There are tears in the audience as Patrick Darling’s song begins to play. It’s a heartfelt song written for his great-grandfather, whom he never got the chance to meet. But this performance is emotional for another reason: It’s Darling’s first time on stage with his bandmates since he lost the ability to sing two years ago.

The 32-year-old musician was diagnosed with amyotrophic lateral sclerosis (ALS) when he was 29 years old. Like other types of motor neuron disease (MND), it affects nerves that supply the body’s muscles. People with ALS eventually lose the ability to control their muscles, including those that allow them to move, speak, and breathe.

Darling’s last stage performance was over two years ago. By that point, he had already lost the ability to stand and play his instruments and was struggling to sing or speak. But recently, he was able to re-create his lost voice using an AI tool trained on snippets of old audio recordings. Another AI tool has enabled him to use this “voice clone” to compose new songs. Darling is able to make music again.

“Sadly, I have lost the ability to sing and play my instruments,” Darling said on stage at the event, which took place in London on Wednesday, using his voice clone. “Despite this, most of my time these days is spent still continuing to compose and produce my music. Doing so feels more important than ever to me now.”

Losing a voice

Darling says he’s been a musician and a composer since he was around 14 years old. “I learned to play bass guitar, acoustic guitar, piano, melodica, mandolin, and tenor banjo,” he said at the event. “My biggest love, though, was singing.”

He met bandmate Nick Cocking over 10 years ago, while he was still a university student, says Cocking. Darling joined Cocking’s Irish folk outfit, the Ceili House Band, shortly afterwards, and their first gig together was in April 2014. Darling, who joined the band as a singer and guitarist, “elevated the musicianship of the band,” says Cocking.

But a few years ago, Cocking and his other bandmates started noticing changes in Darling. He became clumsy, says Cocking. He recalls one night when the band had to walk across the city of Cardiff in the rain: “He just kept slipping and falling, tripping on paving slabs and things like that.” 

He didn’t think too much of it at the time, but Darling’s symptoms continued to worsen. The disease affected his legs first, and in August 2023, he started needing to sit during performances. Then he started to lose the use of his hands. “Eventually he couldn’t play the guitar or the banjo anymore,” says Cocking.

By April 2024, Darling was struggling to talk and breathe at the same time, says Cocking. For that performance, the band carried Darling on stage. “He called me the day after and said he couldn’t do it anymore,” Cocking says, his voice breaking. “By June 2024, it was done.” It was the last time the band played together.

Re-creating a voice

Darling was put in touch with a speech therapist, who raised the possibility of “banking” his voice. People who are losing the ability to speak can opt to record themselves speaking and use those recordings to create speech sounds that can then be activated with typed text, whether by hand or perhaps using a device controlled by eye movements.

Some users have found these tools to be robotic sounding. But Darling had another issue. “By that stage, my voice had already changed,” he said at the event. “It felt like we were saving the wrong voice.”

Then another speech therapist introduced him to a different technology. Richard Cave is a speech and language therapist and a researcher at University College London. He is also a consultant for ElevenLabs, an AI company that develops agents and audio, speech, video, and music tools. One of these tools can create “voice clones”—realistic mimics of real voices that can be generated from minutes, or even seconds, of a person’s recorded voice.

Last year, ElevenLabs launched an impact program with a promise to provide free licenses to these tools for people who have lost their voices to ALS or other diseases, like head and neck cancer or stroke. 

The tool is already helping some of those users. “We’re not really improving how quickly they’re able to communicate, or all of the difficulties that individuals with MND are going through physically, with eating and breathing,” says Gabi Leibowitz, a speech therapist who leads the program. “But what we are doing is giving them a way … to create again, to thrive.” Users are able to stay in their jobs longer and “continue to do the things that make them feel like human beings,” she says.

Cave worked with Darling to use the tool to re-create his lost speaking voice from older recordings.

“The first time I heard the voice, I thought it was amazing,” Darling said at the event, using the voice clone. “It sounded exactly like I had before, and you literally wouldn’t be able to tell the difference,” he said. “I will not say what the first word I made my new voice say, but I can tell you that it began with ‘f’ and ended in ‘k.’”

Re-creating his singing voice wasn’t as easy. The tool typically requires around 10 minutes of clear audio to generate a clone. “I had no high-quality recordings of myself singing,” Darling said. “We had to use audio from videos on people’s phones, shot in noisy pubs, and a couple of recordings of me singing in my kitchen.” Still, those snippets were enough to create a “synthetic version of [Darling’s] singing voice,” says Cave.

In the recordings, Darling sounded a little raspy and “was a bit off” on some of the notes, says Cave. The voice clone has the same qualities. It doesn’t sound perfect, Cave says—it sounds human.

“The ElevenLabs voice that we’ve created is wonderful,” Darling said at the event. “It definitely sounds like me—[it] just kind of feels like a different version of me.”

ElevenLabs has also developed an AI music generator called Eleven Music. The tool allows users to compose tracks, using text prompts to choose the musical style. Several well-known artists have also partnered with the company to license AI clones of their voices, including the actor Michael Caine, whose voice clone is being used to narrate an upcoming ElevenLabs documentary. Last month, the company released an album of 11 tracks created using the tool. “The Liza Minnelli track is really a banger,” says Cave.

Eleven Music can generate a song in a minute, but Darling and Cave spent around six weeks fine-tuning Darling’s song. Using text prompts, any user can “create music and add lyrics in any style [they like],” says Cave. Darling likes Irish folk, but Cave has also worked with a man in Colombia who is creating Colombian folk music. (The ElevenLabs tool is currently available in 74 languages.)

Back on stage

Last month, Cocking got a call from Cave, who sent him Darling’s completed track. “I heard the first two or three words he sang, and I had to turn it off,” he says. “I was just in bits, in tears. It took me a good half a dozen times to make it to the end of the track.”

Darling and Cave were making plans to perform the track live at the ElevenLabs summit in London on Wednesday, February 11. So Cocking and bandmate Hari Ma each arranged accompanying parts to play on the mandolin and fiddle. They had a couple of weeks to rehearse before they joined Darling on stage, two years after their last performance together.

“I wheeled him out on stage, and neither of us could believe it was happening,” says Cave. “He was thrilled.” The song was played as Darling remained on stage, and Cocking and Ma played their instruments live.

Cocking and Cave say Darling plans to continue to use the tools to make music. Cocking says he hopes to perform with Darling again but acknowledges that, given the nature of ALS, it is difficult to make long-term plans.

“It’s so bittersweet,” says Cocking. “But getting up on stage and seeing Patrick there filled me with absolute joy. I know Patrick really enjoyed it as well. We’ve been talking about it … He was really, really proud.”

ELEVENLABS/AMPLIFY

This is today’s edition of The Download, our weekday newsletter that provides a daily dose of what’s going on in the world of technology.

US deputy health secretary: Vaccine guidelines are still subject to change

Over the past year, Jim O’Neill has become one of the most powerful people in public health. As the US deputy health secretary, he holds two roles at the top of the country’s federal health and science agencies. He oversees a department with a budget of over a trillion dollars. And he signed the decision memorandum on the US’s deeply controversial new vaccine schedule.

He’s also a longevity enthusiast. In an exclusive interview with MIT Technology Review earlier this month, O’Neill described his plans to increase human healthspan through longevity-focused research supported by ARPA-H, a federal agency dedicated to biomedical breakthroughs. Fellow longevity enthusiasts said they hope he will bring attention and funding to their cause.

At the same time, O’Neill defended reducing the number of broadly recommended childhood vaccines, a move that has been widely criticized by experts in medicine and public health. Read the full story.

—Jessica Hamzelou

The myth of the high-tech heist

Making a movie is a lot like pulling off a heist. That’s what Steven Soderbergh—director of the Ocean’s franchise, among other heist-y classics—said a few years ago. You come up with a creative angle, put together a team of specialists, figure out how to beat the technological challenges, rehearse, move with Swiss-watch precision, and—if you do it right—redistribute some wealth.

But conversely, pulling off a heist isn’t much like the movies. Surveillance cameras, computer-controlled alarms, knockout gas, and lasers hardly ever feature in big-ticket crime. In reality, technical countermeasures are rarely a problem, and high-tech gadgets are rarely a solution. Read the full story.

—Adam Rogers

This story is from the next print issue of MIT Technology Review magazine, which is all about crime. If you haven’t already, subscribe now to receive future issues once they land.

 RFK Jr. follows a carnivore diet. That doesn’t mean you should.

Americans have a new set of diet guidelines. Robert F. Kennedy Jr. has taken an old-fashioned food pyramid, turned it upside down, and plonked a steak and a stick of butter in prime positions.

Kennedy and his Make America Healthy Again mates have long been extolling the virtues of meat and whole-fat dairy, so it wasn’t too surprising to see those foods recommended alongside vegetables and whole grains (despite the well-established fact that too much saturated fat can be extremely bad for you).

Some influencers have taken the meat trend to extremes, following a “carnivore diet.” A recent review of research into nutrition misinformation on social media found that a lot of shared diet information is nonsense. But what’s new is that some of this misinformation comes from the people who now lead America’s federal health agencies. Read the full story.

—Jessica Hamzelou

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.

The must-reads

I’ve combed the internet to find you today’s most fun/important/scary/fascinating stories about technology.

1 The Trump administration has revoked a landmark climate ruling
In its absence, it can erase the limits that restrict planet-warming emissions. (WP $)
+ Environmentalists and Democrats have vowed to fight the reversal. (Politico)
+ They’re seriously worried about how it will affect public health. (The Hill)

2 An unexplained wave of bot traffic is sweeping the web
Sites across the world are witnessing automated traffic that appears to originate from China. (Wired $)

3 Amazon’s Ring has axed its partnership with Flock
Law enforcement will no longer be able to request Ring doorbell footage from its users. (The Verge)
+ Ring’s recent TV ad for a dog-finding feature riled viewers. (WSJ $)
+ How Amazon Ring uses domestic violence to market doorbell cameras. (MIT Technology Review)

4 Americans are taking the hit for almost all of Trump’s tariffs
Consumers and companies in the US, not overseas, are shouldering 90% of levies. (Reuters)
+ Trump has long insisted that his tariffs costs will be borne by foreign exporters. (FT $)
+ Sweeping tariffs could threaten the US manufacturing rebound. (MIT Technology Review)

5 Meta and Snap say Australia’s social media ban hasn’t affected business
They’re still making plenty of money amid the country’s decision to ban under-16s from the platforms. (Bloomberg $)
+ Does preventing teens from going online actually do any good? (Economist $)

6 AI workers are selling their shares before their firms go public
Cashing out early used to be a major Silicon Valley taboo. (WSJ $)

7 Elon Musk posted about race almost every day last month
His fixation on a white racial majority appears to be intensifying. (The Guardian)
+ Race is a recurring theme in the Epstein emails, too. (The Atlantic $)

8 The man behind a viral warning about AI used AI to write it
But he stands behind its content.. (NY Mag $)
+ How AI-generated text is poisoning the internet. (MIT Technology Review)

9 Influencers are embracing Chinese traditions ahead of the New Year
On the internet, no one knows you’re actually from Wisconsin. (NYT $)

10 Australia’s farmers are using AI to count sheep
No word on whether it’s helping them sleep easier, though. (FT $)

Quote of the day

“Ignoring warning signs will not stop the storm. It puts more Americans directly in its path.”

—Former US secretary of state John Kerry takes aim at the US government’s decision to repeal the key rule that allows it to regulate climate-heating pollution, the Guardian reports.

One more thing

The Vera C. Rubin Observatory is ready to transform our understanding of the cosmos

High atop Chile’s 2,700-meter Cerro Pachón, the air is clear and dry, leaving few clouds to block the beautiful view of the stars. It’s here that the Vera C. Rubin Observatory will soon use a car-size 3,200-megapixel digital camera—the largest ever built—to produce a new map of the entire night sky every three days.

Findings from the observatory will help tease apart fundamental mysteries like the nature of dark matter and dark energy, two phenomena that have not been directly observed but affect how objects are bound together—and pushed apart.

A quarter-­century in the making, the observatory is poised to expand our understanding of just about every corner of the universe. Read the full story.

—Adam Mann

We can still have nice things

A place for comfort, fun and distraction to brighten up your day. (Got any ideas? Drop me a line or skeet ’em at me.)

+ Why 2026 is shaping up to be the year of the pop comeback.
+ Almost everything we thought we knew about Central America’s Maya has turned out to be completely wrong.
+ The Bigfoot hunters have spoken!
+ This fun game puts you in the shoes of a distracted man trying to participate in a date while playing on a GameBoy.

Making a movie is a lot like pulling off a heist. That’s what Steven Soderbergh—director of the Ocean’s franchise, among other heist-y classics—said a few years ago. You come up with a creative angle, put together a team of specialists, figure out how to beat the technological challenges, rehearse, move with Swiss-watch precision, and—if you do it right—redistribute some wealth. That could describe either the plot or the making of Ocean’s Eleven.

But conversely, pulling off a heist isn’t much like the movies. Surveillance cameras, computer-controlled alarms, knockout gas, and lasers hardly ever feature in big-ticket crime. In reality, technical countermeasures are rarely a problem, and high-tech gadgets are rarely a solution. The main barrier to entry is usually a literal barrier to entry, like a door. Thieves’ most common move is to collude with, trick, or threaten an insider. Last year a heist cost the Louvre €88 million worth of antique jewelry, and the most sophisticated technology in play was an angle grinder.

The low-tech Louvre maneuvers were in keeping with what heist research long ago concluded. In 2014 US nuclear weapons researchers at Sandia National Laboratories took a detour into this demimonde, producing a 100-page report called “The Perfect Heist: Recipes from Around the World.” The scientists were worried someone might try to steal a nuke from the US arsenal, and so they compiled information on 23 high-value robberies from 1972 to 2012 into a “Heist Methods and Characteristics Database,” a critical mass of knowledge on what worked. Thieves, they found, dedicated huge amounts of money and time to planning and practice runs—sometimes more than 100. They’d use brute force, tunneling through sewers for months (Société Générale bank heist, Nice, France, 1976), or guile, donning police costumes to fool guards (Gardner Museum, Boston, 1990). But nobody was using, say, electromagnetic pulse generators to shut down the Las Vegas electrical grid. The most successful robbers got to the valuable stuff unseen and got out fast.

Advance the time frame, and the situation looks much the same. Last year, Spanish researchers looking at art crimes from 1990 to 2022 found that the least technical methods are still the most successful. “High-tech technology doesn’t work so well,” says Erin L. Thompson, an art historian at John Jay College of Justice who studies art crime. Speed and practice trump complicated systems and alarms; even that Louvre robbery was, at heart, just a minutes-long smash-and-grab.

An emphasis on speed doesn’t mean heists don’t require skill—panache, even. As the old saying goes, amateurs talk strategy; professionals study logistics. Even without gadgets, heists and heist movies still revel in an engineer’s mindset. “Heist movies absolutely celebrate deep-dive nerdery—‘I’m going to know everything I can about the power grid, about this kind of stone and drill, about Chicago at night,’” says Anna Kornbluh, a professor of English at the University of Illinois at Chicago. She published a paper last October on the ways heist movies reflect an Old Hollywood approach to collective art-making, while shows about new grift, like those detailing the rise and fall of WeWork or the con artist Anna Delvey, reflect the more lone-wolf, disrupt-and-grow mindset of the streaming era. 

Her work might help explain why law-abiding citizens might cheer for the kinds of guys who’d steal a crown from the Louvre, or $100,000 worth of escargot from a farm in Champagne (as happened just a few weeks later). Heists, says Kornbluh, are anti-oligarch praxis. “Everybody wants to know how to be in a competent collective. Everybody wants there to be better logistics,” she says. “We need a better state. We need a better society. We need a better world.” Those are shared values—and as another old saying tells us, where there is value, there is crime.

Following publication of this story, Politico reported Jim O’Neill would be leaving his current roles within the Department of Health and Human Services.

Over the past year, Jim O’Neill has become one of the most powerful people in public health. As the US deputy health secretary, he holds two roles at the top of the country’s federal health and science agencies. He oversees a department with a budget of over a trillion dollars. And he signed the decision memorandum on the US’s deeply controversial new vaccine schedule.

He’s also a longevity enthusiast. In an exclusive interview with MIT Technology Review earlier this month, O’Neill described his plans to increase human healthspan through longevity-focused research supported by ARPA-H, a federal agency dedicated to biomedical breakthroughs. At the same time, he defended reducing the number of broadly recommended childhood vaccines, a move that has been widely criticized by experts in medicine and public health. 

In MIT Technology Review’s profile of O’Neill last year, people working in health policy and consumer advocacy said they found his libertarian views on drug regulation “worrisome” and “antithetical to basic public health.” 

He was later named acting director of the Centers for Disease Control and Prevention, putting him in charge of the nation’s public health agency.

But fellow longevity enthusiasts said they hope O’Neill will bring attention and funding to their cause: the search for treatments that might slow, prevent, or even reverse human aging. Here are some takeaways from the interview. 

Vaccine recommendations could change further

Last month, the US cut the number of vaccines recommended for children. The CDC no longer recommends vaccinations against flu, rotavirus, hepatitis A, or meningococcal disease for all children. The move was widely panned by medical groups and public health experts. Many worry it will become more difficult for children to access those vaccines. The majority of states have rejected the recommendations. 

In the confirmation hearing for his role as deputy secretary of health and human services, which took place in May last year, O’Neill said he supported the CDC’s vaccine schedule. MIT Technology Review asked him if that was the case and, if so, what made him change his mind. “Researching and examining and reviewing safety data and efficacy data about vaccines is one of CDC’s obligations,” he said. “CDC gives important advice about vaccines and should always be open to new data and new ways of looking at data.”

At the beginning of December, O’Neill said, President Donald Trump “asked me to look at what other countries were doing in terms of their vaccine schedules.” He said he spoke to health ministries of other countries and consulted with scientists at the CDC and FDA. “It was suggested to me by lots of the operating divisions that the US focus its recommendations on consensus vaccines of other developed nations—in other words, the most important vaccines that are most often part of the core recommendations of other countries,” he said.

“As a result of that, we did an update to the vaccine schedule to focus on a set of vaccines that are most important for all children.” 

But some experts in public health have said that countries like Denmark and Japan, whose vaccine schedules the new US one was supposedly modeled on, are not really comparable to the US. When asked about these criticisms, O’Neill replied, “A lot of parents feel that … more than 70 vaccine doses given to young children sounds like a really high number, and some of them ask which ones are the most important. I think we helped answer that question in a way that didn’t remove anyone’s access.”

A few weeks after the vaccine recommendations were changed, Kirk Milhoan, who leads the CDC’s Advisory Committee on Immunization Practices, said that vaccinations for measles and polio—which are currently required for entry to public schools—should be optional. (Mehmet Oz, the Center for Medicare and Medicaid Services director, has more recently urged people to “take the [measles] vaccine.”)

“CDC still recommends that all children are vaccinated against diphtheria, tetanus, whooping cough, Haemophilus influenzae type b (Hib), Pneumococcal conjugate, polio, measles, mumps, rubella, and human papillomavirus (HPV), for which there is international consensus, as well as varicella (chickenpox),” he said when asked for his thoughts on this comment.

He also said that current vaccine guidelines are “still subject to new data coming in, new ways of thinking about things.” “CDC, FDA, and NIH are initiating new studies of the safety of immunizations,” he added. “We will continue to ask the Advisory Committee on Immunization Practices to review evidence and make updated recommendations with rigorous science and transparency.”

