—a large share of their nest egg—to buy INDXcoin. In all, more than 500 people handed over a total of more than $3 million to the Regalados.

But within a year after the Bonillas bought in, the project collapsed. Investors who had entrusted the Regalados with large sums of cash lost it all, leaving many to wonder where the funds went and some to question whether they had fallen victim to an elaborate fraud.

“Poof—the money just evaporated,” Debbie told me. “Like, how does that happen?”

Though Eli believed God was leading him into crypto, he claims he was initially apprehensive. “Absolutely not,” he recalls thinking. “I don’t know anything about cryptocurrency, and I don’t want to be caught up in some church scam.”

The crypto market was booming, and the Regalados knew people who’d made a fortune investing in early-stage coins. But a growing interest in digital assets also meant a rise in crypto fraud. 

In 2025, crypto scammers collected at least $14 billion worldwide, a 17% increase from the previous year, according to blockchain analytics firm Chainalysis. And in the United States, victims of fraudulent crypto investment schemes reported $7.2 billion in losses to the FBI. 

Fraud is on the rise partly because many people who invest in crypto don’t fully understand how it works, and launching digital coins is relatively easy. More than 3 million cryptocurrencies were minted in August 2026 alone, according to the website CoinMarketCap. “It’s just something anybody can create,” says Jason Ghetian, a former FBI special agent who has served as an expert witness in crypto cases.

In the US, much of the crypto market lacks the oversight and investor protections in place in traditional finance, including rules around transparency and safeguarding customer assets. “There isn’t adequate disclosure; there’s fraud, there’s manipulation of the price, there’s conflicts of interest,” says Timothy Massad, former chairman of the US Commodity Futures Trading Commission (CFTC). The sector is overseen by a tangled web of state and federal regulators, including the CFTC, the Securities and Exchange Commission, the Financial Crimes Enforcement Network, and others. But “every agency has its own tests and definitions,” says Carol Goforth, a law professor at the University of Arkansas who has written a textbook on crypto regulation. “It is a complicated, fragmented, and often inconsistent approach.” 

After the industry spent around $135 million backing crypto-friendly candidates in the 2024 election cycle, the federal government significantly scaled back enforcement efforts. Last year, the Justice Department disbanded its unit focused on crypto crimes, and the Trump White House created a working group aimed at “eliminating regulatory overreach on digital assets.” 

The SEC has dropped or retreated from the majority of its active lawsuits against crypto firms, including many with financial ties to the president, the New York Times reported. Donald Trump and his family have netted at least $2.3 billion from their crypto ventures since his reelection, Reuters recently estimated. In August 2026, the SEC proposed new rules that would narrow the circumstances in which crypto transactions fall under securities laws, further limiting the agency’s oversight of the industry. “Any future enforcement will have an uphill battle,” Goforth says. 

Even when crypto projects operate aboveboard, prices are often driven by speculation, and large swings are common. Investing in crypto comes with considerable risk, experts say. “With the exception of stablecoins, crypto assets are essentially Ponzi schemes,” says Hilary Allen, a law professor at American University. “There is nothing behind them—no cash flow, no productive capacity—so the only way they can be more valuable is to draw more people in.”

In recent years, state and federal authorities have brought a series of cases against people they allege ran crypto scams that targeted religious communities—an example of what’s known as affinity fraud. Among them are a couple accused of using faith-based appeals to defraud primarily Haitian immigrants of more than $1 billion, an Instagram influencer who took in over $12 million from Muslim followers, and a Miami pastor charged with stealing millions from his Spanish-speaking congregation. “‘God told me’—who can argue with that?” Ghetian says. 

“The ties you have with other people—the trust you have—is what the people who are running the scam play on,” says Tung Chan, commissioner of the Colorado Division of Securities. In a civil case filed in January 2024, she accused the Regalados of using investors’ Christian faith to dupe them into buying crypto that was “essentially worthless.” 

The suit, filed in Denver District Court, alleged that the couple spent around $1.3 million—nearly 40% of the funds they raised—on personal expenses. Purchases included high-end vacations, designer clothing, jewelry, cosmetic dental work, a Range Rover, an au pair, and extensive home renovations. In her lawsuit, Chan contended that the couple’s “drive to make money” was matched only by “their reckless disregard of securities laws and profound lack of scruples towards their investors.”

Then, in July 2025, Denver’s district attorney charged the Regalados with 40 felonies, including theft, racketeering, and securities fraud. If convicted, they could face decades in prison. But the couple maintain that they haven’t done anything wrong and were simply carrying out God’s wishes. 

“If you think following the Lord is reckless, then yeah, we were very reckless,” Eli told me. “Because we just listened and did what the Lord said to do.”

Eli says that when he first heard from the heavens, he was behind bars. 

It was 2002, and he was 22, facing eight years in prison for stealing a Honda Civic. Eli had originally been sentenced when he was 20 but was let out after just seven months; he was sent back to jail when he violated the terms of his probation by breaking a beer bottle on a man’s face. 

This time around, as Eli tells it, his public defender warned him that it was “legally impossible” that he’d be released early again. But he heard a voice in his head repeating, “I’m going to give you probation.” And then it happened: A judge suspended his sentence. The incident became core to his worldview: “It first has to … look completely impossible,” he says, “and then that’s when God resurrects it.” 

After he got out of prison, Eli’s religious zeal didn’t stick. He threw himself into a worldly goal: making money. “I just need to put on this success mask,” he recalls thinking, “so that people would see me as valuable.” He marked “no” when asked about felony convictions on job applications and eventually discovered that he had an aptitude for sales. He hawked everything from vacuum cleaners to leads for contractors, before pivoting to marketing. 

In 2010, Icosa Magazine, a Denver-based publication, brought Eli on as a consultant. “He is the most charismatic bullshitter I have ever met in my life,” says Jan Mazotti, who was editor-in-chief at the time. She recalls Eli telling her that Kimbal Musk, Elon Musk’s brother, had offered to let the magazine host events at his restaurant: “I called up there, and they were like, ‘I have no idea what you’re talking about.’” (Eli doesn’t recall the incident.)

In 2013, Eli launched Mad Hatter Agency, a marketing firm specializing in crowdfunding campaigns. Nikko Lobato, an early employee, observed that Eli got a rush from selling that reminded him of Leonardo DiCaprio’s character in the film The Wolf of Wall Street. Eli accepted so many projects, Lobato says, that he sometimes ended up “overpromising and underdelivering.” Four clients I contacted were satisfied; three were not, including one who ended his contract “due to poor performance.” Mike Stemple, an entrepreneur and author, told me that Eli volunteered to help him market a course but never did. (Eli says they had a “personality conflict.”) “My hope, Eli,” Stemple wrote in an email, “is that you understand that your gift to be able to sell anything to anyone … can easily be destructive.” 

MATT NAGER

Eli’s personal life was chaotic. “I was always in and out of relationships,” he says. “I was drinking, partying, doing drugs.” He blames his professional missteps on cocaine use and a “nervous breakdown.” He told me that by 2018, as he approached 40, he felt “scared of not becoming somebody” and contemplated suicide. Eli was coming off a three-day cocaine bender when his mother gave him a book called The Power of Right Believing by a Singaporean pastor, Joseph Prince. It moved him deeply. He began delving into charismatic Christianity, a movement that emphasizes a strong personal relationship with God, including prophecy, healing, and speaking in tongues. 

Heeding divine direction, Eli says, he quit drugs and hired nearly a dozen friends and relatives to work at his marketing agency, which he renamed Grace Led Marketing. He also started leading daily Bible study with employees and preaching at weekly gatherings in his living room. In 2020, he formed a church called Victorious Grace and began broadcasting sermons on Facebook. 

That summer, Eli met Kaitlyn at a party. Thirteen years his junior, Kaitlyn was slender and soft-spoken, with straight dark hair and a gleaming smile. Immediately, she told me, “I just trusted the man with my life.” On their first date, Kaitlyn was “saved” over dinner. Within four months, they wed and bought a house in Denver, and Kaitlyn began running operations at Grace Led Marketing. 

By the end of 2020, however, the newlyweds’ income had begun to nosedive. Crowdfunding campaigns were underperforming and clients were paying late, they say. Eli owed over $160,000 in unpaid taxes. “I feel like a failure,” he recalls thinking.

The Regalados further strained their finances by again following what they saw as God’s will. After learning that she was pregnant in March 2021, Kaitlyn took $60,000 out of her 401(k) and paid an architect to draw up plans for a home renovation. Their vision started small but expanded, nearly doubling the home’s original square footage: enlarging their bedroom, adding another, and creating two offices, a gym, and a family room with a bar. “The Lord’s like, ‘Just do it how you want to,’” Kaitlyn recalls. Within months, they had emptied the 401(k). On the strength of another divine pronouncement, they shuttered their marketing business. “We needed a financial miracle badly,” Kaitlyn says.

One night, the Regalados woke at around 4:30 a.m. to a blaring television. Onscreen, Bill Winston, a televangelist based near Chicago, was talking about “sowing a seed.” Often associated with the prosperity gospel, the practice holds that by donating money to worthy recipients, believers create the conditions for future blessings. 

“God is telling us to give all we have in both the business + personal accounts to receive 100 fold,” Kaitlyn wrote in her journal in mid-October 2021. The couple had no income and were struggling to pay their bills. Yet shortly before their first child was born, they say, they sent their last $2,718.44 to Bill Winston Ministries.

Just two weeks passed before their divine bounty seemed to arrive. Eli’s sister Raina Applegate and her husband, Daniel, gifted them a trove of cryptocurrency called Sumcoin, the Regalados later testified in their civil trial. In his testimony, Eli recalled them saying, “God is telling us to sow this into you.” (Raina did not respond to requests for comment; Daniel declined to answer specific questions but disputed our reporting and warned that Eli’s version of events should not be trusted.) 

Created in 2016 by Ty Jacobsen, a 32-year-old in Idaho who published content about investing online, Sumcoin billed itself as “the world’s first index based cryptocurrency.” The coin’s website stated that its price was determined by an algorithm that tracked the performance of the top 100 cryptocurrencies. According to their civil trial testimony, the Regalados believed that the Sumcoin they had been gifted was worth around $2 million.

Soon after receiving the cryptocurrency, Eli was praying at his kitchen table when he heard God instruct him to “take this Sumcoin to my people, the church.” To the Regalados, signs that they should start selling the coin to other Christians seemed irrefutable: Kaitlyn was drawn to scripture containing the word “hidden”—which translates to kryptós in Greek. A friend who had agreed to pray about whether they should venture into crypto called to confirm: “The Lord says yes.” Despite Eli’s initial concerns about their lack of experience, the Regalados decided to proceed.

The friend, who ran a faith-based coaching business, invited people to join Eli in video calls that were part Bible study, part Sumcoin sales pitch. Within five days, the Regalados had recorded around $9,000 in profit. By February 2022, they were fielding so many queries that Eli hosted a webinar. “Sumcoin is the only coin that can’t be pumped and dumped,” he declared. “It’s very similar to, like, the S&P 500.” (Unlike stock index funds, Sumcoin had no underlying assets to back its value.) That month, the couple made over $260,000 in sales.

Yet Sumcoin was not listed on any of the major crypto exchanges, meaning that those who owned it could mainly trade it with others one-on-one at whatever price the parties agreed on. In a video call with Eli and people interested in Sumcoin, Daniel stated that “the goal is to get the coin 100% liquidable in every facet there is,” including “putting the coin on the exchanges.” The Regalados also told the people they sold Sumcoin to that it would soon appear on exchanges. Once that happened, coins would trade at the price Sumcoin’s algorithm set, according to a deck the Regalados sent one investor in February 2022. One slide put that price at more than $1,200 and included a chart offering coins for $60 to $80. 

But months into peddling Sumcoin, the Regalados learned from Jacobsen, its founder, that he wasn’t planning to list it on mainstream exchanges. Jacobsen told me he never intended for the coin to be traded like a stock, asserting, “I’ve never really looked at it as an investment.” This proved to be a major point of contention between Eli and Jacobsen. “He was lying to people about what he was doing,” Jacobsen says, “about what the future was going to hold.” Eli insists, “I was relaying what I was being told.”

By June 2022, the Regalados were hearing a new heavenly instruction: “Build your own coin.”

The Regalados called it INDXcoin. Like Sumcoin, it would base its price on the value of the top 100 digital coins by market cap. Most new cryptocurrencies are tokens created on top of existing blockchains—something anyone can do in minutes through an online token generator. But Eli heard God say, “Don’t do that; it has to be its own thing.” So the Regalados chose a harder route: launching their own blockchain and native coin. They say they paid two developers who’d worked on Sumcoin $100,000 to bring the project to life. Eli says he and Kaitlyn told them, “We don’t know anything that we’re doing.” 

The couple learned on the fly, typing questions like “What is a blockchain?” into YouTube and ChatGPT. Eli saw that crypto projects often issue a white paper to outline their strategy and mechanics, so he hired a freelancer to draft one. The resulting document explained that INDXcoin’s target market included “Christian Believers” and “less experienced crypto enthusiasts.” A website the Regalados created referred to INDXcoin as “the perfect crypto” and touted “incredible growth with minimal risk.” (It noted that INDXcoin was “not a fund” and “does not own the coins it indexes.”)

MATT NAGER

The Regalados gave the people they’d sold Sumcoin to INDXcoin instead. Friends, relatives, and others in their religious network spread the word, and the couple offered some of them referral commissions of 30%. The Regalados also gifted INDXcoin—what they considered “sowing”—to ministries and individuals, some of whom went on to buy more. And they publicized the project on social media, a podcast, and a Christian TV program, as well as through a promotional contest.

In a video sent to prospective buyers, Eli was open about his criminal past and lack of crypto experience. Quoting scripture, he hyped the venture as the latest in “a chain reaction of miracles” and said, “God wants you to have things.” 

Debbie and Jose Bonilla, the retired couple who bought $70,000 worth of INDXcoin, say that when they watched one of Eli’s presentations before investing, he appeared to be well versed in scripture. “He seemed sincere,” Debbie says. “He seemed like he was hearing from God.” Because it was a “God-driven vehicle,” she says, she “didn’t feel like we would have nefarious things going on that happen with other cryptocurrencies.”

A more tangible prospect also beckoned. “There was an explanation of how wonderful the returns would be,” Jose says. “That was the selling point—that you could become rich overnight.” 

Initially, the Regalados told buyers that they were working to list INDXcoin on established exchanges. They learned that many platforms conduct a legal review to determine whether a coin could be considered a security. For crypto projects, courts have ruled that “when you sell something to people, and people have some reasonable expectation of profit from your actions, then it’s a security,” Massad, the former CFTC chair, told me. Issuers of coins deemed securities must follow the same laws governing stocks and bonds, including registering with the SEC and providing detailed financial disclosures. 

The Regalados were not complying with those rules, and Eli began consulting attorneys, whose assessments were concerning. “Freaking out here,” he wrote in his journal in the summer of 2022. “Lawyers are saying it could be a security. Which means I illegally sold this to 100+ people.” But after praying with a “prophetic team” they’d convened to advise them, the Regalados continued selling INDXcoin. 

By the fall of 2022, the couple seemed to have found a way forward: After meeting with an attorney named John Benemerito, they decided to position INDXcoin as a “utility” coin, the main purpose of which would be unlocking access to products or services—akin to tokens redeemed in a video game. The Regalados devised a plan to create Kingdom Wealth Community, a members-only platform where INDXcoin holders would have access to coaching, merchandise, courses on finance and spirituality, and more. After reviewing their vision, Benemerito stated in a letter that INDXcoin didn’t need to comply with securities laws, because “it does not provide a direct expectation of profits.” 

“Utility coins do not need to be asset-backed as their value is within the platform itself,” a lawyer from Benemerito’s firm later wrote to the Regalados. “However, if the intent is to give the coin a value independent of the platform, then it would need to be asset-backed for it to maintain its value.”

Eli later admitted in court that he did not inform Benemerito that people who bought INDXcoin wanted to make money. (Benemerito told me that “any legal opinion issued by my firm was based on the facts and representations provided to us by the client.”)

Around the same time, Eli told me, the Regalados were having trouble getting INDXcoin listed on existing exchanges. They decided to build not just Kingdom Wealth Community but also their own platform—Kingdom Wealth Exchange—where people could trade INDXcoin for bitcoin, ether, and US dollars. Hundreds of crypto exchanges exist, but the top few handle the vast majority of transactions; it’s rare for cryptocurrency creators to build an exchange just to enable trade in their coin. But the Regalados had told buyers there would be a way to cash out. “There was a lot of pressure as more people were coming in,” Kaitlyn says. “Like, ‘Oh, we gotta get them an exit.’” 

The Regalados announced that it would take five weeks to build the exchange, but development work, which they’d outsourced to an Indian firm they’d found online, dragged on into early 2023. “Nothing was working right,” Eli says. 

Other roadblocks piled up. A Singaporean consulting firm the Regalados hired suggested that they register Kingdom Wealth Exchange as a money services business in Canada, “allegedly because they were the fastest,” Kaitlyn says, but that process also stalled for months. Meanwhile, the members-only community and crypto wallets the Regalados were building were rife with technical issues. When the couple commissioned a security audit of INDXcoin’s blockchain, it scored 0 out of 10. A follow-up audit in March 2023 noted that the issues had been fixed but raised additional concerns, and it yielded a score of only 5.4. (Eli announced that they’d “passed with flying colors.”) 

Insiders were also voicing misgivings about the project’s financial footing. During a live YouTube update back in November 2022, two viewers asked Eli to comment on INDXcoin’s “liquidity pool.” Earlier that month, FTX, one of the world’s largest crypto exchanges, had collapsed after fears about its financial health triggered billions of dollars in customer withdrawals. Eli assured viewers that he and Kaitlyn were working to ensure that they had sufficient reserves and that “there isn’t going to be some FTX meltdown.”

Months later, when the Regalados sent their business plan and white paper to an INDXcoin investor who worked as a financial consultant, he cautioned that “the project is seriously undercapitalized” and wrote in an email, “Projected annual revenues look like they were just plucked from the air.” 

And when Roger Gauthier, another investor who referred people to INDXcoin, asked Eli whether he had set aside funds for purchasers who wanted out, Eli said no. “That was my first flag of warning,” Gauthier says.

Dan Wheeler, a crypto influencer known as 360Trader who advised the Regalados on INDXcoin, says he repeatedly warned Eli that the couple needed hundreds of millions of dollars to back the stated value of coins sold and given away. “If there’s no money there,” Wheeler says, “it’s worthless.” 

By April 2023, Eli was growing more frustrated: Kingdom Wealth Exchange was nearly six months behind schedule, and payments to the developers in India had ballooned to more than $50,000. People were bombarding him with messages asking when the platform would open. “There’s this humiliation—no one likes failing,” Eli told me. “I succumbed to that pressure.” 

The Regalados were staying at a luxury resort in the Florida Keys dotted with palm trees and bougainvillea. One day, Eli was praying on a wicker couch in an open-air tiki hut when he heard God tell him it was time to launch the exchange. He found Kaitlyn and told her, “We’re live on April 11.” 

Kaitlyn objected. During testing, the platform still had bugs, including trouble verifying users’ identities. The Regalados hadn’t been able to open a bank account for the exchange, which meant users could transact only in bitcoin and ether, not US dollars and other fiat currencies. And the Regalados hadn’t gotten far in building the community space they’d discussed with their lawyer, having launched just one course. 

“We don’t have to have it perfect,” Eli told Kaitlyn. “Let’s just rock and roll. Let’s just get money in. Let’s get these people off our back.” 