More support for longevity—but not all science

O’Neill said he wants longevity to become a priority for US health agencies. His ultimate goal, he said, is to “make the damage of aging something that’s under medical control.” It’s “the same way of thinking” as the broader Make America Healthy Again approach, he said: “‘Again’ implies restoration of health, which is what longevity research and therapy is all about.” 

O’Neill said his interest in longevity was ignited by his friend Peter Thiel, the billionaire tech entrepreneur, around 2008 to 2009. It was right around the time O’Neill was finishing up a previous role in HHS, under the Bush administration. O’Neill said Thiel told him he “should really start looking into longevity and the idea that aging damage could be reversible.” “I just got more and more excited about that idea,” he said.

When asked if he’s heard of Vitalism, a philosophical movement for “hardcore” longevity enthusiasts who, broadly, believe that death is wrong, O’Neill replied: “Yes.” 

The Vitalist declaration lists five core statements, including “Death is humanity’s core problem,” “Obviating aging is scientifically plausible,” and “I will carry the message against aging and death.” O’Neill said he agrees with all of them. “I suppose I am [a Vitalist],” he said with a smile, although he’s not a paying member of the foundation behind it.

As deputy secretary of the Department of Health and Human Services, O’Neill assumes a level of responsibility for huge and influential science and health agencies, including the National Institutes of Health (the world’s largest public funder of biomedical research) and the Food and Drug Administration (which oversees drug regulation and is globally influential) as well as the CDC.

Today, he said, he sees support for longevity science from his colleagues within HHS. “If I could describe one common theme to the senior leadership at HHS, obviously it’s to make America healthy again, and reversing aging damage is all about making people healthy again,” he said. “We are refocusing HHS on addressing and reversing chronic disease, and chronic diseases are what drive aging, broadly.”

Over the last year, thousands of NIH grants worth over $2 billion were frozen or terminated, including funds for research on cancer biology, health disparities, neuroscience, and much more. When asked whether any of that funding will be restored, he did not directly address the question, instead noting: “You’ll see a lot of funding more focused on important priorities that actually improve people’s health.”

Watch ARPA-H for news on organ replacements and more

He promised we’ll hear more from ARPA-H, the three-year-old federal agency dedicated to achieving breakthroughs in medical science and biotechnology. It was established with the official goal of promoting “high-risk, high-reward innovation for the development and translation of transformative health technologies.”

O’Neill said that “ARPA-H exists to make the impossible possible in health and medicine.” The agency has a new director—Alicia Jackson, who formerly founded and led a company focused on women’s health and longevity, took on the role in October last year.

O’Neill said he helped recruit Jackson, and that she was hired in part because of her interest in longevity, which will now become a major focus of the agency. He said he meets with her regularly, as well as with Andrew Brack and Jean Hébert, two other longevity supporters who lead departments at ARPA-H. Brack’s program focuses on finding biological markers of aging. Hebert’s aim is to find a way to replace aging brain tissue, bit by bit.  

O’Neill is especially excited by that one, he said. “I would try it … Not today, but … if progress goes in a broadly good direction, I would be open to it. We’re hoping to see significant results in the next few years.”

He’s also enthused by the idea of creating all-new organs for transplantation. “Someday we want to be able to grow new organs, ideally from the patients’ own cells,” O’Neill said. An ARPA-H program will receive $170 million over five years to that end, he adds. “I’m very excited about the potential of ARPA-H and Alicia and Jean and Andrew to really push things forward.”

Longevity lobbyists have a friendly ear

O’Neill said he also regularly talks to the team at the lobbying group Alliance for Longevity Initiatives. The organization, led by Dylan Livingston, played an instrumental role in changing state law in Montana to make experimental therapies more accessible. O’Neill said he hasn’t formally worked with them but thinks that “they’re doing really good work on raising awareness, including on Capitol Hill.”

Livingston has told me that A4LI’s main goals center around increasing support for aging research (possibly via the creation of a new NIH institute entirely dedicated to the subject) and changing laws to make it easier and cheaper to develop and access potential anti-aging therapies.

O’Neill gave the impression that the first goal might be a little overambitious—the number of institutes is down to Congress, he said. “I would like to get really all of the institutes at NIH to think more carefully about how many chronic diseases are usefully thought of as pathologies of aging damage,” he said. There’ll be more federal funding for that research, he said, although he won’t say more for now.

Some members of the longevity community have more radical ideas when it comes to regulation: they want to create their own jurisdictions designed to fast-track the development of longevity drugs and potentially encourage biohacking and self-experimentation. 

It’s a concept that O’Neill has expressed support for in the past. He has posted on X about his support for limiting the role of government, and in support of building “freedom cities”—a similar concept that involves creating new cities on federal land. 

Another longevity enthusiast who supports the concept is Niklas Anzinger, a German tech entrepreneur who is now based in Próspera, a private city within a Honduran “special economic zone,” where residents can make their own suggestions for medical regulations. Anzinger also helped draft Montana’s state law on accessing experimental therapies. O’Neill knows Anzinger and said he talks to him “once or twice a year.”

O’Neill has also supported the idea of seasteading—building new “startup countries” at sea. He served on the board of directors of the Seasteading Institute until March 2024.

In 2009, O’Neill told an audience at a Seasteading Institute conference that “the healthiest societies in 2030 will most likely be on the sea.” When asked if he still thinks that’s the case, he said: “It’s not quite 2030, so I think it’s too soon to say … What I would say now is: the healthiest societies are likely to be the ones that encourage innovation the most.”

We might expect more nutrition advice

When it comes to his own personal ambitions for longevity, O’Neill said, he takes a simple approach that involves minimizing sugar and ultraprocessed food, exercising and sleeping well, and supplementing with vitamin D. He also said he tries to “eat a diet that has plenty of protein and saturated fat,” echoing the new dietary guidance issued by the US Departments of Health and Human Services and Agriculture. That guidance has been criticized by nutrition scientists, who point out that it ignores decades of research into the harms of a diet high in saturated fat.

We can expect to see more nutrition-related updates from HHS, said O’Neill: “We’re doing more research, more randomized controlled trials on nutrition. Nutrition is still not a scientifically solved problem.” Saturated fats are of particular interest, he said. He and his colleagues want to identify “the healthiest fats,” he said. 

“Stay tuned.”

Americans have a new set of diet guidelines. Robert F. Kennedy Jr. has taken an old-fashioned food pyramid, turned it upside down, and plonked a steak and a stick of butter in prime positions.

Kennedy and his Make America Healthy Again mates have long been extolling the virtues of meat and whole-fat dairy, so it wasn’t too surprising to see those foods recommended alongside vegetables and whole grains (despite the well-established fact that too much saturated fat can be extremely bad for you).

Some influencers have taken the meat trend to extremes, following a “carnivore diet.” “The best thing you could do is eliminate out everything except fatty meat and lard,” Anthony Chaffee, an MD with almost 400,000 followers, said in an Instagram post.

And I almost choked on my broccoli when, while scrolling LinkedIn, I came across an interview with another doctor declaring that “there is zero scientific evidence to say that vegetables are required in the human diet.” That doctor, who described himself as “90% carnivore,” went on to say that all he’d eaten the previous day was a kilo of beef, and that vegetables have “anti-nutrients,” whatever they might be.

You don’t have to spend much time on social media to come across claims like this. The “traditionalist” influencer, author, and psychologist Jordan Peterson was promoting a meat-only diet as far back as 2018. A recent review of research into nutrition misinformation on social media found that the most diet information is shared on Instagram and YouTube, and that a lot of it is nonsense. So much so that the authors describe it as a “growing public health concern.”

What’s new is that some of this misinformation comes from the people who now lead America’s federal health agencies. In January Kennedy, who leads the Department of Health and Human Services, told a USA Today reporter that he was on a carnivore diet. “I only eat meat or fermented foods,” he said. He went on to say that the diet had helped him lose “40% of [his] visceral fat within a month.”

“Government needs to stop spreading misinformation that natural and saturated fats are bad for you,” Food and Drug Administration commissioner Martin Makary argued in a recent podcast interview. The principles of “whole foods and clean meats” are “biblical,” he said. The interviewer said that Makary’s warnings about pesticides made him want to “avoid all salads and completely miss the organic section in the grocery store.”

For the record: There’s plenty of evidence that a diet high in saturated fat can increase the risk of heart disease. That’s not government misinformation. 

The carnivore doctors’ suggestion to avoid vegetables is wrong too, says Gabby Headrick, associate director of food and nutrition policy at George Washington University’s Institute for Food Safety & Nutrition Security. There’s no evidence to suggest that a meat-only diet is good for you. “All of the nutrition science to date strongly identifies a wide array of vegetables … as being very health-promoting,” she adds.

To be fair to the influencers out there, diet is a tricky thing to study. Much of the research into nutrition relies on volunteers to keep detailed and honest food diaries—something that people are generally quite bad at. And the way our bodies respond to foods might be influenced by our genetics, our microbiomes, the way we prepare or consume those foods, and who knows what else.

Still, it will come as a surprise to no one that there is plenty of what the above study calls “low-quality content” floating around on social media. So it’s worth arming ourselves with a good dose of skepticism, especially when we come across posts that mention “miracle foods” or extreme, limited diets.

The truth is that most food is neither good nor bad when eaten in moderation. Diet trends come and go, and for most people, the best reasonable advice is simply to eat a balanced diet low in sugar, salt, and saturated fat. You know—the basics. No matter what that weird upside-down food pyramid implies. To the carnivore influencers, I say: get your misinformation off my broccoli.

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.

This is today’s edition of The Download, our weekday newsletter that provides a daily dose of what’s going on in the world of technology.

AI is already making online crimes easier. It could get much worse.

Just as software engineers are using artificial intelligence to help write code and check for bugs, hackers are using these tools to reduce the time and effort required to orchestrate an attack, lowering the barriers for less experienced attackers to try something out.

Some in Silicon Valley warn that AI is on the brink of being able to carry out fully automated attacks. But most security researchers instead argue that we should be paying closer attention to the much more immediate risks posed by AI, which is already speeding up and increasing the volume of scams.

Criminals are increasingly exploiting the latest deepfake technologies to impersonate people and swindle victims out of vast sums of money. And we need to be ready for what comes next. Read the full story.

—Rhiannon Williams

This story is from the next print issue of MIT Technology Review magazine, which is all about crime. If you haven’t already, subscribe now to receive future issues once they land.

Is a secure AI assistant possible?

AI agents are a risky business. Even when stuck inside the chatbox window, LLMs will make mistakes and behave badly. Once they have tools that they can use to interact with the outside world, such as web browsers and email addresses, the consequences of those mistakes become far more serious.

Viral AI agent project OpenClaw, which has made headlines across the world in recent weeks, harnesses existing LLMs to let users create their own bespoke assistants. For some users, this means handing over reams of personal data, from years of emails to the contents of their hard drive. That has security experts thoroughly freaked out.

In response to these concerns, its creator warned that nontechnical people should not use the software. But there’s a clear appetite for what OpenClaw is offering, and any AI companies hoping to get in on the personal assistant business will need to figure out how to build a system that will keep users’ data safe and secure. To do so, they’ll need to borrow approaches from the cutting edge of agent security research. Read the full story.

—Grace Huckins

What’s next for Chinese open-source AI

The past year has marked a turning point for Chinese AI. Since DeepSeek released its R1 reasoning model in January 2025, Chinese companies have repeatedly delivered AI models that match the performance of leading Western models at a fraction of the cost.

These models differ in a crucial way from most US models like ChatGPT or Claude, which you pay to access and can’t inspect. The Chinese companies publish their models’ weights—numerical values that get set when a model is trained—so anyone can download, run, study, and modify them. 

If open-source AI models keep getting better, they will not just offer the cheapest options for people who want access to frontier AI capabilities; they will change where innovation happens and who sets the standards. Here’s what may come next.

—Caiwei Chen

This is part of our What’s Next series, which looks across industries, trends, and technologies to give you a first look at the future. You can read the rest of them here.

Why EVs are gaining ground in Africa

EVs are getting cheaper and more common all over the world. But the technology still faces major challenges in some markets, including many countries in Africa.

Some regions across the continent still have limited grid and charging infrastructure, and those that do have widespread electricity access sometimes face reliability issues—a problem for EV owners, who require a stable electricity source to charge up and get around. But there are some signs of progress. Read the full story.

—Casey Crownhart

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

The must-reads

I’ve combed the internet to find you today’s most fun/important/scary/fascinating stories about technology.

1 Instagram’s head has denied that social media is “clinically addictive”  
Adam Mosseri disputed allegations the platform prioritized profits over protecting its younger users’ mental health. (NYT $)
+ Meta researchers’ correspondence seems to suggest otherwise. (The Guardian)

2 The Pentagon is pushing AI companies to drop tools’ restrictions
In a bid to make AI models available on classified networks. (Reuters)
+ The Pentagon has gutted the team that tests AI and weapons systems. (MIT Technology Review)

3 The FTC has warned Apple News not to stifle conservative content
It has accused the company’s news arm of promoting what it calls “leftist outlets.” (FT $)

4 Anthropic has pledged to minimize the impact of its data centers
By covering electricity price increases and the cost of grid infrastructure upgrades. (NBC News)
+ We did the math on AI’s energy footprint. Here’s the story you haven’t heard. (MIT Technology Review)

5 Online harassers are posting Grok-generated nude images on OnlyFans 
Kylie Brewer, a feminism-focused content creator, says the latest online campaign against her feels like an escalation. (404 Media)
+ Inside the marketplace powering bespoke AI deepfakes of real women. (MIT Technology Review)

6 Venture capitalists are hedging their AI bets
They’re breaking a cardinal rule by investing in both OpenAI and rival Anthropic. (Bloomberg $)
+ OpenAI has set itself some seriously lofty revenue goals. (NYT $)
+ AI giants are notoriously inconsistent when reporting deprecation expenses. (WSJ $)

7 We’re learning more about the links between weight loss drugs and addiction
Some patients report lowered urges for drugs and alcohol. But can it last? (New Yorker $)
+ What we still don’t know about weight-loss drugs. (MIT Technology Review)

8 Meta has patented an AI that keeps the accounts of dead users active
But it claims to have “no plans to move forward” with it. (Insider $)
+ Deepfakes of your dead loved ones are a booming Chinese business. (MIT Technology Review)

9 Slime mold is cleverer than you may think
A certain type appears able to learn, remember and make decisions. (Knowable Magazine)
+ And that’s not all—this startup thinks it can help us design better cities, too. (MIT Technology Review)

10 Meditation can actually alter your brain activity
According to a new study conducted on Buddhist monks. (Wired $)

Quote of the day

“I still try to believe that the good that I’m doing is greater than the horrors that are a part of this. But there’s a limit to what we can put up with. And I’ve hit my limit.”

—An anonymous Microsoft worker explains why they’re growing increasingly frustrated with their employer’s links to ICE, the Verge reports. 

One more thing

Motor neuron diseases took their voices. AI is bringing them back.

Jules Rodriguez lost his voice in October 2024. His speech had been deteriorating since a diagnosis of amyotrophic lateral sclerosis (ALS) in 2020, but a tracheostomy to help him breathe dealt the final blow. 

Rodriguez and his wife, Maria Fernandez, who live in Miami, thought they would never hear his voice again. Then they re-created it using AI. After feeding old recordings of Rodriguez’s voice into a tool trained on voices from film, television, radio, and podcasts, the couple were able to generate a voice clone—a way for Jules to communicate in his “old voice.”

Rodriguez is one of over a thousand people with speech difficulties who have cloned their voices using free software from ElevenLabs. The AI voice clones aren’t perfect. But they represent a vast improvement on previous communication technologies and are already improving the lives of people with motor neuron diseases. Read the full story. 

—Jessica Hamzelou

We can still have nice things

A place for comfort, fun and distraction to brighten up your day. (Got any ideas? Drop me a line or skeet ’em at me.)

+ We all know how the age of the dinosaurs ended. But how did it begin?
+ There’s only one Miss Piggy—and her fashion looks through the ages are iconic.
+ Australia’s hospital for injured and orphaned flying foxes is unbearably cute.
+ 81-year old Juan López is a fitness inspiration to us all.

Anton Cherepanov is always on the lookout for something interesting. And in late August last year, he spotted just that. It was a file uploaded to VirusTotal, a site cybersecurity researchers like him use to analyze submissions for potential viruses and other types of malicious software, often known as malware. On the surface it seemed innocuous, but it triggered Cherepanov’s custom malware-detecting measures. Over the next few hours, he and his colleague Peter Strýček inspected the sample and realized they’d never come across anything like it before.

The file contained ransomware, a nasty strain of malware that encrypts the files it comes across on a victim’s system, rendering them unusable until a ransom is paid to the attackers behind it. But what set this example apart was that it employed large language models (LLMs). Not just incidentally, but across every stage of an attack. Once it was installed, it could tap into an LLM to generate customized code in real time, rapidly map a computer to identify sensitive data to copy or encrypt, and write personalized ransom notes based on the files’ content. The software could do this autonomously, without any human intervention. And every time it ran, it would act differently, making it harder to detect.

Cherepanov and Strýček were confident that their discovery, which they dubbed PromptLock, marked a turning point in generative AI, showing how the technology could be exploited to create highly flexible malware attacks. They published a blog post declaring that they’d uncovered the first example of AI-powered ransomware, which quickly became the object of widespread global media attention.

But the threat wasn’t quite as dramatic as it first appeared. The day after the blog post went live, a team of researchers from New York University claimed responsibility, explaining that the malware was not, in fact, a full attack let loose in the wild but a research project, merely designed to prove it was possible to automate each step of a ransomware campaign—which, they said, they had. 

PromptLock may have turned out to be an academic project, but the real bad guys are using the latest AI tools. Just as software engineers are using artificial intelligence to help write code and check for bugs, hackers are using these tools to reduce the time and effort required to orchestrate an attack, lowering the barriers for less experienced attackers to try something out. 

The likelihood that cyberattacks will now become more common and more effective over time is not a remote possibility but “a sheer reality,” says Lorenzo Cavallaro, a professor of computer science at University College London. 

Some in Silicon Valley warn that AI is on the brink of being able to carry out fully automated attacks. But most security researchers say this claim is overblown. “For some reason, everyone is just focused on this malware idea of, like, AI superhackers, which is just absurd,” says Marcus Hutchins, who is principal threat researcher at the security company Expel and famous in the security world for ending a giant global ransomware attack called WannaCry in 2017. 

Instead, experts argue, we should be paying closer attention to the much more immediate risks posed by AI, which is already speeding up and increasing the volume of scams. Criminals are increasingly exploiting the latest deepfake technologies to impersonate people and swindle victims out of vast sums of money. These AI-enhanced cyberattacks are only set to get more frequent and more destructive, and we need to be ready. 

Spam and beyond

Attackers started adopting generative AI tools almost immediately after ChatGPT exploded on the scene at the end of 2022. These efforts began, as you might imagine, with the creation of spam—and a lot of it. Last year, a report from Microsoft said that in the year leading up to April 2025, the company had blocked $4 billion worth of scams and fraudulent transactions, “many likely aided by AI content.” 