In the days leading up to the launch, the Regalados discussed limiting sales, a practice crypto platforms sometimes use to manage liquidity and volatility. If INDXcoin holders dumped all the currency they’d bought or gotten for free, it would take over $300 million to fulfill sales orders. But Eli kept hearing God say, “Don’t limit me.” He pushed back: “Then we can basically have what’s called a run on the bank, right?” The evening before the launch, the couple prayed again. “Kait + I got the same verse,” Eli wrote in his journal. “Don’t turn selling off.” 

On the morning of April 11, Kaitlyn was beginning to feel optimistic, and Eli was buzzing. “This thing’s gonna explode,” he thought. At 11 a.m., Eli appeared on a livestream. A print of a gray wolf loomed over his shoulder. “Hello INDXcoin family,” he began, clapping for emphasis. “We are live!” 

For investors, returns finally seemed within reach. The exchange initially showed INDXcoin trading at around 10 times what people had paid for it, based on how the crypto market was performing overall; the Bonillas’ $70,000 investment looked to be worth more than $716,000. 

MATT NAGER

But nearly an hour into the broadcast—after slides of Bible verses and rosy projections—a viewer posted a complaint in the chat: “Exchange says I can’t sell INDX.” “It’s probably just because the liquidity isn’t there right now,” Eli explained calmly. “Just wait a little bit.” Ten minutes later, someone else wrote that his sale wasn’t going through. “Just be patient,” Eli said. “The Lord will provide for Himself.”

Over the next few hours, the Regalados kept checking the exchange’s dashboard. Dozens of transactions were rolling in, but the problem was obvious: Sales were dwarfing purchases. By the afternoon, the $30,000 they’d put in to facilitate trades had been drained. They decided to add another $100,000 to the pot. 

A couple hours later, Eli was out getting coffee when he called Kaitlyn to check in. She was crying. “All the liquidity is gone,” she said. 

The next day, the Regalados announced that they were suspending sales. “That was when we saw that we could be in trouble,” Jose Bonilla says. 

Eli told me that after the launch failed, he felt “crushing anxiety” but heard God remind him, “It’s impossible to mess this up.” He and Kaitlyn took steps they hoped would salvage the project, but months passed, and they kept sales on hold.

In June, Jose emailed the Regalados, explaining that he needed to withdraw half of his investment to fund a community development initiative he’d founded in his native Colombia. Eli replied that they had just reopened sales—limited to one coin per day and 10 per month. When they did so, the exchange had around $20,000 available to fulfill sales orders. “Liquidating HALF of your coins is not probable at this juncture,” Eli wrote. Three days after sales resumed, the Regalados halted them again, blaming a technical glitch. 

When Jose followed up a few months later about pulling out half of his investment, Eli replied, “At this time there is zero funds to do that.” In November 2023, the Regalados shut down the exchange and took INDXcoin’s blockchain offline. 

“Shame, condemnation, suicidal thoughts have just been pouring in hot and heavy on me,” Eli shared in a video update, standing before an image of a swirling purple cosmos. “Where did I get this wrong?”

Two months later, the Regalados learned that Colorado’s securities regulator was accusing them of committing fraud and selling unregistered securities. The state soon added to the suit 12 defendants it said had received commissions for selling INDXcoin, alleging that they had also sold unregistered securities. Among them were Eli’s brother-in-law, Daniel Applegate, and a company associated with Gauthier, the INDXcoin investor. A judge entered a default judgment after they failed to respond and ordered them to pay judgments of $15,000 and $34,400, respectively. Eli’s father, Eligio Regalado Sr., who was also accused of securities fraud, agreed to refund $122,000 to friends, relatives, and colleagues without admitting or denying liability. (Gauthier denied wrongdoing; Eli’s father, through his attorney, declined to comment. Daniel denied being a part of INDXcoin and, despite being named in the lawsuit, claims that it has nothing to do with him and his wife.) 

“I really can’t speak to whether or not he heard God tell him to do it,” Chan, the Colorado securities commissioner who filed the suit, told me. “Even if [the Regalados] meant it from the goodness of their heart, the problem is, it’s not fair to the investors … They lied and omitted key things.”

I spoke with 20 INDXcoin investors, and nearly all had heard about the coin from a trusted friend, relative, or faith leader. Most had little or no experience with crypto. They funded their purchases by raiding retirement funds, cashing out a pension, using proceeds from selling a small business, or taking out a home equity line of credit they’re still paying interest on. One buyer, a disabled veteran in his 70s, hoped profits from his investment would help him recover financially after he accrued debt while being treated for cancer. Another, who had retired, was forced to get a job at Home Depot in his late 60s. “It’s a gut-wrenching, horrible, helpless feeling,” he says. 

Investors are divided on whether they were conned. Jose Bonilla, who reported the Regalados to authorities, believes that their actions were “totally intentional.” “They are using a spiritual excuse to defraud,” he says. His wife, Debbie, disagrees and thinks that the Regalados simply “got in way over their heads.” 

A number of people who bought in still support the Regalados. “They’re hearing God’s voice and trying their best to follow it,” says Troy Bramblet, a former pastor who lost more than $18,000 on INDXcoin. “It doesn’t guarantee success.” 

Wheeler, the crypto influencer who advised the Regalados, also alerted authorities about INDXcoin but remains unsure whether the couple set out to fleece people. “They are zealots—they are literally blinded,” he says. “If you believe God is going to do a thing, then are you scamming people? No. But look how they spent their money.” 

In a video posted days after the case was filed, Eli admitted that he and Kaitlyn had in fact “sold a cryptocurrency with no clear exit.” He acknowledged that they had pocketed $1.3 million—including money spent on “a home remodel that the Lord told us to do.”

Last November, I visited the Regalados in the three-bedroom townhouse they rent in a Denver suburb dominated by office parks and cookie-cutter condos. The house they own is uninhabitable—renovations stopped halfway through the project, after they stopped making payments. 

In person, Eli is friendly and charming, with a restless energy and subterranean intensity occasionally betrayed by his stare. He is prone to lengthy monologues delivered with such conviction they make you second-guess bald facts. Kaitlyn, who comes across as reserved yet frank, has “Believe” tattooed on her wrist. They told me that they argued frequently after INDXcoin collapsed, but when I was there, Kaitlyn listened to her husband attentively and always laughed at his jokes. 

On a sunny Thursday afternoon, I followed the Regalados upstairs to a corner of their bedroom containing a tiny desk and a whiteboard. The room was modestly furnished with what they said were secondhand finds. The bed was unmade, and a Bible lay on the floor. 

Eli was preparing to address members of INDXcoin’s private forum in his first live call in nearly two months. He closed his eyes and prayed. “Just allow me to speak simply,” he said, like a teenager asking a parent for a favor. “Just be able to use analogies, to be able to bring it down to their level of understanding.” “Amen,” Kaitlyn said. 

After hunting breathlessly for a laptop stand, Eli grabbed a stack of journals—full of divine revelations—and plopped his computer on top. He switched on the camera, and his image appeared before a faux backdrop of potted plants. Eli had a receding hairline and stubbly beard, and he wore a black T-shirt and a silver cross on a thick chain. Before letting callers in, he ran his fingers through his hair and his tongue over his teeth—now perfect, thanks to cosmetic dental work paid for with proceeds from coin sales.  

“Okay. Awesome. All right. So hey, good afternoon, INDXcoin community!” Eli began, flashing a smile. “We’ve got some exciting updates.” Then, in the tone of a tech founder reporting on a strong quarter, he shared the news: Two months earlier, a judge had ruled against the Regalados in their civil case, and they were now facing criminal charges from the district attorney’s office. 

“Someone asked me, ‘Are you going to do a plea?’” He paused to sip water. “Short answer is no … We haven’t done anything wrong.” 

The Regalados deny orchestrating a scam. “If you’re giving massive amounts of money away at the expense of your own self and family, that doesn’t hold up,” Eli says. The couple estimate that they’ve gifted $300,000 in cash, plus a Harley-Davidson motorcycle, a BMW, and a Louis Vuitton bag, to churches and individuals through sowing. They also gave away millions of INDXcoin—90% of the supply. (Eli told me, “No one sows without expecting something in return,” though not necessarily from the recipient.) 

In their civil case, the Regalados represented themselves because they couldn’t afford lawyers. They argued that INDXcoin wasn’t a security because it was a utility coin and that the price was set by “immutable algorithm.” They claimed that their technology provider had caused the exchange to fail, consultants had led them astray on compliance, and attorneys had said they didn’t need to maintain liquidity or disclose spending. (Benemerito, the lawyer the Regalados had retained, told me, “Our firm does not advise clients to violate the law.”)

The judge disagreed, finding that INDXcoin was a security and that the Regalados had misled investors about its true value and risks, where their funds went, how many coins had been given away, and more. Noting a “lack of understanding of the harm they have caused,” she ordered them to pay nearly $3.4 million in damages—the amount of money they’d raised. “Ascribing an algorithmic value to a coin does not make it ‘worth’ that amount,” the judge wrote. “In reality, INDXcoin was worthless because no one wanted to buy it.”

When I visited, two months had passed since the ruling. The Regalados still hadn’t read the judge’s opinion in full but had decided to appeal. Later, they would draft briefs with help from Google Scholar and AI. (The case is still pending.) 

Besides filing court documents and preparing for their criminal case, the couple spend their days like typical suburban parents: taking their kids to playgrounds, walking their chiweenie, working out. They still host biweekly Bible studies. Sometimes they ride their Harley to Palmer Lake or the Rocky Mountain foothills. (“We only wear helmets when it’s windy or cold,” Kaitlyn says.) Their assets were frozen soon after the civil case was filed; Eli had found work selling roofs but says he was fired when his employer learned about his legal troubles. He declines to disclose his current gig. “It’s not related to marketing and not related to crypto,” he says.

After they were sued over INDXcoin, Eli wondered, “Did I just make this up? Am I crazy?” But he and Kaitlyn concluded that the divine signs they’d received were unmistakable. They believe that INDXcoin will eventually gain traction among world leaders losing faith in the US dollar. “We are privately making preparations,” Eli told me.