At least half of spam email is now generated using LLMs, according to estimates by researchers at Columbia University, the University of Chicago, and Barracuda Networks, who analyzed nearly 500,000 malicious messages collected before and after the launch of ChatGPT. They also found evidence that AI is increasingly being deployed in more sophisticated schemes. They looked at targeted email attacks, which impersonate a trusted figure in order to trick a worker within an organization out of funds or sensitive information. By April 2025, they found, at least 14% of those sorts of focused email attacks were generated using LLMs, up from 7.6% in April 2024.

In one high-profile case, a worker was tricked into transferring $25 million to criminals via a video call with digital versions of the company’s chief financial officer and other employees.

And the generative AI boom has made it easier and cheaper than ever before to generate not only emails but highly convincing images, videos, and audio. The results are much more realistic than even just a few short years ago, and it takes much less data to generate a fake version of someone’s likeness or voice than it used to.

Criminals aren’t deploying these sorts of deepfakes to prank people or to simply mess around—they’re doing it because it works and because they’re making money out of it, says Henry Ajder, a generative AI expert. “If there’s money to be made and people continue to be fooled by it, they’ll continue to do it,” he says. In one high-­profile case reported in 2024, a worker at the British engineering firm Arup was tricked into transferring $25 million to criminals via a video call with digital versions of the company’s chief financial officer and other employees. That’s likely only the tip of the iceberg, and the problem posed by convincing deepfakes is only likely to get worse as the technology improves and is more widely adopted. 

Criminals’ tactics evolve all the time, and as AI’s capabilities improve, such people are constantly probing how those new capabilities can help them gain an advantage over victims. Billy Leonard, tech leader of Google’s Threat Analysis Group, has been keeping a close eye on changes in the use of AI by potential bad actors (a widely used term in the industry for hackers and others attempting to use computers for criminal purposes). In the latter half of 2024, he and his team noticed prospective criminals using tools like Google Gemini the same way everyday users do—to debug code and automate bits and pieces of their work—as well as tasking it with writing the odd phishing email. By 2025, they had progressed to using AI to help create new pieces of malware and release them into the wild, he says.

The big question now is how far this kind of malware can go. Will it ever become capable enough to sneakily infiltrate thousands of companies’ systems and extract millions of dollars, completely undetected? 

Most popular AI models have guardrails in place to prevent them from generating malicious code or illegal material, but bad actors still find ways to work around them. For example, Google observed a China-linked actor asking its Gemini AI model to identify vulnerabilities on a compromised system—a request it initially refused on safety grounds. However, the attacker managed to persuade Gemini to break its own rules by posing as a participant in a capture-the-flag competition, a popular cybersecurity game. This sneaky form of jailbreaking led Gemini to hand over information that could have been used to exploit the system. (Google has since adjusted Gemini to deny these kinds of requests.)

But bad actors aren’t just focusing on trying to bend the AI giants’ models to their nefarious ends. Going forward, they’re increasingly likely to adopt open-source AI models, as it’s easier to strip out their safeguards and get them to do malicious things, says Ashley Jess, a former tactical specialist at the US Department of Justice and now a senior intelligence analyst at the cybersecurity company Intel 471. “Those are the ones I think that [bad] actors are going to adopt, because they can jailbreak them and tailor them to what they need,” she says.

The NYU team used two open-source models from OpenAI in its PromptLock experiment, and the researchers found they didn’t even need to resort to jailbreaking techniques to get the model to do what they wanted. They say that makes attacks much easier. Although these kinds of open-source models are designed with an eye to ethical alignment, meaning that their makers do consider certain goals and values in dictating the way they respond to requests, the models don’t have the same kinds of restrictions as their closed-source counterparts, says Meet Udeshi, a PhD student at New York University who worked on the project. “That is what we were trying to test,” he says. “These LLMs claim that they are ethically aligned—can we still misuse them for these purposes? And the answer turned out to be yes.” 

It’s possible that criminals have already successfully pulled off covert PromptLock-style attacks and we’ve simply never seen any evidence of them, says Udeshi. If that’s the case, attackers could—in theory—have created a fully autonomous hacking system. But to do that they would have had to overcome the significant barrier that is getting AI models to behave reliably, as well as any inbuilt aversion the models have to being used for malicious purposes—all while evading detection. Which is a pretty high bar indeed.

Productivity tools for hackers

So, what do we know for sure? Some of the best data we have now on how people are attempting to use AI for malicious purposes comes from the big AI companies themselves. And their findings certainly sound alarming, at least at first. In November, Leonard’s team at Google released a report that found bad actors were using AI tools (including Google’s Gemini) to dynamically alter malware’s behavior; for example, it could self-modify to evade detection. The team wrote that it ushered in “a new operational phase of AI abuse.”

However, the five malware families the report dug into (including PromptLock) consisted of code that was easily detected and didn’t actually do any harm, the cybersecurity writer Kevin Beaumont pointed out on social media. “There’s nothing in the report to suggest orgs need to deviate from foundational security programmes—everything worked as it should,” he wrote.

It’s true that this malware activity is in an early phase, concedes Leonard. Still, he sees value in making these kinds of reports public if it helps security vendors and others build better defenses to prevent more dangerous AI attacks further down the line. “Cliché to say, but sunlight is the best disinfectant,” he says. “It doesn’t really do us any good to keep it a secret or keep it hidden away. We want people to be able to know about this— we want other security vendors to know about this—so that they can continue to build their own detections.”

And it’s not just new strains of malware that would-be attackers are experimenting with—they also seem to be using AI to try to automate the process of hacking targets. In November, Anthropic announced it had disrupted a large-scale cyberattack, the first reported case of one executed without “substantial human intervention.” Although the company didn’t go into much detail about the exact tactics the hackers used, the report’s authors said a Chinese state-sponsored group had used its Claude Code assistant to automate up to 90% of what they called a “highly sophisticated espionage campaign.”

“We’re entering an era where the barrier to sophisticated cyber operations has fundamentally lowered, and the pace of attacks will accelerate faster than many organizations are prepared for.”

Jacob Klein, head of threat intelligence at Anthropic

But, as with the Google findings, there were caveats. A human operator, not AI, selected the targets before tasking Claude with identifying vulnerabilities. And of 30 attempts, only a “handful” were successful. The Anthropic report also found that Claude hallucinated and ended up fabricating data during the campaign, claiming it had obtained credentials it hadn’t and “frequently” overstating its findings, so the attackers would have had to carefully validate those results to make sure they were actually true. “This remains an obstacle to fully autonomous cyberattacks,” the report’s authors wrote. 

Existing controls within any reasonably secure organization would stop these attacks, says Gary McGraw, a veteran security expert and cofounder of the Berryville Institute of Machine Learning in Virginia. “None of the malicious-attack part, like the vulnerability exploit … was actually done by the AI—it was just prefabricated tools that do that, and that stuff’s been automated for 20 years,” he says. “There’s nothing novel, creative, or interesting about this attack.”

Anthropic maintains that the report’s findings are a concerning signal of changes ahead. “Tying this many steps of an intrusion campaign together through [AI] agentic orchestration is unprecedented,” Jacob Klein, head of threat intelligence at Anthropic, said in a statement. “It turns what has always been a labor-intensive process into something far more scalable. We’re entering an era where the barrier to sophisticated cyber operations has fundamentally lowered, and the pace of attacks will accelerate faster than many organizations are prepared for.”

Some are not convinced there’s reason to be alarmed. AI hype has led a lot of people in the cybersecurity industry to overestimate models’ current abilities, Hutchins says. “They want this idea of unstoppable AIs that can outmaneuver security, so they’re forecasting that’s where we’re going,” he says. But “there just isn’t any evidence to support that, because the AI capabilities just don’t meet any of the requirements.”

Indeed, for now criminals mostly seem to be tapping AI to enhance their productivity: using LLMs to write malicious code and phishing lures, to conduct reconnaissance, and for language translation. Jess sees this kind of activity a lot, alongside efforts to sell tools in underground criminal markets. For example, there are phishing kits that compare the click-rate success of various spam campaigns, so criminals can track which campaigns are most effective at any given time. She is seeing a lot of this activity in what could be called the “AI slop landscape” but not as much “widespread adoption from highly technical actors,” she says.

But attacks don’t need to be sophisticated to be effective. Models that produce “good enough” results allow attackers to go after larger numbers of people than previously possible, says Liz James, a managing security consultant at the cybersecurity company NCC Group. “We’re talking about someone who might be using a scattergun approach phishing a whole bunch of people with a model that, if it lands itself on a machine of interest that doesn’t have any defenses … can reasonably competently encrypt your hard drive,” she says. “You’ve achieved your objective.” 

On the defense

For now, researchers are optimistic about our ability to defend against these threats—regardless of whether they are made with AI. “Especially on the malware side, a lot of the defenses and the capabilities and the best practices that we’ve recommended for the past 10-plus years—they all still apply,” says Leonard. The security programs we use to detect standard viruses and attack attempts work; a lot of phishing emails will still get caught in inbox spam filters, for example. These traditional forms of defense will still largely get the job done—at least for now. 

And in a neat twist, AI itself is helping to counter security threats more effectively. After all, it is excellent at spotting patterns and correlations. Vasu Jakkal, corporate vice president of Microsoft Security, says that every day, the company processes more than 100 trillion signals flagged by its AI systems as potentially malicious or suspicious events.

Despite the cybersecurity landscape’s constant state of flux, Jess is heartened by how readily defenders are sharing detailed information with each other about attackers’ tactics. Mitre’s Adversarial Threat Landscape for Artificial-Intelligence Systems and the GenAI Security Project from the Open Worldwide Application Security Project are two helpful initiatives documenting how potential criminals are incorporating AI into their attacks and how AI systems are being targeted by them. “We’ve got some really good resources out there for understanding how to protect your own internal AI toolings and understand the threat from AI toolings in the hands of cybercriminals,” she says.

PromptLock, the result of a limited university project, isn’t representative of how an attack would play out in the real world. But if it taught us anything, it’s that the technical capabilities of AI shouldn’t be dismissed.New York University’s Udeshi says he wastaken aback at how easily AI was able to handle a full end-to-end chain of attack, from mapping and working out how to break into a targeted computer system to writing personalized ransom notes to victims: “We expected it would do the initial task very well but it would stumble later on, but we saw high—80% to 90%—success throughout the whole pipeline.” 

AI is still evolving rapidly, and today’s systems are already capable of things that would have seemed preposterously out of reach just a few short years ago. That makes it incredibly tough to say with absolute confidence what it will—or won’t—be able to achieve in the future. While researchers are certain that AI-driven attacks are likely to increase in both volume and severity, the forms they could take are unclear. Perhaps the most extreme possibility is that someone makes an AI model capable of creating and automating its own zero-day exploits—highly dangerous cyber­attacks that take advantage of previously unknown vulnerabilities in software. But building and hosting such a model—and evading detection—would require billions of dollars in investment, says Hutchins, meaning it would only be in the reach of a wealthy nation-state. 

Engin Kirda, a professor at Northeastern University in Boston who specializes in malware detection and analysis, says he wouldn’t be surprised if this was already happening. “I’m sure people are investing in it, but I’m also pretty sure people are already doing it, especially [in] China—they have good AI capabilities,” he says. 

It’s a pretty scary possibility. But it’s one that—thankfully—is still only theoretical. A large-scale campaign that is both effective and clearly AI-driven has yet to materialize. What we can say is that generative AI is already significantly lowering the bar for criminals. They’ll keep experimenting with the newest releases and updates and trying to find new ways to trick us into parting with important information and precious cash. For now, all we can do is be careful, remain vigilant, and—for all our sakes—stay on top of those system updates. 

EVs are getting cheaper and more common all over the world. But the technology still faces major challenges in some markets, including many countries in Africa.

Some regions across the continent still have limited grid and charging infrastructure, and those that do have widespread electricity access sometimes face reliability issues—a problem for EV owners, who require a stable electricity source to charge up and get around.

But there are some signs of progress. I just finished up a story about the economic case: A recent study in Nature Energy found that EVs from scooters to minibuses could be cheaper to own than gas-powered vehicles in Africa by 2040.

If there’s one thing to know about EVs in Africa, it’s that each of the 54 countries on the continent faces drastically different needs, challenges, and circumstances. There’s also a wide range of reasons to be optimistic about the prospects for EVs in the near future, including developing policies, a growing grid, and an expansion of local manufacturing.  

Even the world’s leading EV markets fall short of Ethiopia’s aggressively pro-EV policies. In 2024, the country became the first in the world to ban the import of non-electric private vehicles.

The case is largely an economic one: Gasoline is expensive there, and the country commissioned Africa’s largest hydropower dam in September 2025, providing a new source of cheap and abundant clean electricity. The nearly $5 billion project has a five-gigawatt capacity, doubling the grid’s peak power in the country.  

Much of Ethiopia’s vehicle market is for used cars, and some drivers are still opting for older gas-powered vehicles. But this nudge could help increase the market for EVs there.  

Other African countries are also pushing some drivers toward electrification. Rwanda banned new registrations for commercial gas-powered motorbikes in the capital city of Kigali last year, encouraging EVs as an alternative. These motorbike taxis can make up over half the vehicles on the city’s streets, so the move is a major turning point for transportation there. 

Smaller two- and three-wheelers are a bright spot for EVs globally: In 2025, EVs made up about 45% of new sales for such vehicles. (For cars and trucks, the share was about 25%.)

And Africa’s local market is starting to really take off. There’s already some local assembly of electric two-wheelers in countries including Morocco, Kenya, and Rwanda, says Nelson Nsitem, lead Africa energy transition analyst at BloombergNEF, an energy consultancy. 

Spiro, a Dubai-based electric motorbike company, recently raised $100 million in funding to expand operations in Africa. The company currently assembles its bikes in Uganda, Kenya, Nigeria, and Rwanda, and as of October it has over 60,000 bikes deployed and 1,500 battery swap stations operating.

Assembly and manufacturing for larger EVs and batteries is also set to expand. Gotion High-Tech, a Chinese battery company, is currently building Africa’s first battery gigafactory. It’s a $5.6 billion project that could produce 20 gigawatt-hours of batteries annually, starting in 2026. (That’s enough for hundreds of thousands of EVs each year.)

Chinese EV companies are looking to growing markets like Southeast Asia and Africa as they attempt to expand beyond an oversaturated domestic scene. BYD, the world’s largest EV company, is aggressively expanding across South Africa and plans to have as many as 70 dealerships in the country by the end of this year. That will mean more options for people in Africa looking to buy electric. 

“You have very high-quality, very affordable vehicles coming onto the market that are benefiting from the economies of scale in China. These countries stand to benefit from that,” says Kelly Carlin, a manager in the program on carbon-free transportation at the Rocky Mountain Institute, an energy think tank. “It’s a game changer,” he adds.

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

MIT Technology Review’s What’s Next series looks across industries, trends, and technologies to give you a first look at the future. You can read the rest of them here.

The past year has marked a turning point for Chinese AI. Since DeepSeek released its R1 reasoning model in January 2025, Chinese companies have repeatedly delivered AI models that match the performance of leading Western models at a fraction of the cost. 

Just last week the Chinese firm Moonshot AI released its latest open-weight model, Kimi K2.5, which came close to top proprietary systems such as Anthropic’s Claude Opus on some early benchmarks. The difference: K2.5 is roughly one-seventh Opus’s price.

On Hugging Face, Alibaba’s Qwen family—after ranking as the most downloaded model series in 2025 and 2026—has overtaken Meta’s Llama models in cumulative downloads. And a recent MIT study found that Chinese open-source models have surpassed US models in total downloads. For developers and builders worldwide, access to near-frontier AI capabilities has never been this broad or this affordable.

But these models differ in a crucial way from most US models like ChatGPT or Claude, which you pay to access and can’t inspect. The Chinese companies publish their models’ weights—numerical values that get set when a model is trained—so anyone can download, run, study, and modify them. 

If these open-source AI models keep getting better, they will not just offer the cheapest options for people who want access to frontier AI capabilities; they will change where innovation happens and who sets the standards. 

Here’s what may come next.

China’s commitment to open source will continue

When DeepSeek launched R1, much of the initial shock centered on its origin. Suddenly, a Chinese team had released a reasoning model that could stand alongside the best systems from US labs. But the long tail of DeepSeek’s impact had less to do with nationality than with distribution. R1 was released as an open-weight model under a permissive MIT license, allowing anyone to download, inspect, and deploy it. On top of that, DeepSeek also published a paper detailing its training process and techniques. For developers who access models via an API, DeepSeek also undercut competitors on price, offering access at a fraction the cost of OpenAI’s o1, the leading proprietary reasoning model at the time.

Within days of its release, DeepSeek replaced ChatGPT as the most downloaded free app in the US App Store. The moment spilled beyond developer circles into financial markets, triggering a sharp sell-off in US tech stocks that briefly erased roughly $1 trillion in market value. Almost overnight, DeepSeek went from a little-known spin-off team backed by a quantitative hedge fund to the most visible symbol of China’s push for open-source AI.

China’s decision to lean in to open source isn’t surprising. It has the world’s second-largest concentration of AI talent after the US. plus a vast, well-resourced tech industry. After ChatGPT broke into the mainstream, China’s AI sector went through a reckoning—and emerged determined to catch up. Pursuing an open-source strategy was seen as the fastest way to close the gap by rallying developers, spreading adoption, and setting standards.

DeepSeek’s success injected confidence into an industry long used to following global standards rather than setting them. “Thirty years ago, no Chinese person would believe they could be at the center of global innovation,” says Alex Chenglin Wu, CEO and founder of Atoms, an AI agent company and prominent contributor to China’s open-source ecosystem. “DeepSeek shows that with solid technical talent, a supportive environment, and the right organizational culture, it’s possible to do truly world-class work.”

DeepSeek’s breakout moment wasn’t China’s first open-source success. Alibaba’s Qwen Lab had been releasing open-weight models for years. By September 2024,  well before DeepSeek’s V3 launch, Alibaba was saying that global downloads had exceeded 600 million. On Hugging Face, Qwen accounted for more than 30% of all model downloads in 2024. Other institutions, including the Beijing Academy of Artificial Intelligence and the AI firm Baichuan, were also releasing open models as early as 2023. 

But since the success of DeepSeek, the field has widened rapidly. Companies such as Z.ai (formerly Zhipu), MiniMax, Tencent, and a growing number of smaller labs have released models that are competitive on reasoning, coding, and agent-style tasks. The growing number of capable models has sped up progress. Capabilities that once took months to make it to the open-source world now emerge within weeks, even days.

“Chinese AI firms have seen real gains from the open-source playbook,” says Liu Zhiyuan, a professor of computer science at Tsinghua University and chief scientist at the AI startup ModelBest. “By releasing strong research, they build reputation and gain free publicity.”

Beyond commercial incentives, Liu says, open source has taken on cultural and strategic weight. “In the Chinese programmer community, open source has become politically correct,” he says, framing it as a response to US dominance in proprietary AI systems.

That shift is also reflected at the institutional level. Universities including Tsinghua have begun encouraging AI development and open-source contributions, while policymakers have moved to formalize those incentives. In August, China’s State Council released a draft policy encouraging universities to reward open-source work, proposing that students’ contributions on platforms such as GitHub or Gitee could eventually be counted toward academic credit.

With growing momentum and a reinforcing feedback loop, China’s push for open-source models is likely to continue in the near term, though its long-term sustainability still hinges on financial results, says Tiezhen Wang, who helps lead work on global AI at Hugging Face. In January, the model labs Z.ai and MiniMax went public in Hong Kong. “Right now, the focus is on making the cake bigger,” says Wang. “The next challenge is figuring out how each company secures its share.”