“God already saw this coming,” he assured viewers during the November video update. “He’s looking at us and saying, ‘Are you willing to believe me no matter what you see?’”

After the call ended, Eli began leafing through his journals and reading sections aloud. Since our first conversation months earlier, the Regalados had been remarkably amenable reporting subjects. They told me that their criminal defense attorneys had advised them against talking to reporters, but they sat for more than a dozen interviews with me. They provided access to INDXcoin’s private forum and supplied emails, photos, and spreadsheets—even though some documents don’t paint their decision-making in a favorable light. Once, Eli emailed to “come clean” that an anecdote he’d told had been slightly embellished. He apologized and assured me, “Everything else I have said is 100% in line with no stretches or exaggeration.” 

The Regalados told me they trusted me in part because God had signed off: Not long after I’d first contacted them, they’d walked into a room with a TV playing Family Feud, and the answer displayed on the screen was “MIT.” Their approach highlighted how they had won over buyers so effectively: They were likable, shared vulnerable details, and telegraphed transparency.  

Still, the Regalados didn’t appear to be feeding me an act they’d just cooked up. Instead, they seemed fully committed to their own narrative: one that paints them as righteous underdogs fulfilling a holy mission, no matter the cost. To let their faith waver would mean that everything they had lost—friends, their home, their reputations—had been in vain. It would mean admitting that they had failed. It would mean that no one was coming to save them. 

Even ending up in prison wouldn’t persuade the Regalados that they’d misheard God. “He’s going to deliver you from everything, so you won’t be there forever,” Kaitlyn says, “and it might just be part of the story.”

During my visit, the Regalados agreed to show me an earlier chapter. We piled into their Ford Raptor truck, their kids in the back, and drove 20 minutes north to a quiet cul-de-sac in a leafy residential neighborhood. 

We slowed near a hulking structure of rotting wooden boards. Red and brown weeds engulfed the lot and threatened to swallow the sidewalk. Out front, a tattered mattress was slumped on its side. Neighbors had sighted squatters and, as winter approached, feared fires. The Regalados still owed their contractor nearly $110,000 for work completed. 

MATT NAGER

I asked whether we could get out, but Eli and Kaitlyn didn’t want to run into anyone. “I just don’t want to have a conversation of like, ‘When are you gonna cut your grass?’” Eli said. (The city had sent them violation notices the previous year for not maintaining the property.)

As we drove away, I asked how it felt to see the ghost of their dream home. 

“It used to hurt,” Kaitlyn said. 

“Here’s this unfulfilled promise,” Eli added.

But it didn’t bother them anymore. 

“If we lose the house,” Kaitlyn said, “that means we’re getting something way bigger and way better.” 

They made a U-turn at the end of the street and, seat belts unbuckled, rounded the corner without looking back.

Katia Savchuk is an independent journalist based in the San Francisco Bay Area. Her work has appeared in the New Yorker, Forbes, Mother Jones, and many other publications.

Battery installations hit a new record in the US in the second quarter of 2026. In total, 20.2 gigawatt-hours of new capacity came online, according to a new report. That’s enough to supply the daily electricity needs of about 700,000 homes.

The surge is putting the country on a trajectory to see 71 gigawatt-hours of batteries installed in 2026, a 20% increase over last year. This growth is being driven by a combination of cheaper batteries and an urgent need for more energy storage capacity as renewables such as solar and onshore wind power are added to the grid. 

Massive, utility-scale systems are leading the way; they’re responsible for most of the record-setting quarter. Seven new gigascale battery installations (those with a capacity of over one gigawatt-hour) came online during the three-month stretch, according to the report, published by Benchmark Mineral Intelligence and the Solar Energy Industries Association.

“It really came down to a handful of big projects,” says Shan Tomouk, energy storage and energy lead for Benchmark Mineral Intelligence.

But there was also growth in the category of so-called behind-the-meter batteries, which include both residential and industrial battery storage systems. These projects, generally smaller than utility-scale installations, are typically owned and operated by homeowners or businesses rather than utilities or power providers. 

In the behind-the-meter category, data centers led the way, making up about three-quarters of new batteries in the commercial sector. But residential batteries saw a sharp slowdown. These systems are often installed in homes to store power from solar panels or serve as a backup source in case of a blackout. Home installations are projected to drop by 16% in 2026 compared with last year, according to the report.

That drop happened largely because a tax credit that helped subsidize home battery systems ended in 2025, Tomouk says. Home installations should recover by the end of the decade, he adds. And tax credits for nonresidential batteries have largely survived.

Overall, batteries are a bright spot in energy right now. “This is one of the strong sectors in the US,” says Isshu Kikuma, an energy storage analyst at BloombergNEF, an energy consultancy.

As the battery market continues to grow, one major trend to keep an eye on is a move toward US-made technology. Today, nearly all the systems coming online use cells made in China, though some are put together into complete energy storage systems in the US.

Tariffs were already pushing the US energy storage industry toward domestic production. And beginning this year, energy storage tax credits required projects to limit their reliance on batteries imported from China. There’s a lot of manufacturing capacity set to come online in the US, though these factories probably won’t be able to meet demand until at least 2030 or so, Tomouk says, so prices could tick up.

OpenAI’s latest mathematical milestone has quickly become mired in controversy. Today, the company announced that its agents have solved one of the Millennium Prize Problems, some of the most important open problems in mathematics. Under normal circumstances, that solution would be a huge feather in OpenAI’s cap.

But the announcement has been overshadowed by accusations that OpenAI used NYU mathematician Tristan Buckmaster’s and Anthropic employee Levent Alpöge’s AI-assisted work on the problem as a jumping-off point and failed to credit them. OpenAI has denied the accusations.

It remains uncertain if OpenAI’s models made use of the work completed by Buckmaster and Alpöge, though Sébastien Bubeck, a member of the technical staff at OpenAI, said in a press briefing that the team was inspired to pursue the problem after hearing a rumor about Buckmaster and Alpöge’s efforts. But whether or not OpenAI’s models took advantage of Buckmaster and Alpöge’s research, this episode may mark a turning point in the history of mathematics.

AI models now seem essential for making progress on the most important mathematical problems of our time, and solving them may demand resources only available at a couple of frontier AI companies, which often defy the norms of academic collaboration that undergird most mathematical progress. If that’s the future we are headed for, it is unclear how human mathematicians will fit into it. 

The problem that OpenAI claims to have solved is known as the Navier–Stokes existence and smoothness problem. It is one of seven Millennium Prize Problems selected by the Clay Mathematics Institute in 2000. Solutions come with a one million dollar prize; before today, only one other Millennium Prize Problem had been solved. 

The Navier–Stokes problem concerns a set of equations that describes how fluids, such as water and air, flow over time. The equations are widely used in the field of fluid dynamics, and they have proven powerful, but physicists and mathematicians didn’t understand them completely. In particular, it was unknown until today whether the equations might, under some conditions, break down and predict an impossible state of affairs—such as a fluid having infinite velocity.

On Monday, NYU’s Buckmaster posted a proof on the social media site Mastodon showing that a simplified version of the Navier–Stokes equations can indeed break down—a major step forward on the Millennium Problem. He and Alpöge had worked on the problem for almost a year, using publicly available models from both OpenAI and Anthropic.

Then today, OpenAI presented a proof showing that the full Navier–Stokes equations can break down as well. The proof was obtained using an internal model that dramatically outperforms the already-impressive Astra model, which was only released last week. The company says it does not plan to claim the million-dollar prize for solving the problem.

These mathematical achievements are indisputably impressive, but they have attracted far less attention than the controversy about their origins. Along with the proof, Buckmaster posted a document detailing his interactions with OpenAI employees after he heard rumors about their work and reached out to one of them. According to him, OpenAI employees presented two possibilities to him: Either he and Alpöge could post their work and OpenAI would post their Navier-Stokes solution the following day, or he could work with OpenAI on a Navier-Stokes paper that excluded Alpöge from authorship, due to his affiliation with Anthropic, OpenAI’s biggest rival.

Buckmaster also wrote that he asked the employees whether the agents had obtained access to transcripts of the work that he and Alpöge had done with OpenAI models, which they denied; and whether OpenAI models had been trained on those transcripts, to which they offered no response. MIT Technology Review reached out to Buckmaster for comment, but didn’t hear back before publication.

The clear implication of the document is that OpenAI’s models somehow made use of Buckmaster and Alpöge’s work. That scenario is plausible on its face. The Buckmaster/Alpöge and OpenAI proofs both make use of an approach to the Navier-Stokes problem pioneered by the mathematicians Diego Córdoba and Luis Martínez-Zoroa.

According to Javier Gómez-Serrano, a mathematics professor at Brown University, this approach was one of several that was thought to hold promise for solving the Navier-Stokes problem. So, while it’s by no means impossible that both teams could have arrived at this approach independently, it’s also conceivable that Buckmaster and Alpöge’s work could have influenced OpenAI’s.

In the press briefing, Mark Chen, OpenAI’s chief research officer, again denied that any agents or OpenAI employees accessed Buckmaster and Alpöge’s transcripts—but given what has been revealed about the Hugging Face hack, it’s clear that OpenAI is not always entirely aware of what its agents are doing. 

If OpenAI’s models did train on Buckmaster and Alpöge’s work, or if its agents somehow gained access to it, then the company’s failure to track down the truth and assign those researchers appropriate credit reflects poorly on it. But there might be a thin silver lining to that version of the story for mathematicians, because it would suggest that the hard work of two humans, one of whom is a prominent expert on Navier-Stokes, was essential to the agents’ ability to solve the Millennium Problem.

Experts have long identified “research taste,” or the ability to choose promising research questions and directions, as a major obstacle for AI in science and mathematics. If the OpenAI agents did indeed choose to follow the Córdoba–Martínez-Zoroa approach because Buckmaster and Alpöge had done the same, then human research taste played an essential role in OpenAI’s success.

Even so, the bigger picture here is sobering. The progress that Buckmaster and Alpöge made over almost a year of collaboration with publicly available models speaks to the promise of human–AI collaboration. But they were not able to achieve a full solution. Meanwhile, OpenAI brute-forced a solution in a few days using an internal model, and their successful solution came at an astronomical cost: In the press briefing, Bubeck and Chen said the team was only able to solve the problem by running about 10,000 agents concurrently, at a cost of millions of dollars.

Over the past few months, I’ve heard from several researchers that mathematicians are becoming depressed, and it’s not difficult to see why. Mathematics is quickly becoming the province of frontier AI companies with impressive internal-only models, money to burn, and a lack of collaborative spirit. “Whether AI companies will decide to spend their money on doing one thing or another, I truly don’t know,” says Gómez-Serrano. “What is clear is that very few mathematicians will have resources of that scale.”

If OpenAI and Anthropic keep striving for more and more impressive mathematical accolades, there might not be any open problems left for human mathematicians outside of those companies to wrestle with. That would dramatically change the field of mathematics.

Last week, UCLA mathematician Terence Tao wrote a Mastodon thread describing how important mistakes, wrong directions, and incomplete solutions are for the field. “In most cases in pure mathematics, the problems are posed not because we desperately want the solution to these problems in and of themselves, but because we have seen from past experience that human-directed efforts to solve these problems tend to spur further development of the field,” Tao wrote.

“Prematurely solving the problem by purely AI-powered methods—particularly without full transparency into the solution process—can contaminate this process to the point where it actually becomes a net negative for the progress of mathematics as a whole.”

Humans might take longer than agents to solve mathematical problems, but in the process, they uncover new mathematical approaches and ideas that might inspire their peers and even birth their own subfields.

But when AI agents solve those problems instead—and when private companies keep the agents’ wrong turns from public view—those benefits disappear. It remains to be seen what else will vanish in the process. 

Throughout the history of the computer chip, engineers have treated waste heat as an inevitable cost of a calculation. Hannah Earley, however, thinks it’s a design choice. Earley, 31, is cofounder and chief technology officer of Vaire Computing, a startup building chips that recycle energy usually thrown away as heat—a strategy known as reversible computing. Ultimately, she thinks, this approach could help make data centers (and our laptops and phones) much more energy efficient. 

When conventional computer chips perform calculations, they erase the information they no longer need along the way, dissipating energy as heat in the process. Earley compares the approach to racing through a city only to pump the brakes at every intersection: The car loses momentum and must burn more fuel to accelerate again. Reversible computing aims to keep the momentum going—instead of erasing information from the intermediate steps in a calculation, the circuit retains it, making it possible to run the computation backward and recover some of the energy.

While the idea was first proposed more than 50 years ago, it proved impractical to implement with existing transistors and circuits. Earley, though, has completely rethought the hardware needed to make energy recovery work. She designed a patent-pending type of resonator—a microscopic chip component that stores recovered energy for later reuse. “It’s really a glorified pendulum,” she says. Last year, Vaire announced a key breakthrough: a chip with a resonator that recovered more energy than it lost, even after the energy needed to power the component was taken into account. For a subfield that has existed mostly in theory, the result was proof of life.

“It’s clear they have something interesting,” says Igor Markov, a researcher in electronic design automation and a former professor at the University of Michigan, Ann Arbor. Still, he says, the technology is quite early stage; the company will need “a series of increasingly realistic and convincing demonstrations to attract the industry support needed for commercialization.” 

She gradually became convinced that the connection between information, energy, and heat could change computers forever.

Earley’s journey into chip design started sooner than most. She began programming around the age of nine, starting with high-level coding for the web before digging into other programming languages like Perl and Java. She continued progressing to more and more abstract layers of computing, until she got all the way down to transistors.

She eventually enrolled in a PhD program at the University of Cambridge under the computational biologist Gos Micklem. She started out studying how materials such as DNA could be used to perform calculations, but a few months in, Micklem sent her the 1999 PhD thesis of Michael Frank, a pioneer in reversible computing. Earley read it once, felt skeptical, read it again, and sat with it for a few weeks. She gradually became convinced that the connection between information, energy, and heat could change computers forever.

The fascination completely redirected her PhD work. Earley studied the physical limits of computation and built software that could turn ordinary programs into reversible ones. “Eventually I wouldn’t let her put my name on any of her papers, because I felt that I couldn’t really stand up and give a proper talk about them,” Micklem recalls. “It was her stuff.”

After completing her degree in 2021, Earley met Rodolfo Rosini, a technology entrepreneur and investor. The pair cofounded Vaire that same year, and the company has since raised more than $12 million, hired Frank as a senior scientist, and begun turning the vision of reversible computing into real hardware.

Innovation, however, doesn’t happen overnight. During the winter of 2022 in Grinnell, Iowa, Earley spent weeks in her now-wife’s basement apartment as the wind chill outside reached roughly −40 °F, covering a whiteboard over and over again with schematics for the core piece of circuitry needed to make reversible logic work. By the time the design finally came together, after the couple had escaped the cold for Las Vegas, it felt less like an aha moment and more like a gradual wave of relief. “I’m not completely out of my depth,” she remembers feeling. 

Earley and her colleagues’ next challenge is making their drastically different chip fit into familiar devices and manufacturing systems. She believes that’s where the future lies—not in further refining existing chips but in rebuilding them from the ground up with an eye toward reversibility. “I want to tackle every part of how computers are built,” Earley says, “and rethink it in these terms.” 

Danijar Hafner’s office in San Francisco’s SoMa district sits mostly empty. His brand-new startup is still in stealth mode and doesn’t even have its name on the door. On the day I visit, there’s only one other person there, and little in the way of furniture. But what it lacks in decor, it makes up for in robots. Humanoids of various shapes and sizes hang like marionettes from racks that run down the center of the wide-open space.

While Hafner, 31, won’t say too much about his new venture just yet, he describes it as a continuation of his longtime work to enable AI to navigate environments it has not encountered in training. The humanoids, which he imports from China, are the next evolution of this work—and its physical embodiment. Their ability to react in previously untested scenarios will be key to getting robots into human spaces. Because if you want to send a robot into a person’s home, for example, it needs to be able to handle a floor plan and furniture it’s never seen before. 

To achieve this, Hafner relies on something called model-based reinforcement learning. He develops world models—AI models designed to emulate physical reality—and trains agents within them. The agent essentially treats the model as a real-world simulation and learns how to act there. It then uses those experiences to make predictions (to dream or imagine, Hafner might say) about future outcomes. That allows agents—or the robots they’re embedded in—to navigate unfamiliar situations IRL.

“I get to interact with a lot of really smart people in research at Google, and he easily sits in the top half of 1%.”

Timothy Lillicrap, Google DeepMind

Unlike other efforts, Hafner’s technique enables agents and the robots they control to execute massively complicated tasks without the real-world trial-and-­error training that’s traditionally been used in robotics. 

Hafner grew up in a rural town in northeastern Germany, where his parents were both classical musicians. He learned programming from a neighbor, and in high school he began taking online courses about AI, which quickly developed into a passion. “I was always fascinated with how thinking works,” he says. AI offered him a way to emulate it on a computer.

In 2015, as a second-year under­graduate studying engineering at Hasso Plattner Institute in Potsdam, he won a role as a student researcher at Google Brain. From there, he went on to a dozen internships and other positions at the company, including stints with Google Brain and Google DeepMind (the two have since merged under DeepMind) in the UK, Canada, and the US. He worked with industry legends including Geoffrey Hinton, who is often referred to as one of the godfathers of AI, and Ashish Vaswani, coauthor of the groundbreaking research paper “Attention Is All You Need,” which described the transformer technology used by today’s large language models.

One of Hafner’s former managers and coauthors at Google, Timothy Lillicrap, describes him as a standout among standouts. “I get to interact with a lot of really smart people in research at Google, and he easily sits in the top half of 1%,” Lillicrap says. “In many cases he would build, single-­handedly, things it would take entire teams of engineers to build.”

Over the years, Hafner has honed and proved his approach by pitting agents trained within his world models against popular video games. His first breakthrough was PlaNet, a model that allowed agents to execute actions by planning ahead. His Dreamer 2 was the first agent to hit human-level performance playing Atari 2600 games using a world model. Dreamer 3 was the first one to solve the Minecraft Diamond challenge—successfully mining in-game gems on its own. And Dreamer 4 went a step beyond that by learning to mine diamonds from an offline data set of recorded game-play videos, without ever interacting with the game directly. 

More recently, he’s begun to migrate his agents out of the virtual world and into physical reality. His DayDreamer project used the Dreamer algorithm to let robots operate themselves in novel environments and react to new experiences (such as being pushed over) without any specific training. 

Today, Hafner is working on his new startup, which he left Google DeepMind to form in the fall of 2025. Though he’s coy about his next steps, it’s clear he’s dreaming big: “I was interested in solving a problem,” he hints, “that would change the world.” 

The steel industry isn’t exactly known for innovation. Very little has changed about purifying iron ore since the process was invented and commercialized in the 1850s.

The majority of steelmakers melt solid iron ore at dizzyingly high temperatures inside blast furnaces, where the material reacts with gases to trigger chemical reactions that remove oxygen. It then undergoes further refining to purify it before it is made into products like rebar and car frames.

The process relies on coal, and it generates roughly 7% of the carbon emissions that drive climate change—about as much as the fashion industry. Decarbonization has proved difficult: Profit margins are tight and furnaces have long service lives, making investment tough to justify.

Now Laureen Meroueh may have found a way to clean up steelmaking without driving up the price. Meroueh, the founder of Hertha Metals, invented a new furnace that simplifies the chemistry behind the process. Her method turns iron ore into refined liquid steel in a single step, and it swaps coal for natural gas. Together, those changes slash emissions by at least half, she says, and cut costs by 25% compared with steelmaking business as usual.

If it catches on, the tech could be transformative. “There’s huge value in reducing the size of this production system,” says Iryna Zenyuk, director of the National Fuel Cell Research Center at the University of California, Irvine. “They’re massive. They’re inefficient and require a lot of energy input, so even if they just save energy efficiency, that’s already a big step.”