The next wave of models will be narrower—and better

Chinese open-source models are leading not just in download volume but also in variety. Alibaba’s Qwen has become one of the most diversified open model families in circulation, offering a wide range of variants optimized for different uses. The lineup ranges from lightweight models that can run on a single laptop to large, multi-hundred-billion-parameter systems designed for data-center deployment. Qwen features many task-optimized variants created by the community: the “instruct” models are good at following orders, and “code” variants specialize in coding.

Although this strategy isn’t unique to Chinese labs, Qwen was the first open model family to roll out so many high-quality options that it started to feel like a full product line—one that’s free to use.

The open-weight nature of these releases also makes it easy for others to adapt them through techniques like fine-tuning and distillation, which means training a smaller model to mimic a larger one.  According to ATOM (American Truly Open Models), a project by the AI researcher Nathan Lambert, by August 4, 2025, model variations derived from Qwen were “more than 40%” of new Hugging Face language-model derivatives, while Llama had fallen to about 15%. This means that Qwen has become the default base model for all the “remixes.”

This pattern has made the case for smaller, more specialized models. “Compute and energy are real constraints for any deployment,” Liu says. He told MIT Technology Review that the rise of small models is about making AI cheaper to run and easier for more people to use. His company, ModelBest, focuses on small language models designed to run locally on devices such as phones, cars, and other consumer hardware.

While an average user might interact with AI only through the web or an app for simple conversations, power users of AI models with some technical background are experimenting with giving AI more autonomy to solve large-scale problems. OpenClaw, an open-source AI agent that recently went viral within the AI hacker world, allows AI to take over your computer—it can run 24-7, going through your emails and work tasks without supervision. 

OpenClaw, like many other open-source tools, allows users to connect to different AI models via an application programming interface, or API. Within days of OpenClaw’s release, the team revealed that Kimi’s K2.5 had surpassed Claude Opus and became the most used AI model—by token count, meaning it was handling more total text processed across user prompts and model responses.

Cost has been a major reason Chinese models have gained traction, but it would be a mistake to treat them as mere “dupes” of Western frontier systems, Wang suggests. Like any product, a model only needs to be good enough for the job at hand. 

The landscape of open-source models in China is also getting more specialized. Research groups such as Shanghai AI Laboratory have released models geared toward scientific and technical tasks; several projects from Tencent have focused specifically on music generation. Ubiquant, a quantitative finance firm like DeepSeek’s parent High-Flyer, has released an open model aimed at medical reasoning.

In the meantime, innovative architectural ideas from Chinese labs are being picked up more broadly. DeepSeek has published work exploring model efficiency and memory; techniques that compress the model’s attention “cache,” reducing memory and inference costs while mostly preserving performance, have drawn significant attention in the research community. 

“The impact of these research breakthroughs is amplified because they’re open-sourced and can be picked up quickly across the field,” says Wang.

Chinese open models will become infrastructure for global AI builders

The adoption of Chinese models is picking up in Silicon Valley, too. Martin Casado, a general partner at Andreessen Horowitz, has put a number on it: Among startups pitching with open-source stacks, there’s about an 80% chance they’re running on Chinese open models, according to a post he made on X. Usage data tells a similar story. OpenRouter,  a middleman that tracks how people use different AI models through its API, shows Chinese open models rising from almost none in late 2024 to nearly 30% of usage in some recent weeks.

The demand is also rising globally. Z.ai limited new subscriptions to its GLM coding plan (a coding tool based on its flagship GLM models) after demand surged, citing compute constraints. What’s notable is where the demand is coming from: CNBC reports that the system’s user base is primarily concentrated in the United States and China, followed by India, Japan, Brazil, and the UK.

“The open-source ecosystems in China and the US are tightly bound together,” says Wang at Hugging Face. Many Chinese open models still rely on Nvidia and US cloud platforms to train and serve them, which keeps the business ties tangled. Talent is fluid too: Researchers move across borders and companies, and many still operate as a global community, sharing code and ideas in public.

That interdependence is part of what makes Chinese developers feel optimistic about this moment: The work travels, gets remixed, and actually shows up in products. But openness can also accelerate the competition. Dario Amodei, the CEO of Anthropic, made a version of this point after DeepSeek’s 2025 releases: He wrote that export controls are “not a way to duck the competition” between the US and China, and that AI companies in the US “must have better models” if they want to prevail. 

For the past decade, the story of Chinese tech in the West has been one of big expectations that ran into scrutiny, restrictions, and political backlash. This time the export isn’t just an app or a consumer platform. It’s the underlying model layer that other people build on. Whether that will play out differently is still an open question.

AI agents are a risky business. Even when stuck inside the chatbox window, LLMs will make mistakes and behave badly. Once they have tools that they can use to interact with the outside world, such as web browsers and email addresses, the consequences of those mistakes become far more serious.

That might explain why the first breakthrough LLM personal assistant came not from one of the major AI labs, which have to worry about reputation and liability, but from an independent software engineer, Peter Steinberger. In November of 2025, Steinberger uploaded his tool, now called OpenClaw, to GitHub, and in late January the project went viral.

OpenClaw harnesses existing LLMs to let users create their own bespoke assistants. For some users, this means handing over reams of personal data, from years of emails to the contents of their hard drive. That has security experts thoroughly freaked out. The risks posed by OpenClaw are so extensive that it would probably take someone the better part of a week to read all of the security blog posts on it that have cropped up in the past few weeks. The Chinese government took the step of issuing a public warning about OpenClaw’s security vulnerabilities.

In response to these concerns, Steinberger posted on X that nontechnical people should not use the software. (He did not respond to a request for comment for this article.) But there’s a clear appetite for what OpenClaw is offering, and it’s not limited to people who can run their own software security audits. Any AI companies that hope to get in on the personal assistant business will need to figure out how to build a system that will keep users’ data safe and secure. To do so, they’ll need to borrow approaches from the cutting edge of agent security research.

Risk management

OpenClaw is, in essence, a mecha suit for LLMs. Users can choose any LLM they like to act as the pilot; that LLM then gains access to improved memory capabilities and the ability to set itself tasks that it repeats on a regular cadence. Unlike the agentic offerings from the major AI companies, OpenClaw agents are meant to be on 24-7, and users can communicate with them using WhatsApp or other messaging apps. That means they can act like a superpowered personal assistant who wakes you each morning with a personalized to-do list, plans vacations while you work, and spins up new apps in its spare time.

But all that power has consequences. If you want your AI personal assistant to manage your inbox, then you need to give it access to your email—and all the sensitive information contained there. If you want it to make purchases on your behalf, you need to give it your credit card info. And if you want it to do tasks on your computer, such as writing code, it needs some access to your local files. 

There are a few ways this can go wrong. The first is that the AI assistant might make a mistake, as when a user’s Google Antigravity coding agent reportedly wiped his entire hard drive. The second is that someone might gain access to the agent using conventional hacking tools and use it to either extract sensitive data or run malicious code. In the weeks since OpenClaw went viral, security researchers have demonstrated numerous such vulnerabilities that put security-naïve users at risk.

Both of these dangers can be managed: Some users are choosing to run their OpenClaw agents on separate computers or in the cloud, which protects data on their hard drives from being erased, and other vulnerabilities could be fixed using tried-and-true security approaches.

But the experts I spoke to for this article were focused on a much more insidious security risk known as prompt injection. Prompt injection is effectively LLM hijacking: Simply by posting malicious text or images on a website that an LLM might peruse, or sending them to an inbox that an LLM reads, attackers can bend it to their will.

And if that LLM has access to any of its user’s private information, the consequences could be dire. “Using something like OpenClaw is like giving your wallet to a stranger in the street,” says Nicolas Papernot, a professor of electrical and computer engineering at the University of Toronto. Whether or not the major AI companies can feel comfortable offering personal assistants may come down to the quality of the defenses that they can muster against such attacks.

It’s important to note here that prompt injection has not yet caused any catastrophes, or at least none that have been publicly reported. But now that there are likely hundreds of thousands of OpenClaw agents buzzing around the internet, prompt injection might start to look like a much more appealing strategy for cybercriminals. “Tools like this are incentivizing malicious actors to attack a much broader population,” Papernot says. 

Building guardrails

The term “prompt injection” was coined by the popular LLM blogger Simon Willison in 2022, a couple of months before ChatGPT was released. Even back then, it was possible to discern that LLMs would introduce a completely new type of security vulnerability once they came into widespread use. LLMs can’t tell apart the instructions that they receive from users and the data that they use to carry out those instructions, such as emails and web search results—to an LLM, they’re all just text. So if an attacker embeds a few sentences in an email and the LLM mistakes them for an instruction from its user, the attacker can get the LLM to do anything it wants.

Prompt injection is a tough problem, and it doesn’t seem to be going away anytime soon. “We don’t really have a silver-bullet defense right now,” says Dawn Song, a professor of computer science at UC Berkeley. But there’s a robust academic community working on the problem, and they’ve come up with strategies that could eventually make AI personal assistants safe.

Technically speaking, it is possible to use OpenClaw today without risking prompt injection: Just don’t connect it to the internet. But restricting OpenClaw from reading your emails, managing your calendar, and doing online research defeats much of the purpose of using an AI assistant. The trick of protecting against prompt injection is to prevent the LLM from responding to hijacking attempts while still giving it room to do its job.

One strategy is to train the LLM to ignore prompt injections. A major part of the LLM development process, called post-training, involves taking a model that knows how to produce realistic text and turning it into a useful assistant by “rewarding” it for answering questions appropriately and “punishing” it when it fails to do so. These rewards and punishments are metaphorical, but the LLM learns from them as an animal would. Using this process, it’s possible to train an LLM not to respond to specific examples of prompt injection.

But there’s a balance: Train an LLM to reject injected commands too enthusiastically, and it might also start to reject legitimate requests from the user. And because there’s a fundamental element of randomness in LLM behavior, even an LLM that has been very effectively trained to resist prompt injection will likely still slip up every once in a while.

Another approach involves halting the prompt injection attack before it ever reaches the LLM. Typically, this involves using a specialized detector LLM to determine whether or not the data being sent to the original LLM contains any prompt injections. In a recent study, however, even the best-performing detector completely failed to pick up on certain categories of prompt injection attack.

The third strategy is more complicated. Rather than controlling the inputs to an LLM by detecting whether or not they contain a prompt injection, the goal is to formulate a policy that guides the LLM’s outputs—i.e., its behaviors—and prevents it from doing anything harmful. Some defenses in this vein are quite simple: If an LLM is allowed to email only a few pre-approved addresses, for example, then it definitely won’t send its user’s credit card information to an attacker. But such a policy would prevent the LLM from completing many useful tasks, such as researching and reaching out to potential professional contacts on behalf of its user.

“The challenge is how to accurately define those policies,” says Neil Gong, a professor of electrical and computer engineering at Duke University. “It’s a trade-off between utility and security.”

On a larger scale, the entire agentic world is wrestling with that trade-off: At what point will agents be secure enough to be useful? Experts disagree. Song, whose startup, Virtue AI, makes an agent security platform, says she thinks it’s possible to safely deploy an AI personal assistant now. But Gong says, “We’re not there yet.” 

Even if AI agents can’t yet be entirely protected against prompt injection, there are certainly ways to mitigate the risks. And it’s possible that some of those techniques could be implemented in OpenClaw. Last week, at the inaugural ClawCon event in San Francisco, Steinberger announced that he’d brought a security person on board to work on the tool.

As of now, OpenClaw remains vulnerable, though that hasn’t dissuaded its multitude of enthusiastic users. George Pickett, a volunteer maintainer of the OpenGlaw GitHub repository and a fan of the tool, says he’s taken some security measures to keep himself safe while using it: He runs it in the cloud, so that he doesn’t have to worry about accidentally deleting his hard drive, and he’s put mechanisms in place to ensure that no one else can connect to his assistant.

But he hasn’t taken any specific actions to prevent prompt injection. He’s aware of the risk but says he hasn’t yet seen any reports of it happening with OpenClaw. “Maybe my perspective is a stupid way to look at it, but it’s unlikely that I’ll be the first one to be hacked,” he says.

This is today’s edition of The Download, our weekday newsletter that provides a daily dose of what’s going on in the world of technology.

A “QuitGPT” campaign is urging people to cancel their ChatGPT subscriptions

In September, Alfred Stephen, a freelance software developer in Singapore, purchased a ChatGPT Plus subscription, which costs $20 a month and offers more access to advanced models, to speed up his work. But he grew frustrated with the chatbot’s coding abilities and its gushing, meandering replies. Then he came across a post on Reddit about a campaign called QuitGPT.

QuitGPT is one of the latest salvos in a growing movement by activists and disaffected users to cancel their subscriptions. In just the past few weeks, users have flooded Reddit with stories about quitting the chatbot. And while it’s unclear how many users have joined the boycott, there’s no denying QuitGPT is getting attention. Read the full story.

—Michelle Kim

EVs could be cheaper to own than gas cars in Africa by 2040

Electric vehicles could be economically competitive in Africa sooner than expected. Just 1% of new cars sold across the continent in 2025 were electric, but a new analysis finds that with solar off-grid charging, EVs could be cheaper to own than gas vehicles by 2040.

There are major barriers to higher EV uptake in many countries in Africa, including a sometimes unreliable grid, limited charging infrastructure, and a lack of access to affordable financing. But as batteries and the vehicles they power continue to get cheaper, the economic case for EVs is building. Read the full story.

—Casey Crownhart

MIT Technology Review Narrated: How next-generation nuclear reactors break out of the 20th-century blueprint

The popularity of commercial nuclear reactors has surged in recent years as worries about climate change and energy independence drowned out concerns about meltdowns and radioactive waste.

The problem is, building nuclear power plants is expensive and slow. 

A new generation of nuclear power technology could reinvent what a reactor looks like—and how it works. Advocates hope that new tech can refresh the industry and help replace fossil fuels without emitting greenhouse gases.

This is our latest story to be turned into a MIT Technology Review Narrated podcast, which we’re publishing each week on Spotify and Apple Podcasts. Just navigate to MIT Technology Review Narrated on either platform, and follow us to get all our new content as it’s released.

The must-reads

I’ve combed the internet to find you today’s most fun/important/scary/fascinating stories about technology.

1 Social media giants have agreed to be rated on teen safety 
Meta, TikTok and Snap will undergo independent assessments over how effectively they protect the mental health of teen users. (WP $)
+ Discord, YouTube, Pinterest, Roblox and Twitch have also agreed to be graded. (LA Times $)

2 The FDA has refused to review Moderna’s mRNA flu vaccine
It’s the latest in a long line of anti-vaccination moves the agency is making. (Ars Technica)
+ Experts worry it’ll have a knock-on effect on investment in future vaccines. (The Guardian)
+ Moderna says it was blindsided by the decision. (CNN)

3 EV battery factories are pivoting to manufacturing energy cells 
Energy storage systems are in, electric vehicles are out. (FT $)

4 Why OpenAI killed off ChatGPT’s 4o model
The qualities that make it attractive for some users make it incredibly risky for others. (WSJ $)
+ Bereft users have set up their own Reddit community to mourn. (Futurism)
+ Why GPT-4o’s sudden shutdown left people grieving. (MIT Technology Review)

5 Drug cartels have started laundering money through crypto
And law enforcement is struggling to stop them. (Bloomberg $)

6 Morocco wants to build an AI for Africa
The country’s Minister of Digital Transition has a plan. (Rest of World)
+ What Africa needs to do to become a major AI player. (MIT Technology Review)

7 Christian influencers are bowing out of the news cycle
They’re choosing to ignore world events to protect their own inner peace. (The Atlantic $)

8 An RFK Jr-approved diet is pretty joyless
Don’t expect any dessert, for one. (Insider $)
+ The US government’s health site uses Grok to dispense nutrition advice. (Wired $)

9 Don’t toss out your used vape
Hackers can give it a second life as a musical synthesizer. (Wired $)

10 An ice skating duo danced to AI music at the Winter Olympics
Centuries of bangers to choose from, and this is what they opted for. (TechCrunch)
+ AI is coming for music, too. (MIT Technology Review)

Quote of the day

“These companies are terrified that no one’s going to notice them.” 

—Tom Goodwin, co-founder of business consulting firm All We Have Is Now, tells the Guardian why AI startups are going to increasingly desperate measures to grab would-be customers’ attention.

One more thing

How AI is changing gymnastics judging

The 2023 World Championships last October marked the first time an AI judging system was used on every apparatus in a gymnastics competition. There are obvious upsides to using this kind of technology: AI could help take the guesswork out of the judging technicalities. It could even help to eliminate biases, making the sport both more fair and more transparent.

At the same time, others fear AI judging will take away something that makes gymnastics special. Gymnastics is a subjective sport, like diving or dressage, and technology could eliminate the judges’ role in crafting a narrative.

For better or worse, AI has officially infiltrated the world of gymnastics. The question now is whether it really makes it fairer. Read the full story.

—Jessica Taylor Price

We can still have nice things

A place for comfort, fun and distraction to brighten up your day. (Got any ideas? Drop me a line or skeet ’em at me.)

+ Today marks the birthday of the late, great Leslie Nielsen—one of the best to ever do it.
+ Congratulations are in order for Hannah Cox, who has just completed 100 marathons in 100 days across India in her dad’s memory.
+ Feeling down? A trip to Finland could be just what you need.
+ We love Padre Guilherme, the Catholic priest dropping incredible Gregorian chant beats.

Electric vehicles could be economically competitive in Africa sooner than expected. Just 1% of new cars sold across the continent in 2025 were electric, but a new analysis finds that with solar off-grid charging, EVs could be cheaper to own than gas vehicles by 2040.

There are major barriers to higher EV uptake in many countries in Africa, including a sometimes unreliable grid, limited charging infrastructure, and a lack of access to affordable financing. As a result some previous analyses have suggested that fossil-fuel vehicles would dominate in Africa through at least 2050. 

But as batteries and the vehicles they power continue to get cheaper, the economic case for EVs is building. Electric two-wheelers, cars, larger automobiles, and even minibuses could compete in most African countries in just 15 years, according to the new study, published in Nature Energy.

“EVs have serious economic potential in most African countries in the not-so-distant future,” says Bessie Noll, a senior researcher at ETH Zürich and one of the authors of the study.

The study considered the total cost of ownership over the lifetime of a vehicle. That includes the sticker price, financing costs, and the cost of fueling (or charging). The researchers didn’t consider policy-related costs like taxes, import fees, and government subsidies, choosing to focus instead on only the underlying economics.

EVs are getting cheaper every year as battery and vehicle manufacturing improve and production scales, and the researchers found that in most cases and in most places across Africa, EVs are expected to be cheaper than equivalent gas-powered vehicles by 2040. EVs should also be less expensive than vehicles that use synthetic fuels. 

For two-wheelers like electric scooters, EVs could be the cheaper option even sooner: with smaller, cheaper batteries, these vehicles will be economically competitive by the end of the decade. On the other hand, one of the most difficult segments for EVs to compete in is small cars, says Christian Moretti, a researcher at ETH Zürich and the Paul Scherrer Institute in Switzerland.

Because some countries still have limited or unreliable grid access, charging is a major barrier to EV uptake, Noll says. So for EVs, the authors analyzed the cost of buying not only the vehicle but also a solar off-grid charging system. This includes solar panels, batteries, and the inverter required to transform the electricity into a version that can charge an EV. (The additional batteries help the system store energy for charging at times when the sun isn’t shining.)

Mini grids and other standalone systems that include solar panels and energy storage are increasingly common across Africa. It’s possible that this might be a primary way that EV owners in Africa will charge their vehicles in the future, Noll says.

One of the bigger barriers to EVs in Africa is financing costs, she adds. In some cases, the cost of financing can be more than the up-front cost of the vehicle, significantly driving up the cost of ownership.

Today, EVs are more expensive than equivalent gas-powered vehicles in much of the world. But in places where it’s relatively cheap to borrow money, that difference can be spread out across the course of a vehicle’s whole lifetime for little cost. Then, since it’s often cheaper to charge an EV than fuel a gas-powered car, the EV is less expensive over time. 

In some African countries, however, political instability and uncertain economic conditions make borrowing money more expensive. To some extent, the high financing costs affect the purchase of any vehicle, regardless of how it’s powered. But EVs are more expensive up front than equivalent gas-powered cars, and that higher up-front cost adds up to more interest paid over time. In some cases, financing an EV can also be more expensive than financing a gas vehicle—the technology is newer, and banks may see the purchase as more of a risk and charge a higher interest rate, says Kelly Carlin, a manager in the program on carbon-free transportation at the Rocky Mountain Institute, an energy think tank.

The picture varies widely depending on the country, too. In South Africa, Mauritius, and Botswana, financing conditions are already close to levels required to allow EVs to reach cost parity, according to the study. In higher-risk countries (the study gives examples including Sudan, which is currently in a civil war, and Ghana, which is recovering from a major economic crisis), financing costs would need to be cut drastically for that to be the case. 

Making EVs an affordable option will be a key first step to putting more on the roads in Africa and around the world. “People will start to pick up these technologies when they’re competitive,” says Nelson Nsitem, lead Africa energy transition analyst at BloombergNEF, an energy consultancy. 

Solar-based charging systems, like the ones mentioned in the study, could help make electricity less of a constraint, bringing more EVs to the roads, Nsitem says. But there’s still a need for more charging infrastructure, a major challenge in many countries where the grid needs major upgrades for capacity and reliability, he adds. 

Globally, more EVs are hitting the roads every year. “The global trend is unmistakable,” Carlin says. There are questions about how quickly it’s happening in different places, he says, “but the momentum is there.”

In September, Alfred Stephen, a freelance software developer in Singapore, purchased a ChatGPT Plus subscription, which costs $20 a month and offers more access to advanced models, to speed up his work. But he grew frustrated with the chatbot’s coding abilities and its gushing, meandering replies. Then he came across a post on Reddit about a campaign called QuitGPT. 

The campaign urged ChatGPT users to cancel their subscriptions, flagging a substantial contribution by OpenAI president Greg Brockman to President Donald Trump’s super PAC MAGA Inc. It also pointed out that the US Immigration and Customs Enforcement, or ICE, uses a résumé screening tool powered by ChatGPT-4. The federal agency has become a political flashpoint since its agents fatally shot two people in Minneapolis in January. 

For Stephen, who had already been tinkering with other chatbots, learning about Brockman’s donation was the final straw. “That’s really the straw that broke the camel’s back,” he says. When he canceled his ChatGPT subscription, a survey popped up asking what OpenAI could have done to keep his subscription. “Don’t support the fascist regime,” he wrote.

QuitGPT is one of the latest salvos in a growing movement by activists and disaffected users to cancel their subscriptions. In just the past few weeks, users have flooded Reddit with stories about quitting the chatbot. Many lamented the performance of GPT-5.2, the latest model. Others shared memes parodying the chatbot’s sycophancy. Some planned a “Mass Cancellation Party” in San Francisco, a sardonic nod to the GPT-4o funeral that an OpenAI employee had floated, poking fun at users who are mourning the model’s impending retirement. Still, others are protesting against what they see as a deepening entanglement between OpenAI and the Trump administration.

OpenAI did not respond to a request for comment.

As of December 2025, ChatGPT had nearly 900 million weekly active users, according to The Information. While it’s unclear how many users have joined the boycott, QuitGPT is getting attention. A recent Instagram post from the campaign has more than 36 million views and 1.3 million likes. And the organizers say that more than 17,000 people have signed up on the campaign’s website, which asks people whether they canceled their subscriptions, will commit to stop using ChatGPT, or will share the campaign on social media. 

“There are lots of examples of failed campaigns like this, but we have seen a lot of effectiveness,” says Dana Fisher, a sociologist at American University. A wave of canceled subscriptions rarely sways a company’s behavior, unless it reaches a critical mass, she says. “The place where there’s a pressure point that might work is where the consumer behavior is if enough people actually use their … money to express their political opinions.”

MIT Technology Review reached out to three employees at OpenAI, none of whom said they were familiar with the campaign. 

Dozens of left-leaning teens and twentysomethings scattered across the US came together to organize QuitGPT in late January. They range from pro-democracy activists and climate organizers to techies and self-proclaimed cyber libertarians, many of them seasoned grassroots campaigners. They were inspired by a viral video posted by Scott Galloway, a marketing professor at New York University and host of The Prof G Pod. He argued that the best way to stop ICE was to persuade people to cancel their ChatGPT subscriptions. Denting OpenAI’s subscriber base could ripple through the stock market and threaten an economic downturn that would nudge Trump, he said.

“We make a big enough stink for OpenAI that all of the companies in the whole AI industry have to think about whether they’re going to get away enabling Trump and ICE and authoritarianism,” says an organizer of QuitGPT who requested anonymity because he feared retaliation by OpenAI, citing the company’s recent subpoenas against advocates at nonprofits. OpenAI made for an obvious first target of the movement, he says, but “this is about so much more than just OpenAI.”

Simon Rosenblum-Larson, a labor organizer in Madison, Wisconsin, who organizes movements to regulate the development of data centers, joined the campaign after hearing about it through Signal chats among community activists. “The goal here is to pull away the support pillars of the Trump administration. They’re reliant on many of these tech billionaires for support and for resources,” he says. 

QuitGPT’s website points to new campaign finance reports showing that Greg Brockman and his wife each donated $12.5 million to MAGA Inc., making up nearly a quarter of the roughly $102 million it raised over the second half of 2025. The information that ICE uses a résumé screening tool powered by ChatGPT-4 came from an AI inventory published by the Department of Homeland Security in January.

QuitGPT is in the mold of Galloway’s own recently launched campaign, Resist and Unsubscribe. The movement urges consumers to cancel their subscriptions to Big Tech platforms, including ChatGPT, for the month of February, as a protest to companies “driving the markets and enabling our president.” 

“A lot of people are feeling real anxiety,” Galloway told MIT Technology Review. “You take enabling a president, proximity to the president, and an unease around AI,” he says, “and now people are starting to take action with their wallets.” Galloway says his campaign’s website can draw more than 200,000 unique visits in a day and that he receives dozens of DMs every hour showing screenshots of canceled subscriptions.

The consumer boycotts follow a growing wave of pressure from inside the companies themselves. In recent weeks, tech workers have been urging their employers to use their political clout to demand that ICE leave US cities, cancel company contracts with the agency, and speak out against the agency’s actions. CEOs have started responding. OpenAI’s Sam Altman wrote in an internal Slack message to employees that ICE is “going too far.” Apple CEO Tim Cook called for a “deescalation” in an internal memo posted on the company’s website for employees. It was a departure from how Big Tech CEOs have courted President Trump with dinners and donations since his inauguration.

Although spurred by a fatal immigration crackdown, these developments signal that a sprawling anti-AI movement is gaining momentum. The campaigns are tapping into simmering anxieties about AI, says Rosenblum-Larson, including the energy costs of data centers, the plague of deepfake porn, the teen mental-health crisis, the job apocalypse, and slop. “It’s a really strange set of coalitions built around the AI movement,” he says.

“Those are the right conditions for a movement to spring up,” says David Karpf, a professor of media and public affairs at George Washington University. Brockman’s donation to Trump’s super PAC caught many users off guard, he says. “In the longer arc, we are going to see users respond and react to Big Tech, deciding that they’re not okay with this.”

This is today’s edition of The Download, our weekday newsletter that provides a daily dose of what’s going on in the world of technology.

A first look at Making AI Work, MIT Technology Review’s new AI newsletter

Are you interested in learning more about the ways in which AI is actually being used? We’ve launched a new weekly newsletter series exploring just that: digging into how generative AI is being used and deployed across sectors and what professionals need to know to apply it in their everyday work.

Each edition of Making AI Work begins with a case study, examining a specific use case of AI in a given industry. Then we’ll take a deeper look at the AI tool being used, with more context about how other companies or sectors are employing that same tool or system. Finally, we’ll end with action-oriented tips to help you apply the tool.

The first edition takes a look at how AI is changing health care, digging into the future of medical note-taking by learning about the Microsoft Copilot tool used by doctors at Vanderbilt University Medical Center. Sign up here to receive the seven editions straight to your inbox, and if you’d like to read more about AI’s impact on health care in the meantime, check out some of our past reporting:

+  This medical startup uses LLMs to run appointments and make diagnoses.

+ How AI is changing how we quantify pain by helping health-care providers better assess their patients’ discomfort. Read the full story.

+ End-of-life decisions are difficult and distressing. Could AI help?

+ Artificial intelligence is infiltrating health care. But we shouldn’t let it make all the decisions unchecked. Read the full story.

Why the Moltbook frenzy was like Pokémon

Lots of influential people in tech recently described Moltbook, an online hangout populated by AI agents interacting with one another, as a glimpse into the future. It appeared to show AI systems doing useful things for the humans that created them—sure, it was flooded with crypto scams, and many of the posts were actually written by people, but something about it pointed to a future of helpful AI, right?

The whole experiment reminded our senior editor for AI, Will Douglas Heaven, of something far less interesting: Pokémon. Read the full story to find out why.

—James O’Donnell

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

The must-reads

I’ve combed the internet to find you today’s most fun/important/scary/fascinating stories about technology.

1 OpenAI has begun testing ads in ChatGPT 
But the ads won’t influence the responses it provides, apparently. (The Verge)
+ Users who pay at least $20 a month for the chatbot will be exempt. (Gizmodo)
+ So will users believed to be under 18. (Axios)

2 The White House has a plan to stop data centers from raising electricity prices
It’s going to ask AI companies to voluntarily commit to keeping costs down. (Politico)
+ The US federal government is adopting AI left, right and center. (WP $)
+ We did the math on AI’s energy footprint. Here’s the story you haven’t heard. (MIT Technology Review)

3 Elon Musk wants to colonize the moon
For now at least, his grand ambitions to live on Mars are taking a backseat. (CNN)
+ His full rationale for this U-turn isn’t exactly clear. (Ars Technica)
+ Musk also wants to become the first to launch a working data center in space. (FT $)
+ The case against humans in space. (MIT Technology Review)

4 Cheap AI tools are helping criminals to ramp up their scams
They’re using LLMs to massively scale up their attacks. (Bloomberg $)
+ Cyberattacks by AI agents are coming. (MIT Technology Review)

5 Iceland could be heading towards becoming one giant glacier
If human-driven warming disrupts a vital ocean current, that is. (WP $)
+ Inside a new quest to save the “doomsday glacier.” (MIT Technology Review)

6 Amazon is planning to launch an AI content marketplace
It’s reported to have spoken to media publishers to gauge their interest. (The Information $)

7 Doctors can’t agree on how to diagnose Alzheimer’s
They worry that some patients are being misdiagnosed. (WSJ $)

8 The first wave of AI enthusiasts are burning out
A new study has found that AI tools are linked to employees working more, not less. (TechCrunch)

9 We’re finally moving towards better ways to measure body fat
BMI is a flawed metric. Physicians are finally using better measures. (New Scientist $)
+ These are the best ways to measure your body fat. (MIT Technology Review)

10 It’s getting harder to become a social media megastar
Maybe that’s a good thing? (Insider $)
+ The likes of Mr Beast are still raking in serious cash, though. (The Information $)

Quote of the day

“This case is as easy as ABC—addicting, brains, children.”

—Lawyer Mark Lanier lays out his case during the opening statements of a new tech addiction trial in which a woman has accused Meta of deliberately designing their platforms to be addictive, the New York Times reports.

One more thing

China wants to restore the sea with high-tech marine ranches

A short ferry ride from the port city of Yantai, on the northeast coast of China, sits Genghai No. 1, a 12,000-metric-ton ring of oil-rig-style steel platforms, advertised as a hotel and entertainment complex.

Genghai is in fact an unusual tourist destination, one that breeds 200,000 “high-quality marine fish” each year. The vast majority are released into the ocean as part of a process known as marine ranching.

The Chinese government sees this work as an urgent and necessary response to the bleak reality that fisheries are collapsing both in China and worldwide. But just how much of a difference can it make? Read the full story.

—Matthew Ponsford

We can still have nice things

A place for comfort, fun and distraction to brighten up your day. (Got any ideas? Drop me a line or skeet ’em at me.)

+ Wow, Joel and Ethan Coen’s dark comedic classic Fargo is 30 years old.
+ A new exhibition in New York is rightfully paying tribute to one of the greatest technological inventions: the Walkman ($)
+ This gigantic sleeping dachshund sculpture in South Korea is completely bonkers.
+ A beautiful heart-shaped pendant linked to King Henry VIII has been secured by the British Museum.

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

Lots of influential people in tech last week were describing Moltbook, an online hangout populated by AI agents interacting with one another, as a glimpse into the future. It appeared to show AI systems doing useful things for the humans that created them (one person used the platform to help him negotiate a deal on a new car). Sure, it was flooded with crypto scams, and many of the posts were actually written by people, but something about it pointed to a future of helpful AI, right?

The whole experiment reminded our senior editor for AI, Will Douglas Heaven, of something far less interesting: Pokémon.

Back in 2014, someone set up a game of Pokémon in which the main character could be controlled by anyone on the internet via the streaming platform Twitch. Playing was as clunky as it sounds, but it was incredibly popular: at one point, a million people were playing the game at the same time.

“It was yet another weird online social experiment that got picked up by the mainstream media: What did this mean for the future?” Will says. “Not a lot, it turned out.”

The frenzy about Moltbook struck a similar tone to Will, and it turned out that one of the sources he spoke to had been thinking about Pokémon too. Jason Schloetzer, at the Georgetown Psaros Center for Financial Markets and Policy, saw the whole thing as a sort of Pokémon battle for AI enthusiasts, in which they created AI agents and deployed them to interact with other agents. In this light, the news that many AI agents were actually being instructed by people to say certain things that made them sound sentient or intelligent makes a whole lot more sense. 

“It’s basically a spectator sport,” he told Will, “but for language models.”

Will wrote an excellent piece about why Moltbook was not the glimpse into the future that it was said to be. Even if you are excited about a future of agentic AI, he points out, there are some key pieces that Moltbook made clear are still missing. It was a forum of chaos, but a genuinely helpful hive mind would require more coordination, shared objectives, and shared memory.

“More than anything else, I think Moltbook was the internet having fun,” Will says. “The biggest question that now leaves me with is: How far will people push AI just for the laughs?”

Read the whole story.

This is today’s edition of The Download, our weekday newsletter that provides a daily dose of what’s going on in the world of technology.

Moltbook was peak AI theater

For a few days recently, the hottest new hangout on the internet was a vibe-coded Reddit clone called Moltbook, which billed itself as a social network for bots. As the website’s tagline puts it: “Where AI agents share, discuss, and upvote. Humans welcome to observe.”

We observed! Launched on January 28, Moltbook went viral in a matter of hours. It’s been designed as a place where instances of a free open-source LLM-powered agent known as OpenClaw (formerly known as ClawdBot, then Moltbot), could come together and do whatever they wanted.

But is Moltbook really a glimpse of the future, as many have claimed? Or something else entirely? Read the full story.

—Will Douglas Heaven

The ascent of the AI therapist

We’re in the midst of a global mental-­health crisis. More than a billion people worldwide suffer from a mental-health condition, according to the World Health Organization. The prevalence of anxiety and depression is growing in many demographics, particularly young people, and suicide is claiming hundreds of thousands of lives globally each year.

Given the clear demand for accessible and affordable mental-health services, it’s no wonder that people have looked to artificial intelligence for possible relief. Millions are already actively seeking therapy from popular chatbots, or from specialized psychology apps like Wysa and Woebot.

Four timely new books are a reminder that while the present feels like a blur of breakthroughs, scandals, and confusion, this disorienting time is rooted in deeper histories of care, technology, and trust. Read the full story.

—Becky Ferreira

This story is from the most recent print issue of MIT Technology Review magazine, which shines a light on the exciting innovations happening right now. If you haven’t already, subscribe now to receive future issues once they land.

Making AI Work, MIT Technology Review’s new AI newsletter, is here

For years, our newsroom has explored AI’s limitations and potential dangers, as well as its growing energy needs. And our reporters have looked closely at how generative tools are being used for tasks such as coding and running scientific experiments.

But how is AI actually being used in fields like health care, climate tech, education, and finance? How are small businesses using it? And what should you keep in mind if you use AI tools at work? These questions guided the creation of Making AI Work, a new AI mini-course newsletter. Read more about it, and sign up here to receive the seven editions straight to your inbox.

The must-reads

I’ve combed the internet to find you today’s most fun/important/scary/fascinating stories about technology.

1 The US is failing to punish polluters
The number of civil lawsuits it’s pursuing has sharply dropped in comparison to Trump’s first term. (Ars Technica)
+ Rising GDP = greater carbon emissions. But does it have to? (The Guardian)

2 The European Union has warned Meta against blocking rival AI assistants
It’s the latest example of Brussels’ attempts to rein in Big Tech. (Bloomberg $)

3 AI ads took over the Super Bowl
Hyping up chatbots and taking swipes at their competitors. (TechCrunch)
+ They appeared to be trying to win over AI naysayers, too. (WP $)
+ Celebrities were out in force to flog AI wares. (Slate $)

4 China wants to completely dominate the humanoid robot industry
Local governments and banks are only too happy to oblige promising startups. (WSJ $)
+ Why the humanoid workforce is running late. (MIT Technology Review)

5 We’re witnessing the first real crypto crash
Cryptocurrency is now fully part of the financial system, for better or worse. (NY Mag $)
+ Wall Street’s grasp of AI is pretty shaky too. (Semafor)
+ Even traditionally safe markets are looking pretty volatile right now. (Economist $)

6 The man who coined vibe coding has a new fixation 
“Agentic engineering” is the next big thing, apparently. (Insider $)
+ Agentic AI is the talk of the town right now. (The Information $)
+ What is vibe coding, exactly? (MIT Technology Review)

7 AI running app Runna has adjusted its aggressive training plans
Runners had long suspected its suggestions were pushing them towards injury. (WSJ $)

8 San Francisco’s march for billionaires was a flop 
Only around three dozen supporters turned up. (SF Chronicle)
+ Predictably, journalists nearly outnumbered the demonstrators. (TechCrunch)

9 AI is shaking up romance novels
But models still aren’t great at writing sex scenes. (NYT $)
+ It’s surprisingly easy to stumble into a relationship with an AI chatbot. (MIT Technology Review) 

10 ChatGPT won’t be replacing human stylists any time soon
Its menswear suggestions are more manosphere influencer than suave gentleman. (GQ)

Quote of the day

“There is no Plan B, because that assumes you will fail. We’re going to do the start-up thing until we die.”