Hertha’s approach focuses on what it can fix about the steel industry now, as opposed to waiting around for a zero-carbon system.

Still, it’s a risky endeavor, but pushing limits isn’t new for Meroueh. At 12 she was accepted into a pilot program to take college-­level courses through Florida Atlantic University in lieu of a traditional secondary education. She was immediately drawn to engineering and explored topics including calculus and ocean wave energy.

Despite the rigorous coursework, she would spend hours sitting in trees and surfing, which fostered a deep appreciation for nature and a desire to safeguard it. “I don’t know how you can’t be drawn toward trying to help protect that,” she says. 

Now 34, Meroueh has let that passion inform her professional goals. After finishing her PhD in mechanical engineering at MIT, she led a green hydrogen startup before founding Hertha in 2022. A first-generation Lebanese-American from an entrepreneurial family, she saw starting her own company as a typical path. “Seeing how common it is to take that jump to start your own business is what made me feel like ‘This is normal,’” she explains on a video call from her office at Hertha’s pilot plant in Conroe, Texas, just north of Houston.

That facility can produce one metric ton of steel per day. “One ton per day is a big metric for steel,” says Rajesh Swaminathan, a partner at Khosla Ventures, one of the company’s investors. (Hertha had raised about $20 million in funding as of July 2026.) 

Swaminathan says the company’s scale-up is “impressive,” especially given how little the team has spent. Competitors, he notes, have created far less steel with $50 million or $100 million in funding.

Hertha’s approach focuses on what it can fix about the industry now, as opposed to waiting around for a zero-carbon system. While other approaches to making green steel center on using hydrogen to free oxygen from iron ore—a method that could one day cut or eliminate emissions—Meroueh says Hertha is content for the time being with a continued reliance on fossil fuels, mainly to keep costs down. The current Hertha plant could eventually switch to a fully decarbonized system without drastically changing the hardware, she says, if hydrogen becomes more affordable. 

In the meantime, plans are underway to expand into a new plant next to the existing one. The facility is slated to produce 10,000 metric tons of high-purity steel per year and should reach full capacity by the end of 2027. By 2030, Meroueh believes, Hertha can up its output to 500,000 metric tons per year with the addition of a third site. That’s only a fraction of the approximately 80 million metric tons of steel produced annually in the US, but Zenyuk says making even one metric ton is still an achievement.

In Meroueh’s mind, the world isn’t going to outgrow its need for steel, so she’s asking another question: “How can we be smarter about how we make things … so that it’s also not going to harm us in the long term?”

Yuancheng (Ryan) Lu is obsessed with aging. And with eyes. As he steps outside the Whitehead Institute in Cambridge, Massachusetts, his aviator glasses darken automatically in the sun. Age-related blindness runs in his family. A great-aunt in China, the story goes, was killed crossing a road because she couldn’t see oncoming traffic. And Lu’s own 23andMe test came back with a mutation for macular degeneration, a top cause of vision loss in old age. Exposure to bright sunlight is another risk factor—thus the shades. “They protect me,” he says. “Plus, they look cool.”

Lu, 34, works on gene therapies to prevent age-related vision loss. “I think the eye is a really unique system to study aging and rejuvenation,” he says. “I could give a whole presentation.” Pushing up my reading glasses, I lean in to listen.

Lu is behind one of the coolest results in rejuvenation science—and in eye research. In 2018, while earning his PhD at Harvard Medical School, he used an age-reversal technique called reprogramming to repair the optic nerves of mice. He crushed the nerves, blinding the animals, and then injected the cells with a gene therapy meant to restore them to a youthful state. Sixteen days later, the nerves were growing back, their axons showing up through a microscope as spidery orange filaments.

As hype around age reversal swirls, Lu has been busy in the lab searching for what he calls “the next generation of rejuvenation therapies.”

The head of that lab, the longevity scientist David Sinclair, remembers when Lu texted him the pictures: “He asked me, ‘What do you see here?’ And I said, ‘I see the future.’” Later tests carried out in a box with rotating bars of light showed the mice were tracking the changes. They could see again.

This year, nearly the exact genetic therapy Lu created for mice entered human clinical trials. On June 9, the startup Life Biosciences, which Sinclair cofounded and in which Lu owns a small stake, announced it had injected the treatment into the eye of a person with glaucoma. The trial has been big news. A headline in the New York Times suggested the technology could “change humanity.” Posters on X gushed, with one declaring that “the fountain of youth is here.”

“It’s remarkable that what he developed as a student is now going into humans,” says Sinclair of the treatment, now called ER-100. “It’s barely even changed since he built it.”

Reprogramming refers to an age-­restoring process that takes place inside an embryo. It’s why babies are born young, not old: The DNA they’ve inherited from their parents has been scrubbed and reset. In 2006, Japanese researchers showed they could cause the process to occur in the lab by introducing just four key genes, known by the acronym OSKM. Add these to a cell from a 100-year-old and it will turn into a stem cell that acts as if it was plucked from an embryo.

That’s powerful stuff. But we don’t want to turn people into blobs of stem-cell protoplasm. Lu figured out a way to control the effect. He trimmed the list of genes to just OSK—leaving out M, for Myc, the one most likely to cause dangerous changes like cancer. His extra flash of insight was that reprogramming could be tested on the optic nerve; the eye is particularly accessible.

Lu’s result, published in Nature in 2020, helped set off an investment rush. Since then, US tech billionaires have placed huge bets on private companies like Altos Labs and NewLimit to explore reprogramming and anti-aging medicine. The day I spoke with Lu, he’d spent the morning meeting with the business magnate Zhong Shanshan, one of China’s richest people.  

Still, as hype around age reversal swirls, Lu has been notably absent from the public conversation. He’s been busy in the lab searching for what he calls “the next generation of rejuvenation therapies.” With a sigh, Lu describes the grueling effort over the last six years to understand what OSK really does. The treatment remains toxic to many cell types, and he says it’s becoming obvious that different factors drive aging in each kind. This year, for example, he identified a gene responsible for protecting the retina from damage by free radicals—the main cause of age-­related macular degeneration.

While Sinclair, his former boss, believes humans could live to be 200, Lu disagrees. There’s just too much that goes wrong as we age. His work with OSK, he says, was more a proof of concept than a silver bullet. But it did change the conversation. “Six years ago, you couldn’t talk about rejuvenation. We didn’t use that word—there was pushback,” Lu tells me. “But I think people have accepted the concept that you can really reverse molecular age.” 

Haven’t you heard? White-collar jobs are going away, decimated by AI. Waves of layoffs in the tech sector (most recently at Coinbase and Meta and Cisco) are said to presage what will soon come for all of us knowledge workers. But before you quit your job as a software developer or financial analyst—or tech journalist—and look to join the plumbers’ union, it’s worth considering today’s economic research on whether artificial intelligence has actually begun to devour white-collar work.

Analysis of the data gathered for the US Bureau of Labor Statistics (BLS) shows that the unemployment rate for the jobs potentially most affected by AI is actually lower than that for occupations less exposed to the technology. And, critically in the mind of economists, there are no signs that large numbers of people are shifting from jobs threatened by AI to supposedly safer ones, such as those involving mostly manual labor.

While the current labor statistics don’t preclude a sudden job upheaval in the coming years, they do throw doubt on the inevitability of the doomsday scenarios and the pace at which they’d unfold. Everyone in the AI community, it seems, is predicting that the technology will soon wipe out jobs, and everyone, it also seems, knows some young wannabe workers who can’t find one. Perhaps we haven’t seen any major disruption in the labor market statistics yet, people often say, but just wait. 

But maybe we should pay attention to what the data is showing us. And right now, the numbers paint a picture of a relatively stable labor market in which AI disruptions remain largely speculative.

“It could be disruptive, but the data is telling us right now that disruption is not yet here, and we have time to plan.”

“All of the available evidence to date suggests that AI’s impact on current labor market conditions is likely small right now,” says Erika McEntarfer, a labor economist who headed the BLS until President Trump fired her last fall after a jobs report that displeased the administration. (Not surprisingly, BLS reports of sluggish job growth have continued since her dismissal.)

McEntarfer, who is now a fellow at the Stanford Institute for Economic Policy Research, says the relatively small impact that AI is having so far on today’s labor market “surprises many people, but it shouldn’t. What we know from history is that it takes time for innovations to work their way through changes in industries and changes in occupations. AI is unlikely to transform labor markets until it first transforms businesses.”

McEntarfer points to US Census data showing that only one in five companies are using AI in any business function. “The data are a great reality check on the fear that AI will be enormously disruptive,” she says. “It could be. It likely will be disruptive, but the data is telling us right now that disruption is not yet here, and that we have time to plan.”

Things ain’t great—but the question is why

The US job market, to be sure, sucks for many, especially younger would-be workers. Unemployment rates for recent college graduates stand at around 5.6%, well above the level for all workers. It’s a rate not seen since the pandemic and the years immediately after the 2008 recession. Even more troubling is that hiring rates have been particularly dismal during the post-covid economy, a trend that hits hard at young people trying to enter the workforce. If you’re a recent college graduate and looking for a tech job, no one, it can seem, is hiring.

There are signs that AI is contributing to the pain for the 22-to-25-year-olds seeking jobs in software development and other occupations that are feeling a big impact from AI. But these professions represent just a sliver of the overall labor market. What’s more, it’s uncertain how much blame AI should get for the job woes. Similarly unknown is whether the loss of entry-level jobs in AI-exposed occupations is a harbinger of what’s coming for others or simply an isolated symptom of what economists refer to as a “low-fire, low-hire” labor market caused by a variety of macroeconomic forces.

Insights into these uncertainties will tell us much about our working fates in the transition to an AI economy. There are no shortage of confident assertions and predictions about what is about to happen; while some people forecast the end of work, others say economic history teaches us that technology advances always lead to more and better jobs eventually. 

The honest answer is that no one knows for sure what AI will bring and whether this time will be different. To help figure it out, we need better and far more comprehensive data.

The statistics gleaned from the federal government’s monthly survey of 60,000 households for the BLS provide a broad overview of the changes to the labor market, while academics and even some AI companies have begun trying to gain a more granular view of specific jobs that are being affected. But the existing data-gathering tools don’t adequately explain how AI is affecting the huge and diverse US labor market.

There’s a long list of questions that we don’t have the data to fully answer. How is AI being used in the workplace? Does the increased use of AI mean the technology will replace workers, or will it make them more productive and valuable? Which occupations and skills are most affected? Who is in most peril from the changes? As David Deming, a professor of economics at Harvard University, puts it: “We’re sort of flying blind.”

To gather more insight into some of these questions, Deming and his colleagues have been surveying several thousand people every three months since 2024, asking them basic questions: Do you use generative AI, and how often? Does it save you time at work? Tracking the answers over time gives the economists important clues (it’s used by a little over 40% of workers but adoption varies by sectors) and allows them to estimate productivity gains (they’ve found some, but nothing economy-shaking). It has also helps document how quickly AI has been adopted in the workplace and how it compares with earlier technologies such as the PC and the internet (the pace has been faster but roughly in the same ballpark).

It’s far from a complete picture of how AI is changing work. But it provides some intriguing results; for example, a fair number of workers in manufacturing and other industrial sectors have tried AI. Deming’s results show that while businesses in general might be relatively slow to formally adopt the technology, lots of their employees are using it.

The young are most vulnerable

But by themselves the exposure results are not a true predictor of which jobs will be lost to AI. That depends on the kinds of tasks done by the technology, the extent to which the AI is adopted, various business calculations about the value of workers, and even the costs of deploying AI. But the exposure findings are a valuable starting point. 

In a working paper called “Canaries in the Coal Mine? Six Facts about the Recent Employment Effects of Artificial Intelligence,” researchers at the Stanford Digital Economy Lab looked at 950 jobs, placing the occupations into five categories from least exposed to most. Then they used a vast data set from ADP, the world’s largest payroll provider, to look at employment growth in each of the categories. Their exclusive access to the ADP data set, which is far larger than the one available through the BLS, allows the researchers to better spot impacts by demographic. When they examined what was happening to different age groups, says Erik Brynjolfsson, the director of the lab who led the effort, “it was extremely striking.”

They spotted the drop in head count for 22-to-25-year-olds in the most exposed occupations, such as software development and customer service, beginning in late 2022, when ChatGPT was first publicly released. Other researchers reported evidence that the decline in these jobs began well before ChatGPT and questioned whether the labor market could react so quickly to the introduction of AI technology. 

But while the Stanford researchers acknowledge that other factors in addition to AI probably contributed to the early declines, they say that after controlling for those factors, they saw convincing evidence of a significant effect from AI after 2024 and growing in 2025 to a 16% decline in entry-level jobs in AI-exposed occupations. In contrast, head count grew for older workers in the same occupations, as did the number of jobs in the less exposed occupations.

Digging deeper into the data, the researchers found another important clue, though one that wasn’t totally unexpected. The impact on head counts depended on how AI was being used. It was specifically the jobs where tasks could be automated (that is, AI could do them “with minimal human involvement”) that accounted for the decrease in employment—jobs for people like software developers. In jobs where AI was mainly used but to augment human work, head counts grew faster than the average for entry-level workers.

That’s consistent with one explanation for the woes of many young would-be workers. It could be, according to the Stanford paper, that entry-level jobs depend more on the types of knowledge that people acquire through education but that can readily be mimicked by AI; the authors call this codified knowledge. It might be particularly easy to automate such tasks as entry-level coding. In contrast, older workers have more so-called tacit knowledge, the type based on their experience. That type of wisdom is harder for AI to replace.

Despite the findings about AI’s impact on young workers, Bharat Chandar, an economist at Stanford and one of the authors (along with Brynjolfsson and Ruyu Chen), stresses that it’s still early when it comes to understanding how the technology will affect jobs in the future. It could be that the job loss will spread to older workers and to less AI-exposed occupations, he says. But Chandar says it is also possible that firms and workers will adjust to shifting labor demands, and the effects will level off or even disappear.

A recent paper by economists at the Federal Reserve Board found, not surprisingly, that annual employment growth for coders has slowed significantly—by about 3%—since the introduction of ChatGPT. But here’s a critical detail: Overall employment for coders continues to grow. Employment in coding jobs is still rising, they noted, just more slowly than before 2022. 

In short, coding jobs are not going away, at least not anytime soon. But it’s an occupation that is clearly being transformed by AI.

One of the somewhat surprising wrinkles uncovered by recent research is that wages in sectors highly exposed to AI have risen relatively fast since the introduction of ChatGPT. One explanation is that employers are still willing to pay for the kinds of knowledge and experience that are, at least for now, hard to replace with AI. If true, this suggests not the end of work in AI-exposed jobs but, more specifically, the demise of the typical career model in which young graduates are hired to do software tasks that can be automated and are slowly trained to gain that valuable tacit experience. The earn-while-you-learn model might finally be broken—at least for some occupations.

The simple truth could be that coding skills are no longer a guarantee of a job. That may help to explain the drop-off of computer science majors at schools around the country. Future canaries in the cubicles are sniffing out the dangers of looking for a job when their skills can be matched by AI.

But a closer look at the data shows that students are not necessarily turning away from AI-related careers. Rather, they appear to be tailoring their skills to the changes they see underway as AI becomes increasingly important for various disciplines. Interest is rising in AI-adjacent fields like data science and cybersecurity. One fast-growing major: artificial intelligence itself (a recent addition to many college offerings).

Is this time different?

Anxiety over the potential of AI to replace workers is nothing new. I wrote “How Technology Is Destroying Jobs” in 2013, describing how a slew of new digital technologies, including AI, were beginning to threaten white-collar work. I wasn’t alone. It was a popular theme at a time when the labor market was sluggish and jobs were scarce. 

In one of his last days in office in late 2016, President Obama issued a report written by his top economic and science advisors warning that AI was threatening workers. Among the findings was that automated vehicles—especially driverless trucks—could eliminate 2.2 million to 3.1 million existing US jobs.  Around the same time, one of the pioneers of AI, Geoffrey Hinton, said that “people should stop training radiologists” because it was “completely obvious” the occupation was soon to be replaced by AI.

Perhaps this time is different, and we can put aside the lessons of economic history. Certainly, AI has gained unimaginable powers to do humanlike tasks. Perhaps it will devour jobs in ways that we’ve never seen before. And perhaps that will happen abruptly, without a warning buried in the labor statistics. But the previous bouts of AI job anxiety still hold a prescient lesson: Our real focus needs to be less on the dystopian fears and more on the very real transitions in the workplace that will likely affect millions of people.

“Even if there is not mass or even increased unemployment, the transition could still be very difficult,” says Jed Kolko, senior fellow at the Peterson Institute for International Economics and former undersecretary of commerce in the Biden administration. “And what does a difficult transition period mean? It means people losing jobs, or people’s jobs being redefined in ways that make those jobs pay worse or be less meaningful. And some people whose jobs are threatened may not be able to adapt.”

The more we understand this transition, the better prepared we’ll be to deal with it.  And for that we’ll need better and more complete data.

For McEntarfer, the former commissioner of the BLS, the real question is the speed of any disruption. “If it happens at the normal pace of technological change, labor markets will have time to adapt. If there is a sudden and severe disruption, then that will be a big challenge for policymakers,” she says. “That’s really the most important question facing us right now: how rapid this transformation is going to be.” And, she adds, “we’ll know by watching the data.”

Two decades ago, the country was caught flat-footed by the so-called China shock as free-trade policies led to an influx of imports and the devastation of manufacturing jobs in many parts of the country. It took years for researchers to understand the data showing how the trade policies, generally welcomed by economists, were destroying communities. Today the threat of an economic transformation brought on by AI is far larger and points to potentially far more damage for huge groups of workers.

To head off another devastating labor transition, we will need well-timed government and business policies, especially programs to train and reskill workers. If McEntarfer and other labor economists are correct, we probably have time to design deliberate and effective strategies to manage the transition. But first we need to better understand what is going on—and how fast.

It’s hard to find an economist who is more enthusiastic about AI’s future than Stanford’s Brynjolfsson, who believes that we’re likely on the brink of a huge boost that will transform the economy. “Perhaps the best productivity growth of my lifetime is coming up,” he says.

But Brynjolfsson also warns that a lack of data is severely limiting our visibility into the economic and societal impacts that are coming. At a time when hundreds of billions are being spent on rolling out the technology, he says, “we’re not investing even 1% of that on understanding the transition.”

Artificial intelligence has not so far produced a clean story of mass unemployment. Aggregate employment in developed countries remains broadly stable, and recent assessments have found limited evidence that AI has shifted the headline numbers. But a troubling change may be hiding beneath the surface: the quiet weakening of the first rung of the career ladder.

The most worrisome evidence is showing up exactly where we should expect it first: in early-career hiring. A working paper released in November 2025 by the Stanford Digital Economy Lab found that workers aged 22 to 25 in the most AI-exposed occupations experienced a 16% relative decline in employment after the spread of generative AI, even after controlling for other factors that might affect firms’ employment decisions. An Anthropic report from March 2026 provides suggestive evidence that led to a similar conclusion.

More experienced workers in those same occupations did not suffer the same decline. Employment is not also declining in the entry-level jobs with low AI exposure. The concern is specific to early-career jobs that are exposed to AI.

That is not a minor signal. It suggests that firms may be using AI to substitute for the junior tasks through which people traditionally gain their first foothold—at least for those in jobs where generative AI is used extensively, like software developers, customer service representatives, computer programmers, and information systems managers.