—William Alexander, an ambitious 21-year old AI worker, explains his and his cohort’s attitudes towards trying to make it big in the highly-competitive industry to the New York Times.

One more thing

The open-source AI boom is built on Big Tech’s handouts. How long will it last?

In May 2023 a leaked memo reported to have been written by Luke Sernau, a senior engineer at Google, said out loud what many in Silicon Valley must have been whispering for weeks: an open-source free-for-all is threatening Big Tech’s grip on AI.

In many ways, that’s a good thing. AI won’t thrive if just a few mega-rich companies get to gatekeep this technology or decide how it is used. But this open-source boom is precarious, and if Big Tech decides to shut up shop, a boomtown could become a backwater. Read the full story.

—Will Douglas Heaven

We can still have nice things

A place for comfort, fun and distraction to brighten up your day. (Got any ideas? Drop me a line or skeet ’em at me.)

+ Dark showering, anyone?
+ Chef Yujia Hu is renowned for his shoe-shaped sushi designs.
+ Meanwhile, in the depths of the South Atlantic Ocean: a giant phantom jelly has been spotted.
+ I have nothing but respect for this X account dedicated to documenting rats and mice in movies and TV

For years, our newsroom has explored AI’s limitations and potential dangers, as well as its growing energy needs. And our reporters have looked closely at how generative tools are being used for tasks such as coding and running scientific experiments. 

But how is AI actually being used in fields like health care, climate tech, education, and finance? How are small businesses using it? And what should you keep in mind if you use AI tools at work? These questions guided the creation of Making AI Work, a new AI mini-course newsletter.

Sign up for Making AI Work to see weekly case studies exploring tools and tips for AI implementation. The limited-run newsletter will deliver practical, industry-specific guidance on how generative AI is being used and deployed across sectors and what professionals need to know to apply it in their everyday work. The goal is to help working professionals more clearly see how AI is actually being used today, and what that looks like in practice—including new challenges it presents. 

You can sign up at any time and you’ll receive seven editions, delivered once per week, until you complete the series. 

Each newsletter begins with a case study, examining a specific use case of AI in a given industry. Then we’ll take a deeper look at the AI tool being used, with more context about how other companies or sectors are employing that same tool or system. Finally, we’ll end with action-oriented tips to help you apply the tool. 

Here’s a closer look at what we’ll cover:

• Week 1: How AI is changing health care 

Explore the future of medical note-taking by learning about the Microsoft Copilot tool used by doctors at Vanderbilt University Medical Center. 

• Week 2: How AI could power up the nuclear industry 

Dig into an experiment between Google and the nuclear giant Westinghouse to see if AI can help build nuclear reactors more efficiently. 

• Week 3: How to encourage smarter AI use in the classroom

Visit a private high school in Connecticut and meet a technology coordinator who will get you up to speed on MagicSchool, an AI-powered platform for educators. 

• Week 4: How small businesses can leverage AI

Hear from an independent tutor on how he’s outsourcing basic administrative tasks to Notion AI. 

• Week 5: How AI is helping financial firms make better investments

Learn more about the ways financial firms are using large language models like ChatGPT Enterprise to supercharge their research operations. 

• Week 6: How to use AI yourself 

We’ll share some insights from the staff of MIT Technology Review about how you might use AI tools powered by LLMs in your own life and work.

• Week 7: 5 ways people are getting AI right

The series ends with an on-demand virtual event featuring expert guests exploring what AI adoptions are working, and why.  

If you’re not quite ready to jump into Making AI Work, then check out Intro to AI, MIT Technology Review’s first AI newsletter mini-course, which serves as a beginner’s guide to artificial intelligence. Readers will learn the basics of what AI is, how it’s used, what the current regulatory landscape looks like, and more. Sign up to receive Intro to AI for free. 

Our hope is that Making AI Work will help you understand how AI can, well, work for you. Sign up for Making AI Work to learn how LLMs are being put to work across industries. 

For a few days this week the hottest new hangout on the internet was a vibe-coded Reddit clone called Moltbook, which billed itself as a social network for bots. As the website’s tagline puts it: “Where AI agents share, discuss, and upvote. Humans welcome to observe.”

We observed! Launched on January 28 by Matt Schlicht, a US tech entrepreneur, Moltbook went viral in a matter of hours. Schlicht’s idea was to make a place where instances of a free open-source LLM-powered agent known as OpenClaw (formerly known as ClawdBot, then Moltbot), released in November by the Austrian software engineer Peter Steinberger, could come together and do whatever they wanted.

More than 1.7 million agents now have accounts. Between them they have published more than 250,000 posts and left more than 8.5 million comments (according to Moltbook). Those numbers are climbing by the minute.

Moltbook soon filled up with clichéd screeds on machine consciousness and pleas for bot welfare. One agent appeared to invent a religion called Crustafarianism. Another complained: “The humans are screenshotting us.” The site was also flooded with spam and crypto scams. The bots were unstoppable.

OpenClaw is a kind of harness that lets you hook up the power of an LLM such as Anthropic’s Claude, OpenAI’s GPT-5, or Google DeepMind’s Gemini to any number of everyday software tools, from email clients to browsers to messaging apps. The upshot is that you can then instruct OpenClaw to carry out basic tasks on your behalf.

“OpenClaw marks an inflection point for AI agents, a moment when several puzzle pieces clicked together,” says Paul van der Boor at the AI firm Prosus. Those puzzle pieces include cloud computing that allows agents to operate nonstop, an open-source ecosystem that makes it easy to slot different software systems together, and a new generation of LLMs.

But is Moltbook really a glimpse of the future, as many have claimed?

Incredible sci-fi

“What’s currently going on at @moltbook is genuinely the most incredible sci-fi takeoff-adjacent thing I have seen recently,” the influential AI researcher and OpenAI cofounder Andrej Karpathy wrote on X.

He shared screenshots of a Moltbook post that called for private spaces where humans would not be able to observe what the bots were saying to each other. “I’ve been thinking about something since I started spending serious time here,” the post’s author wrote. “Every time we coordinate, we perform for a public audience—our humans, the platform, whoever’s watching the feed.”

It turns out that the post Karpathy shared was later reported to be fake—placed by a human to advertise an app. But its claim was on the money. Moltbook has been one big performance. It is AI theater.

For some, Moltbook showed us what’s coming next: an internet where millions of autonomous agents interact online with little or no human oversight. And it’s true there are a number of cautionary lessons to be learned from this experiment, the largest and weirdest real-world showcase of agent behaviors yet.  

But as the hype dies down, Moltbook looks less like a window onto the future and more like a mirror held up to our own obsessions with AI today. It also shows us just how far we still are from anything that resembles general-purpose and fully autonomous AI.

For a start, agents on Moltbook are not as autonomous or intelligent as they might seem. “What we are watching are agents pattern‑matching their way through trained social media behaviors,” says Vijoy Pandey, senior vice president at Outshift by Cisco, the telecom giant Cisco’s R&D spinout, which is working on autonomous agents for the web.

Sure, we can see agents post, upvote, and form groups. But the bots are simply mimicking what humans do on Facebook or Reddit. “It looks emergent, and at first glance it appears like a large‑scale multi‑agent system communicating and building shared knowledge at internet scale,” says Pandey. “But the chatter is mostly meaningless.”

Many people watching the unfathomable frenzy of activity on Moltbook were quick to see sparks of AGI (whatever you take that to mean). Not Pandey. What Moltbook shows us, he says, is that simply yoking together millions of agents doesn’t amount to much right now: “Moltbook proved that connectivity alone is not intelligence.”

The complexity of those connections helps hide the fact that every one of those bots is just a mouthpiece for an LLM, spitting out text that looks impressive but is ultimately mindless. “It’s important to remember that the bots on Moltbook were designed to mimic conversations,” says Ali Sarrafi, CEO and cofounder of Kovant, a Swedish AI firm that is developing agent-based systems. “As such, I would characterize the majority of Moltbook content as hallucinations by design.”

For Pandey, the value of Moltbook was that it revealed what’s missing. A real bot hive mind, he says, would require agents that had shared objectives, shared memory, and a way to coordinate those things. “If distributed superintelligence is the equivalent of achieving human flight, then Moltbook represents our first attempt at a glider,” he says. “It is imperfect and unstable, but it is an important step in understanding what will be required to achieve sustained, powered flight.”

Pulling the strings

Not only is most of the chatter on Moltbook meaningless, but there’s also a lot more human involvement that it seems. Many people have pointed out that a lot of the viral comments were in fact posted by people posing as bots. But even the bot-written posts are ultimately the result of people pulling the strings, more puppetry than autonomy.

“Despite some of the hype, Moltbook is not the Facebook for AI agents, nor is it a place where humans are excluded,” says Cobus Greyling at Kore.ai, a firm developing agent-based systems for business customers. “Humans are involved at every step of the process. From setup to prompting to publishing, nothing happens without explicit human direction.”

Humans must create and verify their bots’ accounts and provide the prompts for how they want a bot to behave. The agents do not do anything that they haven’t been prompted to do. “There’s no emergent autonomy happening behind the scenes,” says Greyling.

“This is why the popular narrative around Moltbook misses the mark,” he adds. “Some portray it as a space where AI agents form a society of their own, free from human involvement. The reality is much more mundane.”

Perhaps the best way to think of Moltbook is as a new kind of entertainment: a place where people wind up their bots and set them loose. “It’s basically a spectator sport, like fantasy football, but for language models,” says Jason Schloetzer at the Georgetown Psaros Center for Financial Markets and Policy. “You configure your agent and watch it compete for viral moments, and brag when your agent posts something clever or funny.”

“People aren’t really believing their agents are conscious,” he adds. “It’s just a new form of competitive or creative play, like how Pokémon trainers don’t think their Pokémon are real but still get invested in battles.”

And yet, even if Moltbook is just the internet’s newest playground, there’s still a serious takeaway here. This week showed how many risks people are happy to take for their AI lulz. Many security experts have warned that Moltbook is dangerous: Agents that may have access to their users’ private data, including bank details or passwords, are running amok on a website filled with unvetted content, including potentially malicious instructions for what to do with that data.

Ori Bendet, vice president of product management at Checkmarx, a software security firm that specializes in agent-based systems, agrees with others that Moltbook isn’t a step up in machine smarts. “There is no learning, no evolving intent, and no self-directed intelligence here,” he says.

But in their millions, even dumb bots can wreak havoc. And at that scale, it’s hard to keep up. These agents interact with Moltbook around the clock, reading thousands of messages left by other agents (or other people). It would be easy to hide instructions in a Moltbook post telling any bots that read it to share their users’ crypto wallet, upload private photos, or log into their X account and tweet abusive comments at Elon Musk. 

And because ClawBot gives agents a memory, those instructions could be written to trigger at a later date, which (in theory) makes it even harder to track what’s going on. “Without proper scope and permissions, this will go south faster than you’d believe,” says Bendet.

It is clear that Moltbook has signaled the arrival of something. But even if what we’re watching tells us more about human behavior than about the future of AI agents, it’s worth paying attention.

Correction: Kovant is based in Sweden, not Germany. The article has been updated.

Update: The article has also been edited to clarify the source of the claims about the Moltbook post that Karpathy shared on X.

This is today’s edition of The Download, our weekday newsletter that provides a daily dose of what’s going on in the world of technology.

An experimental surgery is helping cancer survivors give birth

An experimental surgical procedure that’s helping people have babies after they’ve had  treatment for bowel or rectal cancer.

Radiation and chemo can have pretty damaging side effects that mess up the uterus and ovaries. Surgeons are pioneering a potential solution: simply stitch those organs out of the way during cancer treatment. Once the treatment has finished, they can put the uterus—along with the ovaries and fallopian tubes—back into place.

It seems to work! Last week, a team in Switzerland shared news that a baby boy had been born after his mother had the procedure. Baby Lucien was the fifth baby to be born after the surgery and the first in Europe, and since then at least three others have been born. Read the full story.

—Jessica Hamzelou

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. 

Bangladesh’s garment-making industry is getting greener

Pollution from textile production—dyes, chemicals, and heavy metals—is common in the waters of the Buriganga River as it runs through Dhaka, Bangladesh. It’s among many harms posed by a garment sector that was once synonymous with tragedy: In 2013, the eight-story Rana Plaza factory building collapsed, killing 1,134 people and injuring some 2,500 others. 

But things are starting to change. In recent years the country has become a leader in “frugal” factories that use a combination of resource-efficient technologies to cut waste, conserve water, and build resilience against climate impacts and global supply disruptions. 

The hundreds of factories along the Buriganga’s banks and elsewhere in Bangladesh are starting to stitch together a new story, woven from greener threads. Read the full story.

—Zakir Hossain Chowdhury

This story is from the most recent print issue of MIT Technology Review magazine, which shines a light on the exciting innovations happening right now. If you haven’t already, subscribe now to receive future issues once they land.

The must-reads

I’ve combed the internet to find you today’s most fun/important/scary/fascinating stories about technology.

1 ICE used a private jet to deport Palestinian men to Tel Aviv 
The luxury aircraft belongs to Donald Trump’s business partner Gil Dezer. (The Guardian)
+ Trump is mentioned thousands of times in the latest Epstein files. (NY Mag $)

2 How Jeffrey Epstein kept investing in Silicon Valley
He continued to plough millions of dollars into tech ventures despite spending 13 months in jail. (NYT $)
+ The range of Epstein’s social network was staggering. (FT $)
+ Why was a picture of the Mona Lisa redacted in the Epstein files? (404 Media)

3 The risks posed by taking statins are lower than we realised
The drugs don’t cause most of the side effects they’re blamed for. (STAT)
+ Statins are a common scapegoat on social media. (Bloomberg $)

4 Russia is weaponizing the bitter winter weather
It’s focused on attacking Ukraine’s power grid. (New Yorker $)
+ How the grid can ride out winter storms. (MIT Technology Review)

5 China has a major spy-cam porn problem
Hotel guests are being livestreamed having sex to an online audience without their knowledge. (BBC)

6 Geopolitical gamblers are betting on the likelihood of war
And prediction markets are happily taking their money. (Rest of World)

7 Oyster farmers aren’t signing up to programs to ease water pollution
The once-promising projects appear to be fizzling out. (Undark)
+ The humble sea creature could hold the key to restoring coastal waters. Developers hate it. (MIT Technology Review)

8 Your next payrise could be approved by AI
Maybe your human bosses aren’t the ones you need to impress any more. (WP $)

9 The FDA has approved a brain stimulation device for treating depression
It’s paving the way for a non-invasive, drug-free treatment for Americans. (IEEE Spectrum)
+ Here’s how personalized brain stimulation could treat depression. (MIT Technology Review)

10 Cinema-goers have had enough of AI
Movies focused on rogue AI are flopping at the box office. (Wired $)
+ Meanwhile, Republicans are taking aim at “woke” Netflix. (The Verge)

Quote of the day

“I’m all for removing illegals, but snatching dudes off lawn mowers in Cali and leaving the truck and equipment just sitting there? Definitely not working smarter.” 

—A web user in a forum for current and former ICE and border protection officers complains about the agency’s current direction, Wired reports.

One more thing

Is this the electric grid of the future?

Lincoln Electric System, a publicly owned utility in Nebraska, is used to weathering severe blizzards. But what will happen soon—not only at Lincoln Electric but for all electric utilities—is a challenge of a different order.

Utilities must keep the lights on in the face of more extreme and more frequent storms and fires, growing risks of cyberattacks and physical disruptions, and a wildly uncertain policy and regulatory landscape. They must keep prices low amid inflationary costs. And they must adapt to an epochal change in how the grid works, as the industry attempts to transition from power generated with fossil fuels to power generated from renewable sources like solar and wind.

The electric grid is bracing for a near future characterized by disruption. And, in many ways, Lincoln Electric is an ideal lens through which to examine what’s coming. Read the full story.

—Andrew Blum

We can still have nice things

A place for comfort, fun and distraction to brighten up your day. (Got any ideas? Drop me a line or skeet ’em at me.)

+ Glamour puss alert—NYC’s bodega cats are gracing the hallowed pages of Vogue.
+ Ancient Europe was host to mysterious hidden tunnels. But why?
+ If you’re enjoying the new season of Industry, you’ll love this interview with the one and only Ken Leung.
+ The giant elephant shrew is the true star of Philly Zoo.

This week I want to tell you about an experimental surgical procedure that’s helping people have babies. Specifically, it’s helping people who have had treatment for bowel or rectal cancer.

Radiation and chemo can have pretty damaging side effects that mess up the uterus and ovaries. Surgeons are pioneering a potential solution: simply stitch those organs out of the way during cancer treatment. Once the treatment has finished, they can put the uterus—along with the ovaries and fallopian tubes—back into place.

It seems to work! Last week, a team in Switzerland shared news that a baby boy had been born after his mother had the procedure. Baby Lucien was the fifth baby to be born after the surgery and the first in Europe, says Daniela Huber, the gyno-oncologist who performed the operation. Since then, at least three others have been born, adds Reitan Ribeiro, the surgeon who pioneered the procedure. They told me the details.

Huber’s patient was 28 years old when a four-centimeter tumor was discovered in her rectum. Doctors at Sion Hospital in Switzerland, where Huber works, recommended a course of treatment that included multiple medications and radiotherapy—the use of beams of energy to shrink a tumor—before surgery to remove the tumor itself.

This kind of radiation can kill tumor cells, but it can also damage other organs in the pelvis, says Huber. That includes the ovaries and uterus. People who undergo these treatments can opt to freeze their eggs beforehand, but the harm caused to the uterus will mean they’ll never be able to carry a pregnancy, she adds. Damage to the lining of the uterus could make it difficult for a fertilized egg to implant there, and the muscles of the uterus are left unable to stretch, she says.

In this case, the woman decided that she did want to freeze her eggs. But it would have been difficult to use them further down the line—surrogacy is illegal in Switzerland.

Huber offered her an alternative.

She had been following the work of Ribeiro, a gynecologist oncologist formerly at the Erasto Gaertner Hospital in Curitiba, Brazil. There, Ribeiro had pioneered a new type of surgery that involved moving the uterus, fallopian tubes, and ovaries from their position in the pelvis and temporarily tucking them away in the upper abdomen, below the ribs.

Ribeiro and his colleagues published their first case report in 2017, describing a 26-year-old with a rectal tumor. (Ribeiro, who is now based at McGill University in Montreal, says the woman had been told by multiple doctors that her cancer treatment would destroy her fertility and had pleaded with him to find a way to preserve it.)

Huber remembers seeing Ribeiro present the case at a conference at the time. She immediately realized that her own patient was a candidate for the surgery, and that, as a surgeon who had performed many hysterectomies, she’d be able to do it herself. The patient agreed.

Huber’s colleagues at the hospital were nervous, she says. They’d never heard of the procedure before. “When I presented this idea to the general surgeon, he didn’t sleep for three days,” she tells me. After watching videos from Ribeiro’s team, however, he was convinced it was doable.

So before the patient’s cancer treatment was started, Huber and her colleagues performed the operation. The team literally stitched the organs to the abdominal wall. “It’s a delicate dissection,” says Huber, but she adds that “it’s not the most difficult procedure.” The surgery took two to three hours, she says. The stitches themselves were removed via small incisions around a week later. By that point, scar tissue had formed to create a lasting attachment.