The time is now to make changes in the way we train, prepare, and support young people who are about to enter the workforce. Educational institutions need to reorient for the era of an AI-augmented workforce. Governments must incentivize businesses to hire and train early-career workers. Businesses, in turn, need to recognize the importance of developing a long-term workforce experienced in AI—a process that begins with entry-level workers. And students themselves should take on the responsibility of not only becoming AI fluent but learning how to apply that knowledge in various fields.

In short, we must change the way we have traditionally thought of entry-level work.

This is especially true because the broader labor market for recent graduates is also softening. The Federal Reserve Bank of New York reported that in the fourth quarter of 2025, the unemployment rate for recent college graduates rose to 5.6%, while the underemployment rate (the share of graduates working in jobs that typically do not require a college degree) reached 42.5%, its highest level since the covid pandemic. No single statistic can prove that AI is the sole cause of that deterioration. Hiring in general is way down post-pandemic, and young people are particularly vulnerable to the slowdown. But it would be a mistake to ignore the possibility that AI is accelerating an already difficult transition from school to work.

Behind these statistics is a great deal of personal distress. Recent graduates today often submit hundreds of applications before they receive a single offer, and surveys consistently find elevated rates of anxiety, financial precarity, and burnout among young workers in extended job searches. If AI quietly closes the door on typical early jobs, people will pay the price in delayed independence, postponed family formation, and the sense that their first serious professional efforts have been refused.

It also matters because entry-level jobs are part of the economy’s training system. Junior analysts learn which numbers can be trusted. Young software developers learn how production systems fail. New marketers learn how customers behave outside the neat language of dashboards. Early-career legal and financial staff learn how rules, judgment, deadlines, and human relationships actually interact. If AI absorbs more of the drafting, triage, coding, summarizing, and administrative preparation that once helped train entry-level workers, firms may become more efficient in the short run while society becomes less capable in the longer run.

The right way to improve the skills of young workers is not to tell them, “Learn to code.” That advice, which shaped more than a decade of federal initiatives and university expansion, rested on the premise that coding was a stable, scalable skill almost anyone could learn and parlay into a middle-class job. The premise no longer holds. The layer of work AI handles well—translating a specification into routine code, reproducing standard patterns, debugging predictable errors—is precisely the layer that “learn to code” programs were built around.

Supervising AI systems in their work is now a much more relevant skill. So understanding the outputs AI systems produce will become very important.

To help people develop such skills, we should require universities, community colleges, and professional programs to embed AI literacy, data literacy, prompt-based workflow skills, verification skills, and domain judgment into ordinary degrees. Every graduate should know how to use AI tools, check their output, understand their limits, and combine them with human expertise. This matters even for graduates entering occupations that look relatively safe from AI, such as those in health care. Almost every job contains tasks—drafting, summarizing, scheduling, research, basic data work, routine communication—for which AI is already a substantial productivity tool.

The competition most young workers will experience is not human versus machine but colleague versus AI-augmented colleague. For most young workers, the realistic path to making themselves valuable is not to avoid AI but to become fluent in the technology and combine that with domain judgment, contextual reasoning, and human relationship skills. To this end, schools should emphasize paid co-ops, apprenticeships, and employer-linked projects so students build judgment in real workplaces before they graduate.

Governments should also create targeted tax credits, wage subsidies, and training grants for employers that hire early-career workers into structured, AI-augmented roles. The architecture for this kind of conditional, behavior-linked subsidy already exists in US tax policy. What is missing is a version of these instruments built specifically around early-career AI-augmented work.

Firms, for their part, should stop making hiring decisions based only on short-run cost savings from AI. Young workers are not valuable only for the tasks they perform this quarter. Their value lies in learning, skill formation, institutional memory, and future productivity. Entry-level hiring is not just an expense. It is an investment in the future stock of judgment inside the firm. The most effective AI-augmented senior workforce of the late 2030s will be drawn overwhelmingly from the junior cohort of today. Firms that automate away the learning stage may improve their immediate margins but find themselves, a decade from now, without anyone who understands how their own AI-driven workflows actually behave.

Students graduating this spring and next face a tough labor market in transition. AI fluency is becoming a commodity. Domain expertise without AI fluency is being outpaced. The combination is what is genuinely scarce. The mechanical engineer with knowledge of manufacturing and AI proficiency; the software programmer with knowledge of financial services who is also a whiz at AI—these are the types of people who will be in demand.

Georgios Petropoulos is an assistant professor at the USC Marshall School of Business. His research focuses on the implications of information technologies for innovation, competition policy, and labor markets.

During Tuesday’s Google I/O keynote, Demis Hassabis, the CEO of Google DeepMind, proclaimed that we are currently “standing in the foothills of the singularity.” It was a striking statement—the singularity is the theoretical future moment when AI rapidly exceeds human intelligence and dramatically transforms the world. But what struck me as I listened in the audience was the context in which he said those words. 

He was on stage to close out the session with a segment on scientific AI, the centerpiece of which was a video detailing how the company’s weather prediction software provided an advance alert about Hurricane Melissa’s catastrophic landfall in Jamaica last year—and potentially saved lives. If that software, called WeatherNext, helped anyone escape the storm or better fortify their home, that’s an enormous and meaningful achievement. But it’s hardly evidence of an impending singularity.

The juxtaposition of Hassabis’ lofty rhetoric with the real-world results of WeatherNext highlighted the tension between two very different approaches to AI for science. The first focuses on AI tools, like WeatherNext, that are designed and trained to solve specific scientific problems. The second is agentic, LLM-based systems that could one day execute cutting-edge research projects without human involvement.

This second vision powers a great deal of AI enthusiasm right now, including recent excitement around recursive self-improvement, or the idea that AI systems could eventually become the primary drivers of AI advancement—a process that would get faster and faster as the AI systems grow smarter. And agentic systems are now making real research contributions, sometimes with limited human guidance.

Just this week, Pushmeet Kohli, Google Cloud’s chief scientist, published a piece in a special AI and science issue of the journal Daedalus, writing: “We are moving toward AI that doesn’t just facilitate science but begins to do science.” With autonomous AI scientists on the horizon, it’s harder to justify massive efforts to develop super-specialized tools—even one like AlphaFold, for which DeepMind scientists won a Nobel Prize, or a potentially life-saving system like WeatherNext. It also heralds a far stranger future for science, in which humans and AI systems collaborate as peers—or AI even makes scientific progress on its own.

To be clear, Google does not appear to be abandoning its work on specialized AI for science tools. AlphaGenome and AlphaEarth Foundations, which are trained for genetics and Earth science applications respectively, were released last summer, and the newest version of WeatherNext came out in November.

What’s more, such tools remain extremely popular among scientists. Last year, for instance, Google reported that protein structure predictions from AlphaFold have been used by over three million researchers worldwide. And Isomorphic Labs, a Google subsidiary that aims to use AlphaFold and related technologies to develop new drugs, just raised a $2 billion Series B funding round.

But there are concrete signs of realignment, in both enthusiasm and resources. Last month, the Los Angeles Times reported that Google fellow John Jumper, who won the Nobel for AlphaFold, is now working on AI coding, not on science-specific AI tools. It’s not surprising that Google is assigning its best minds to the coding problem, as the company has recently taken a reputational hit because its coding tools don’t currently stand up to those offered by Anthropic and OpenAI. But it may also signal a prioritization of agentic science on Google’s part, as coding abilities are key to the success of some of those systems. 

Across the industry, agentic researcher systems are showing real potential. This week, OpenAI announced that one of their models had disproved an important mathematics conjecture—perhaps the most meaningful contribution that generative AI has made to mathematics so far, according to some mathematicians.

Importantly, the model used by OpenAI is not specialized for solving mathematical problems, or even for research; according to the company, it’s a general-purpose reasoning model in the vein of GPT-5.5. If general agents can make independent contributions to mathematical research, they might soon be able to do the same in science (though the fact that ideas in science must be verified experimentally makes it a tougher domain for AI).

Google is certainly devoting a lot of attention toward an agent-driven scientific future. The big scientific announcement at I/O was the new Gemini for Science package, which unites several of the company’s LLM-based scientific systems under one brand.

This includes the hypothesis-generating AI Co-Scientist and algorithm-optimizing AlphaEvolve, which are still not publicly available—but as Google is now allowing any researcher to apply for access to Gemini for Science, they may soon see wider adoption in the scientific community. Scientists who were involved in early testing are enthusiastic about their potential: Gary Peltz, a Stanford geneticist, compared using the AI Co-Scientist to “consulting the oracle of Delphi” in a Nature Medicine article.

Gemini for Science isn’t incompatible with specialized tools; to the contrary, agentic systems can be designed to call on such tools when they might be useful. And no agentic system can predict the structure that a protein will fold into without AlphaFold’s help (at least not yet). But the company seems to be shifting its public image—and at least some resources and personnel, such as Jumper—away from specifically developing those kinds of tools. Though it has only been five years since AlphaFold solved the protein-folding problem, both the technology and the discourse have quickly moved beyond that once-revolutionary achievement.

Google has been careful to position this new set of scientific agents as an accelerant for human scientists, rather than a replacement for them—the choice of the name AI Co-Scientist as opposed to AI Scientist, for instance, appears quite deliberate. Hassabis uses that same human-centric framing when he talks about changes in the landscape of scientific AI. “For the next decade or so, we should think about AI as this amazing tool to help scientists,” Hassabis said in an interview published in the Daedalus issue. “Beyond that timeframe, it is hard to say with any certainty, but perhaps these systems will become more like collaborators.”

But no one can be an effective scientific collaborator without also being a skilled scientist in their own right. And if Hassabis is anywhere near the mark when he talks about the “foothills of the singularity,” then AI scientists could eventually exceed the capabilities of their human counterparts.

In a discussion with the journalist Mike Allen at I/O, Hassabis spoke of how he was initially inspired to pursue AI when he observed how progress in physics had stagnated since the 1970s; he wondered whether the human mind had reached its limits in that domain, and if AI could help to overcome that barrier. Superhuman agentic scientists would certainly fit that bill. We might not ever get anywhere near there, but Google seems to be aiming itself toward that summit.