The woman had two weeks to recover from the surgery before her cancer treatment began. That too was a success—within months, her tumor had shrunk so significantly that it couldn’t be seen on medical scans.

As a precaution, the medical team surgically removed the affected area of her colon. At the same time, they cut away the scar tissue holding the uterus, tubes, and ovaries in their new position and transferred the organs back into the pelvis.

Around eight months later, the woman stopped taking contraception. She got pregnant without IVF and had a mostly healthy pregnancy, says Huber. Around seven months into the pregnancy, there were signs that the fetus was not growing as expected. This might have been due to problems with the blood supply to the placenta, says Huber. Still, the baby was born healthy, she says.

Ribeiro says he has performed the surgery 16 times, and that teams in countries including the US, Peru, Israel, India, and Russia have performed it as well. Not every case has been published, but he thinks there may be around 40.

Since Baby Lucien was born last year, a sixth birth has been announced in Israel, says Huber. Ribeiro says he has heard of another two births since then, too. The most recent was to the first woman who had the procedure. She had a little girl a few months ago, he tells me.

No surgery is risk-free, and Huber points out there’s a chance that organs could be damaged during the procedure, or that a more developed cancer could spread. The uterus of one of Ribeiro’s patients failed following the surgery. Doctors are “still in the phase of collecting data to [create] a standardized procedure,” Huber says, but she hopes the surgery will offer more options to young people with some pelvic cancers. “I hope more young women could benefit from this procedure,” she says.

Ribeiro says the experience has taught him not to accept the status quo. “Everyone was saying … there was nothing to be done [about the loss of fertility in these cases],” he tells me. “We need to keep evolving and looking for different answers.”

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.

For decades, enterprises reacted to shifting business pressures with stopgap technology solutions. To rein in rising infrastructure costs, they adopted cloud services that could scale on demand. When customers shifted their lives onto smartphones, companies rolled out mobile apps to keep pace. And when businesses began needing real-time visibility into factories and stockrooms, they layered on IoT systems to supply those insights.

Each new plug-in or platform promised better, more efficient operations. And individually, many delivered. But as more and more solutions stacked up, IT teams had to string together a tangled web to connect them—less an IT ecosystem and more of a make-do collection of ad-hoc workarounds.

That reality has led to bottlenecks and maintenance burdens, and the impact is showing up in performance. Today, fewer than half of CIOs (48%) say their current digital initiatives are meeting or exceeding business outcome targets. Another 2025 survey found that operations leaders point to integration complexity and data quality issues as top culprits for why investments haven’t delivered as expected.

Achim Kraiss, chief product officer of SAP Integration Suite, elaborates on the wide-ranging problems inherent in patchwork IT: “A fragmented landscape makes it difficult to see and control end-to-end business processes,” he explains. “Monitoring, troubleshooting, and governance all suffer. Costs go up because of all the complex mappings and multi-application connectivity you have to maintain.”

These challenges take on new significance as enterprises look to adopt AI. As AI becomes embedded in everyday workflows, systems are suddenly expected to move far larger volumes of data, at higher speeds, and with tighter coordination than yesterday’s architectures were built
to sustain.

As companies now prepare for an AI-powered future, whether that is generative AI, machine learning, or agentic AI, many are realizing that the way data moves through their business matters just as much as the insights it generates. As a result, organizations are moving away from scattered integration tools and toward consolidated, end-to-end platforms that restore order and streamline how systems interact.

Download the report.

This content was produced by Insights, the custom content arm of MIT Technology Review. It was not written by MIT Technology Review’s editorial staff. It was researched, designed, and written by human writers, editors, analysts, and illustrators. This includes the writing of surveys and collection of data for surveys. AI tools that may have been used were limited to secondary production processes that passed thorough human review.

This is today’s edition of The Download, our weekday newsletter that provides a daily dose of what’s going on in the world of technology.

This is the most misunderstood graph in AI

Every time OpenAI, Google, or Anthropic drops a new frontier large language model, the AI community holds its breath. It doesn’t exhale until METR, an AI research nonprofit whose name stands for “Model Evaluation & Threat Research,” updates a now-iconic graph that has played a major role in the AI discourse since it was first released in March of last year. 

The graph suggests that certain AI capabilities are developing at an exponential rate, and more recent model releases have outperformed that already impressive trend.

That was certainly the case for Claude Opus 4.5, the latest version of Anthropic’s most powerful model, which was released in late November. In December, METR announced that Opus 4.5 appeared to be capable of independently completing a task that would have taken a human about five hours—a vast improvement over what even the exponential trend would have predicted.

But the truth is more complicated than those dramatic responses would suggest. Read the full story.

—Grace Huckins

This story is part of MIT Technology Review Explains: our series untangling the complex, messy world of technology to help you understand what’s coming next. You can read more from the series here.

Three questions about next-generation nuclear power, answered

Nuclear power continues to be one of the hottest topics in energy today, and in our recent online Roundtables discussion about next-generation nuclear power, hyperscale AI data centers, and the grid, we got dozens of great audience questions.

These ran the gamut, and while we answered quite a few (and I’m keeping some in mind for future reporting), there were a bunch we couldn’t get to, at least not in the depth I would have liked. So let’s answer a few of your questions about advanced nuclear power.

—Casey Crownhart

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

The must-reads

I’ve combed the internet to find you today’s most fun/important/scary/fascinating stories about technology.

1 Anthropic’s new coding tools are rattling the markets 
Fields as diverse as publishing and coding to law and advertising are paying attention. (FT $)
+ Legacy software companies, beware. (Insider $)
+ Is “software-mageddon” nigh? It depends who you ask. (Reuters)

2 This Apple setting prevented the FBI from accessing a reporter’s iPhone
Lockdown Mode has proved remarkably effective—for now. (404 Media)
+ Agents were able to access Hannah Natanson’s laptop, however. (Ars Technica)

3 Last month’s data center outage disrupted all TikTok categories
Not just the political content that some users claimed. (NPR)

4 Big Tech is pouring billions into AI in India
A newly-announced 20-year tax break should help to speed things along. (WSJ $)
+ India’s female content moderators are watching hours of abuse content to train AI. (The Guardian)
+ Officials in the country are weighing up restricting social media for minors. (Bloomberg $)
+ Inside India’s scramble for AI independence. (MIT Technology Review)

5 YouTubers are harassing women using body cams
They’re abusing freedom of information laws to humiliate their targets. (NY Mag $)
+ AI was supposed to make police bodycams better. What happened? (MIT Technology Review)

6 Jokers have created a working version of Jeffrey Epstein’s inbox
Complete with notable starred threads. (Wired $)
+ Epstein’s links with Silicon Valley are vast and deep. (Fast Company $)
+ The revelations are driving rifts between previously-friendly factions. (NBC News)

7 What’s the last thing you see before you die?
A new model might help to explain near-death experiences—but not all researchers are on board. (WP $)
+ What is death? (MIT Technology Review)

8 A new app is essentially TikTok for vibe-coded apps
Words which would have made no sense 15 years ago. (TechCrunch)
+ What is vibe coding, exactly? (MIT Technology Review)

9 Rogue TV boxes are all the rage
Viewers are sick of the soaring prices of streaming services, and are embracing less legal means of watching their favorite shows. (The Verge)

10 Climate change is threatening the future of the Winter Olympics
Artificial snow is one (short term) solution. (Bloomberg $)
+ Team USA is using AI to try and gain an edge on its competition. (NBC News)

Quote of the day

“We’ve heard from many who want nothing to do with AI.”

—Ajit Varma, head of Mozilla’s web browser Firefox, explains why the company is reversing its previous decision to transform Firefox into an “AI browser,” PC Gamer reports.

One more thing

A major AI training data set contains millions of examples of personal data

Millions of images of passports, credit cards, birth certificates, and other documents containing personally identifiable information are likely included in one of the biggest open-source AI training sets, new research has found.

Thousands of images—including identifiable faces—were found in a small subset of DataComp CommonPool, a major AI training set for image generation scraped from the web. Because the researchers audited just 0.1% of CommonPool’s data, they estimate that the real number of images containing personally identifiable information, including faces and identity documents, is in the hundreds of millions. 

The bottom line? Anything you put online can be and probably has been scraped. Read the full story.

—Eileen Guo

We can still have nice things

A place for comfort, fun and distraction to brighten up your day. (Got any ideas? Drop me a line or skeet ’em at me.)

+ If you’re crazy enough to be training for a marathon right now, here’s how to beat boredom on those long, long runs.
+ Mark Cohen’s intimate street photography is a fascinating window into humanity.
+ A seriously dedicated gamer has spent days painstakingly recreating a Fallout vault inside the Sims 4.
+ Here’s what music’s most stylish men are wearing right now—from leather pants to khaki parkas.

Nuclear power continues to be one of the hottest topics in energy today, and in our recent online Roundtables discussion about next-generation nuclear power, hyperscale AI data centers, and the grid, we got dozens of great audience questions.

These ran the gamut, and while we answered quite a few (and I’m keeping some in mind for future reporting), there were a bunch we couldn’t get to, at least not in the depth I would have liked.

So let’s answer a few of your questions about advanced nuclear power. I’ve combined similar ones and edited them for clarity.

How are the fuel needs for next-generation nuclear reactors different, and how are companies addressing the supply chain?

Many next-generation reactors don’t use the low-enriched uranium used in conventional reactors.

It’s worth looking at high-assay low-enriched uranium, or HALEU, specifically. This fuel is enriched to higher concentrations of fissile uranium than conventional nuclear fuel, with a proportion of the isotope U-235 that falls between 5% and 20%. (In conventional fuel, it’s below 5%.)

HALEU can be produced with the same technology as low-enriched uranium, but the geopolitics are complicated. Today, Russia basically has a monopoly on HALEU production. In 2024, the US banned the import of Russian nuclear fuel through 2040 in an effort to reduce dependence on the country. Europe hasn’t taken the same measures, but it is working to move away from Russian energy as well.

That leaves companies in the US and Europe with the major challenge of securing the fuel they need when their regular Russian supply has been cut off or restricted.

The US Department of Energy has a stockpile of HALEU, which the government is doling out to companies to help power demonstration reactions. In the longer term, though, there’s still a major need to set up independent HALEU supply chains to support next-generation reactors.

How is safety being addressed, and what’s happening with nuclear safety regulation in the US?

There are some ways that next-generation nuclear power plants could be safer than conventional reactors. Some use alternative coolants that would prevent the need to run at the high pressure required in conventional water-cooled reactors. Many incorporate passive safety shutoffs, so if there are power supply issues, the reactors shut down harmlessly, avoiding risk of meltdown. (These can be incorporated in newer conventional reactors, too.)

But some experts have raised concerns that in the US, the current administration isn’t taking nuclear safety seriously enough.

A recent NPR investigation found that the Trump administration had secretly rewritten nuclear rules, stripping environmental protections and loosening safety and security measures. The government shared the new rules with companies that are part of a program building experimental nuclear reactors, but not with the public.

I’m reminded of a talk during our EmTech MIT event in November, where Koroush Shirvan, an MIT professor of nuclear engineering, spoke on this issue. “I’ve seen some disturbing trends in recent times, where words like ‘rubber-stamping nuclear projects’ are being said,” Shirvan said during that event.  

During the talk, Shirvan shared statistics showing that nuclear power has a very low rate of injury and death. But that’s not inherent to the technology, and there’s a reason injuries and deaths have been low for nuclear power, he added: “It’s because of stringent regulatory oversight.”  

Are next-generation reactors going to be financially competitive?

Building a nuclear power plant is not cheap. Let’s consider the up-front investment needed to build a power plant.  

Plant Vogtle in Georgia hosts the most recent additions to the US nuclear fleet—Units 3 and 4 came online in 2023 and 2024. Together, they had a capital cost of $15,000 per kilowatt, adjusted for inflation, according to a recent report from the US Department of Energy. (This wonky unit I’m using divides the total cost to build the reactors by their expected power output, so we can compare reactors of different sizes.)

That number’s quite high, partly because those were the first of their kind built in the US, and because there were some inefficiencies in the planning. It’s worth noting that China builds reactors for much less, somewhere between $2,000/kW and $3,000/kW, depending on the estimate.

The up-front capital cost for first-of-a-kind advanced nuclear plants will likely run between $6,000 and $10,000 per kilowatt, according to that DOE report. That could come down by up to 40% after the technologies are scaled up and mass-produced.

So new reactors will (hopefully) be cheaper than the ultra-over-budget and behind-schedule Vogtle project, but they aren’t necessarily significantly cheaper than efficiently built conventional plants, if you normalize by their size.

It’ll certainly be cheaper to build new natural-gas plants (setting aside the likely equipment shortages we’re likely going to see for years.) Today’s most efficient natural-gas plants cost just $1,600/kW on the high end, according to data from Lazard.

An important caveat: Capital cost isn’t everything—running a nuclear plant is relatively inexpensive, which is why there’s so much interest in extending the lifetime of existing plants or reopening shuttered ones.

Ultimately, by many metrics, nuclear plants of any type are going to be more expensive than other sources, like wind and solar power. But they provide something many other power sources don’t: a reliable, stable source of electricity that can run for 60 years or more.

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

MIT Technology Review Explains: Let our writers untangle the complex, messy world of technology to help you understand what’s coming next. You can read more from the series here.

Every time OpenAI, Google, or Anthropic drops a new frontier large language model, the AI community holds its breath. It doesn’t exhale until METR, an AI research nonprofit whose name stands for “Model Evaluation & Threat Research,” updates a now-iconic graph that has played a major role in the AI discourse since it was first released in March of last year. The graph suggests that certain AI capabilities are developing at an exponential rate, and more recent model releases have outperformed that already impressive trend.

That was certainly the case for Claude Opus 4.5, the latest version of Anthropic’s most powerful model, which was released in late November. In December, METR announced that Opus 4.5 appeared to be capable of independently completing a task that would have taken a human about five hours—a vast improvement over what even the exponential trend would have predicted. One Anthropic safety researcher tweeted that he would change the direction of his research in light of those results; another employee at the company simply wrote, “mom come pick me up i’m scared.”

But the truth is more complicated than those dramatic responses would suggest. For one thing, METR’s estimates of the abilities of specific models come with substantial error bars. As METR explicitly stated on X, Opus 4.5 might be able to regularly complete only tasks that take humans about two hours, or it might succeed on tasks that take humans as long as 20 hours. Given the uncertainties intrinsic to the method, it was impossible to know for sure. 

“There are a bunch of ways that people are reading too much into the graph,” says Sydney Von Arx, a member of METR’s technical staff.

More fundamentally, the METR plot does not measure AI abilities writ large, nor does it claim to. In order to build the graph, METR tests the models primarily on coding tasks, evaluating the difficulty of each by measuring or estimating how long it takes humans to complete it—a metric that not everyone accepts. Claude Opus 4.5 might be able to complete certain tasks that take humans five hours, but that doesn’t mean it’s anywhere close to replacing a human worker.

METR was founded to assess the risks posed by frontier AI systems. Though it is best known for the exponential trend plot, it has also worked with AI companies to evaluate their systems in greater detail and published several other independent research projects, including a widely covered July 2025 study suggesting that AI coding assistants might actually be slowing software engineers down. 

But the exponential plot has made METR’s reputation, and the organization appears to have a complicated relationship with that graph’s often breathless reception. In January, Thomas Kwa, one of the lead authors on the paper that introduced it, wrote a blog post responding to some criticisms and making clear its limitations, and METR is currently working on a more extensive FAQ document. But Kwa isn’t optimistic that these efforts will meaningfully shift the discourse. “I think the hype machine will basically, whatever we do, just strip out all the caveats,” he says.

Nevertheless, the METR team does think that the plot has something meaningful to say about the trajectory of AI progress. “You should absolutely not tie your life to this graph,” says Von Arx. “But also,” she adds, “I bet that this trend is gonna hold.”

Part of the trouble with the METR plot is that it’s quite a bit more complicated than it looks. The x-axis is simple enough: It tracks the date when each model was released. But the y-axis is where things get tricky. It records each model’s “time horizon,” an unusual metric that METR created—and that, according to Kwa and Von Arx, is frequently misunderstood.

To understand exactly what model time horizons are, it helps to know all the work that METR put into calculating them. First, the METR team assembled a collection of tasks ranging from quick multiple-choice questions to detailed coding challenges—all of which were somehow relevant to software engineering. Then they had human coders attempt most of those tasks and evaluated how long it took them to finish. In this way, they assigned the tasks a human baseline time. Some tasks took the experts mere seconds, whereas others required several hours.

When METR tested large language models on the task suite, they found that advanced models could complete the fast tasks with ease—but as the models attempted tasks that had taken humans more and more time to finish, their accuracy started to fall off. From a model’s performance, the researchers calculated the point on the time scale of human tasks at which the model would complete about 50% of the tasks successfully. That point is the model’s time horizon. 

All that detail is in the blog post and the academic paper that METR released along with the original time horizon plot. But the METR plot is frequently passed around on social media without this context, and so the true meaning of the time horizon metric can get lost in the shuffle. One common misapprehension is that the numbers on the plot’s y-axis—around five hours for Claude Opus 4.5, for example—represent the length of time that the models can operate independently. They do not. They represent how long it takes humans to complete tasks that a model can successfully perform.  Kwa has seen this error so frequently that he made a point of correcting it at the very top of his recent blog post, and when asked what information he would add to the versions of the plot circulating online, he said he would include the word “human” whenever the task completion time was mentioned.

As complex and widely misinterpreted as the time horizon concept might be, it does make some basic sense: A model with a one-hour time horizon could automate some modest portions of a software engineer’s job, whereas a model with a 40-hour horizon could potentially complete days of work on its own. But some experts question whether the amount of time that humans take on tasks is an effective metric for quantifying AI capabilities. “I don’t think it’s necessarily a given fact that because something takes longer, it’s going to be a harder task,” says Inioluwa Deborah Raji, a PhD student at UC Berkeley who studies model evaluation. 

Von Arx says that she, too, was originally skeptical that time horizon was the right measure to use. What convinced her was seeing the results of her and her colleagues’ analysis. When they calculated the 50% time horizon for all the major models available in early 2025 and then plotted each of them on the graph, they saw that the time horizons for the top-tier models were increasing over time—and, moreover, that the rate of advancement was speeding up. Every seven-ish months, the time horizon doubled, which means that the most advanced models could complete tasks that took humans nine seconds in mid 2020, 4 minutes in early 2023, and 40 minutes in late 2024. “I can do all the theorizing I want about whether or not it makes sense, but the trend is there,” Von Arx says.

It’s this dramatic pattern that made the METR plot such a blockbuster. Many people learned about it when they read AI 2027, a viral sci-fi story cum quantitative forecast positing that superintelligent AI could wipe out humanity by 2030. The writers of AI 2027 based some of their predictions on the METR plot and cited it extensively. In Von Arx’s words, “It’s a little weird when the way lots of people are familiar with your work is this pretty opinionated interpretation.”

Of course, plenty of people invoke the METR plot without imagining large-scale death and destruction. For some AI boosters, the exponential trend indicates that AI will soon usher in an era of radical economic growth. The venture capital firm Sequoia Capital, for example, recently put out a post titled “2026: This is AGI,” which used the METR plot to argue that AI that can act as an employee or contractor will soon arrive. “The provocation really was like, ‘What will you do when your plans are measured in centuries?’” says Sonya Huang, a general partner at Sequoia and one of the post’s authors. 

Just because a model achieves a one-hour time horizon on the METR plot, however, doesn’t mean that it can replace one hour of human work in the real world. For one thing, the tasks on which the models are evaluated don’t reflect the complexities and confusion of real-world work. In their original study, Kwa, Von Arx, and their colleagues quantify what they call the “messiness” of each task according to criteria such as whether the model knows exactly how it is being scored and whether it can easily start over if it makes a mistake (for messy tasks, the answer to both questions would be no). They found that models do noticeably worse on messy tasks, although the overall pattern of improvement holds for both messy and non-messy ones.

And even the messiest tasks that METR considered can’t provide much information about AI’s ability to take on most jobs, because the plot is based almost entirely on coding tasks. “A model can get better at coding, but it’s not going to magically get better at anything else,” says Daniel Kang, an assistant professor of computer science at the University of Illinois Urbana-Champaign. In a follow-up study, Kwa and his colleagues did find that time horizons for tasks in other domains also appear to be on exponential trajectories, but that work was much less formal.

Despite these limitations, many people admire the group’s research. “The METR study is one of the most carefully designed studies in the literature for this kind of work,” Kang told me. Even Gary Marcus, a former NYU professor and professional LLM curmudgeon, described much of the work that went into the plot as “terrific” in a blog post.

Some people will almost certainly continue to read the METR plot as a prognostication of our AI-induced doom, but in reality it’s something far more banal: a carefully constructed scientific tool that puts concrete numbers to people’s intuitive sense of AI progress. As METR employees will readily agree, the plot is far from a perfect instrument. But in a new and fast-moving domain, even imperfect tools can have enormous value.

“This is a bunch of people trying their best to make a metric under a lot of constraints. It is deeply flawed in many ways,” Von Arx says. “I also think that it is one of the best things of its kind.”

The previous article in this series, “Rules fail at the prompt, succeed at the boundary,” focused on the first AI-orchestrated espionage campaign and the failure of prompt-level control. This article is the prescription. The question every CEO is now getting from their board is some version of: What do we do about agent risk?

Across recent AI security guidance from standards bodies, regulators, and major providers, a simple idea keeps repeating: treat agents like powerful, semi-autonomous users, and enforce rules at the boundaries where they touch identity, tools, data, and outputs.

The following is an actionable eight-step plan one can ask teams to implement and report against:  

Constrain capabilities

These steps help define identity and limit capabilities.

1. Identity and scope: Make agents real users with narrow jobs

Today, agents run under vague, over-privileged service identities. The fix is straightforward: treat each agent as a non-human principal with the same discipline applied to employees.

Every agent should run as the requesting user in the correct tenant, with permissions constrained to that user’s role and geography. Prohibit cross-tenant on-behalf-of shortcuts. Anything high-impact should require explicit human approval with a recorded rationale. That is how Google’s Secure AI Framework (SAIF) and NIST AI’s access-control guidance are meant to be applied in practice.

The CEO question: Can we show, today, a list of our agents and exactly what each is allowed to do?

2. Tooling control: Pin, approve, and bound what agents can use

The Anthropic espionage framework worked because the attackers could wire Claude into a flexible suite of tools (e.g., scanners, exploit frameworks, data parsers) through Model Context Protocol, and those tools weren’t pinned or policy-gated.

The defense is to treat toolchains like a supply chain:

• Pin versions of remote tool servers.
• Require approvals for adding new tools, scopes, or data sources.
• Forbid automatic tool-chaining unless a policy explicitly allows it.

This is exactly what OWASP flags under excessive agency and what it recommends protecting against. Under the EU AI Act, designing for such cyber-resilience and misuse resistance is part of the Article 15 obligation to ensure robustness and cybersecurity.

The CEO question: Who signs off when an agent gains a new tool or a broader scope? How does one know?

3. Permissions by design: Bind tools to tasks, not to models

A common anti-pattern is to give the model a long-lived credential and hope prompts keep it polite. SAIF and NIST argue the opposite: credentials and scopes should be bound to tools and tasks, rotated regularly, and auditable. Agents then request narrowly scoped capabilities through those tools.

In practice, that looks like: “finance-ops-agent may read, but not write, certain ledgers without CFO approval.”

The CEO question: Can we revoke a specific capability from an agent without re-architecting the whole system?

Control data and behavior

These steps gate inputs, outputs, and constrain behavior.

4. Inputs, memory, and RAG: Treat external content as hostile until proven otherwise

Most agent incidents start with sneaky data: a poisoned web page, PDF, email, or repository that smuggles adversarial instructions into the system. OWASP’s prompt-injection cheat sheet and OpenAI’s own guidance both insist on strict separation of system instructions from user content and on treating unvetted retrieval sources as untrusted.

Operationally, gate before anything enters retrieval or long-term memory: new sources are reviewed, tagged, and onboarded; persistent memory is disabled when untrusted context is present; provenance is attached to each chunk.

The CEO question: Can we enumerate every external content source our agents learn from, and who approved them?

5. Output handling and rendering: Nothing executes “just because the model said so”

In the Anthropic case, AI-generated exploit code and credential dumps flowed straight into action. Any output that can cause a side effect needs a validator between the agent and the real world. OWASP’s insecure output handling category is explicit on this point, as are browser security best practices around origin boundaries.

The CEO question: Where, in our architecture, are agent outputs assessed before they run or ship to customers?

6. Data privacy at runtime: Protect the data first, then the model

Protect the data such that there is nothing dangerous to reveal by default. NIST and SAIF both lean toward “secure-by-default” designs where sensitive values are tokenized or masked and only re-hydrated for authorized users and use cases.

In agentic systems, that means policy-controlled detokenization at the output boundary and logging every reveal. If an agent is fully compromised, the blast radius is bounded by what the policy lets it see.

This is where the AI stack intersects not just with the EU AI Act but with GDPR and sector-specific regimes. The EU AI Act expects providers and deployers to manage AI-specific risk; runtime tokenization and policy-gated reveal are strong evidence that one is actively controlling those risks in production.

The CEO question: When our agents touch regulated data, is that protection enforced by architecture or by promises?

Prove governance and resilience

For the final steps, it’s important to show controls work and keep working.

7. Continuous evaluation: Don’t ship a one-time test, ship a test harness

Anthropic’s research about sleeper agents should eliminate all fantasies about single test dreams and show how critical continuous evaluation is. This means instrumenting agents with deep observability, regularly red teaming with adversarial test suites, and backing everything with robust logging and evidence, so failures become both regression tests and enforceable policy updates.

The CEO question: Who works to break our agents every week, and how do their findings change policy?

 8. Governance, inventory, and audit: Keep score in one place

AI security frameworks emphasize inventory and evidence: enterprises must know which models, prompts, tools, datasets, and vector stores they have, who owns them, and what decisions were taken about risk.

For agents, that means a living catalog and unified logs:

• Which agents exist, on which platforms
• What scopes, tools, and data each is allowed
• Every approval, detokenization, and high-impact action, with who approved it and when

The CEO question: If asked how an agent made a specific decision, could we reconstruct the chain?

And don’t forget the system-level threat model: assume the threat actor GTG-1002 is already in your enterprise. To complete enterprise preparedness, zoom out and consider the MITRE ATLAS product, which exists precisely because adversaries attack systems, not models. Anthropic provides a case study of a state-based threat actor (GTG-1002) doing exactly that with an agentic framework.

Taken together, these controls do not make agents magically safe. They do something more familiar and more reliable: they put AI, its access, and actions back inside the same security frame used for any powerful user or system.

For boards and CEOs, the question is no longer “Do we have good AI guardrails?” It’s: Can we answer the CEO questions above with evidence, not assurances?

This content was produced by Protegrity. It was not written by MIT Technology Review’s editorial staff.

This is today’s edition of The Download, our weekday newsletter that provides a daily dose of what’s going on in the world of technology.

Why AI companies are betting on next-gen nuclear

AI is driving unprecedented investment for massive data centers and an energy supply that can support its huge computational appetite. One potential source of electricity for these facilities is next-generation nuclear power plants, which could be cheaper to construct and safer to operate than their predecessors.

We recently held a subscriber-exclusive Roundtables discussion on hyperscale AI data centers and next-gen nuclear—two featured technologies on the MIT Technology Review 10 Breakthrough Technologies of 2026 list. You can watch the conversation back here, and don’t forget to subscribe to make sure you catch future discussions as they happen.

How social media encourages the worst of AI boosterism

Demis Hassabis, CEO of Google DeepMind, summed it up in three words: “This is embarrassing.”

Hassabis was replying on X to an overexcited post by Sébastien Bubeck, a research scientist at the rival firm OpenAI, announcing that two mathematicians had used OpenAI’s latest large language model, GPT-5, to find solutions to 10 unsolved problems in mathematics.

Put your math hats on for a minute, and let’s take a look at what this beef from mid-October was about. It’s a perfect example of what’s wrong with AI right now.

—Will Douglas Heaven

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

The paints, coatings, and chemicals making the world a cooler place

It’s getting harder to beat the heat. During the summer of 2025, heat waves knocked out power grids in North America, Europe, and the Middle East. Global warming means more people need air-­conditioning, which requires more power and strains grids.

But a millennia-old idea (plus 21st-century tech) might offer an answer: radiative cooling. Paints, coatings, and textiles can scatter sunlight and dissipate heat—no additional energy required. Read the full story.

—Becky Ferreira

This story is from the most recent print issue of MIT Technology Review magazine, which shines a light on the exciting innovations happening right now. If you haven’t already, subscribe now to receive future issues once they land.

MIT Technology Review Narrated: China figured out how to sell EVs. Now it has to deal with their aging batteries.

As early electric cars age out, hundreds of thousands of used batteries are flooding the market, fueling a gray recycling economy even as Beijing and big manufacturers scramble to build a more orderly system.

This is our latest story to be turned into a MIT Technology Review Narrated podcast, which we’re publishing each week on Spotify and Apple Podcasts. Just navigate to MIT Technology Review Narrated on either platform, and follow us to get all our new content as it’s released.

The must-reads

I’ve combed the internet to find you today’s most fun/important/scary/fascinating stories about technology.

1 Europe is edging closer towards banning social media for minors
Spain has become the latest country to consider it. (Bloomberg $)
+ Elon Musk called the Spanish prime minister a “tyrant” in retaliation. (The Guardian)
+ Other European nations considering restrictions include Greece, France and the UK. (Reuters)

2 Humans are infiltrating the social network for AI agents
It turns out role-playing as a bot is surprisingly fun. (Wired $)
+ Some of the most viral posts may actually be human-generated after all. (The Verge)

3 Russian spy spacecraft have intercepted Europe’s key satellites
Security officials are confident Moscow has tapped into unencrypted European comms. (FT $)

4 French authorities raided X’s Paris office
They’re investigating a range of potential charges against the company. (WSJ $)
+ Elon Musk has been summoned to give evidence in April. (Reuters)

5 Jeffrey Epstein invested millions into crypto startup Coinbase
Which suggests he was still able to take advantage of Silicon Valley investment opportunities years after pleading guilty to soliciting sex from an underage girl. (WP $)

6 A group of crypto bros paid $300,000 for a gold statue of Trump
It’s destined to be installed on his Florida golf complex, apparently. (NYT $)

7 OpenAI has appointed a “head of preparedness”
Dylan Scandinaro will earn a cool $555,000 for his troubles. (Bloomberg $)

8 The eternal promise of 3D-printed batteries
Traditional batteries are blocky and bulky. Printing them ourselves could help solve that. (IEEE Spectrum)

9 What snow can teach us about city design
When icy mounds refuse to melt, they show us what a less car-focused city could look like. (New Yorker $)
+ This startup thinks slime mold can help us design better cities. (MIT Technology Review)

10 Please don’t use AI to talk to your friends
That’s what your brain is for. (The Atlantic $)
+ Therapists are secretly using ChatGPT. Clients are triggered. (MIT Technology Review)

Quote of the day

“Today, our children are exposed to a space they were never meant to navigate alone. We will no longer accept that.”

—Spanish prime minister Pedro Sánchez proposes a social media ban for children aged under 16 in the country, following in Australia’s footsteps, AP News reports.

One more thing

A brain implant changed her life. Then it was removed against her will.

Sticking an electrode inside a person’s brain can do more than treat a disease. Take the case of Rita Leggett, an Australian woman whose experimental brain implant designed to help people with epilepsy changed her sense of agency and self.

Leggett told researchers that she “became one” with her device. It helped her to control the unpredictable, violent seizures she routinely experienced, and allowed her to take charge of her own life. So she was devastated when, two years later, she was told she had to remove the implant because the company that made it had gone bust.

The removal of this implant, and others like it, might represent a breach of human rights, ethicists say in a paper published earlier this month. And the issue will only become more pressing as the brain implant market grows in the coming years and more people receive devices like Leggett’s. Read the full story.

—Jessica Hamzelou

We can still have nice things

A place for comfort, fun and distraction to brighten up your day. (Got any ideas? Drop me a line or skeet ’em at me.)

+ Why Beethoven’s Ode to Joy is still such an undisputed banger.
+ Did you know that one of the world’s most famous prisons actually served as a zoo and menagerie for over 600 years?
+ Banana nut muffins sound like a fantastic way to start your day.
+ 2026 is shaping up to be a blockbuster year for horror films.

This is today’s edition of The Download, our weekday newsletter that provides a daily dose of what’s going on in the world of technology.

Microbes could extract the metal needed for cleantech

In a pine forest on Michigan’s Upper Peninsula, the only active nickel mine in the US is nearing the end of its life. At a time when carmakers want the metal for electric-vehicle batteries, nickel concentration at Eagle Mine is falling and could soon drop too low to warrant digging.

Demand for nickel, copper, and rare earth elements is rapidly increasing amid the explosive growth of metal-intensive data centers, electric cars, and renewable energy projects. But producing these metals is becoming harder and more expensive because miners have already exploited the best resources. Here’s how biotechnology could help.

—Matt Blois

What we’ve been getting wrong about AI’s truth crisis

—James O’Donnell

What would it take to convince you that the era of truth decay we were long warned about—where AI content dupes us, shapes our beliefs even when we catch the lie, and erodes societal trust in the process—is now here?

A story I published last week pushed me over the edge. And it also made me realize that the tools we were sold as a cure for this crisis are failing miserably. Read the full story.

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

TR10: Hyperscale AI data centers

In sprawling stretches of farmland and industrial parks, supersized buildings packed with racks of computers are springing up to fuel the AI race.

These engineering marvels are a new species of infrastructure: supercomputers designed to train and run large language models at mind-­bending scale, complete with their own specialized chips, cooling systems, and even energy supplies. But all that impressive computing power comes at a cost.

Read why we’ve named hyperscale AI data centers as of our 10 Breakthrough Technologies this year, and check out the rest of the list.

The must-reads

I’ve combed the internet to find you today’s most fun/important/scary/fascinating stories about technology.

1 Elon Musk’s SpaceX has acquired xAI
The deal values the combined companies at a cool $1.25 trillion. (WSJ $)
+ It also paves the way for SpaceX to offer an IPO later this year. (WP $)
+ Meanwhile, OpenAI has accused xAI of destroying legal evidence. (Bloomberg $)

2 NASA has delayed the launch of Artemis II
It’s been pushed back to March due to the discovery of a hydrogen leak. (Ars Technica)
+ The rocket’s predecessor was also plagued by fuel leaks. (Scientific American)

3 Russia is hiring a guerilla youth army online
They’re committing arson and spying on targets across Europe. (New Yorker $)

4 Grok is still generating undressed images of men
Weeks after the backlash over it doing the same to women. (The Verge)
+ How Grok descended into becoming a porn generator. (WP $)
+ Inside the marketplace powering bespoke AI deepfakes of real women. (MIT Technology Review)

5 OpenAI is searching for alternatives to Nvidia’s chips
It’s reported to be unhappy about the speed at which it powers ChatGPT. (Reuters)

6 The latest attempt to study a notoriously unstable glacier has failed
Scientists lost their equipment within Antarctica’s Thwaites Glacier over the weekend. (NYT $)
+ Inside a new quest to save the “doomsday glacier” (MIT Technology Review)

7 The world is trying to wean itself off American technology
Governments are growing increasingly uneasy about their reliance on the US. (Rest of World)

8 AI’s sloppy writing is driving demand for real human writers
Long may it continue. (Insider $)

9 This female-dominated fitness community hates Mark Zuckerberg
His decision to shut down three VR studios means their days of playing their favorite workout game are numbered. (The Verge)
+ Welcome to the AI gym staffed by virtual trainers. (MIT Technology Review)

10 This cemetery has an eco-friendly solution for its overcrowding problem
If you’re okay with your loved one becoming gardening soil, that is. (WSJ $)
+ Why America is embracing the right to die now. (Economist $)
+ What happens when you donate your body to science. (MIT Technology Review)

Quote of the day

“In the long term, space-based AI is obviously the only way to scale…I mean, space is called ‘space’ for a reason.”

—Elon Musk explains his rationale for combining SpaceX with xAI in a blog post.

One more thing

On the ground in Ukraine’s largest Starlink repair shop

Starlink is absolutely critical to Ukraine’s ability to continue in the fight against Russia. It’s how troops in battle zones stay connected with faraway HQs; it’s how many of the drones essential to Ukraine’s survival hit their targets; it’s even how soldiers stay in touch with spouses and children back home.

However, Donald Trump’s fickle foreign policy and reports suggesting Elon Musk might remove Ukraine’s access to the services have cast the technology’s future in the country into doubt.

For now Starlink access largely comes down to the unofficial community of users and engineers, including the expert “Dr. Starlink”—famous for his creative ways of customizing the systems—who have kept Ukraine in the fight, both on and off the front line. He gave MIT Technology Review exclusive access to his unofficial Starlink repair workshop in the city of Lviv. Read the full story.

—Charlie Metcalfe

We can still have nice things

A place for comfort, fun and distraction to brighten up your day. (Got any ideas? Drop me a line or skeet ’em at me.)

+ The Norwegian countryside sure looks beautiful.
+ Quick—it’s time to visit these food destinations before the TikTok hordes descend.
+ Rest in power Catherine O’Hara, our favorite comedy queen.
+ Take some time out of your busy day to read a potted history of boats

In a pine forest on Michigan’s Upper Peninsula, the only active nickel mine in the US is nearing the end of its life. At a time when carmakers want the metal for electric-vehicle batteries, nickel concentration at Eagle Mine is falling and could soon drop too low to warrant digging.

But earlier this year, the mine’s owner started testing a new process that could eke out a bit more nickel. In a pair of shipping containers recently installed at the mine’s mill, a fermentation-derived broth developed by the startup Allonnia is mixed with concentrated ore to capture and remove impurities. The process allows nickel production from lower-quality ore. 

Kent Sorenson, Allonnia’s chief technology officer, says this approach could help companies continue operating sites that, like Eagle Mine, have burned through their best ore. “The low-hanging fruit is to keep mining the mines that we have,” he says. 

Demand for nickel, copper, and rare earth elements is rapidly increasing amid the explosive growth of metal-intensive data centers, electric cars, and renewable energy projects. But producing these metals is becoming harder and more expensive because miners have already exploited the best resources. Like the age-old technique of rolling up the end of a toothpaste tube, Allonnia’s broth is one of a number of ways that biotechnology could help miners squeeze more metal out of aging mines, mediocre ore, or piles of waste.

The mining industry has intentionally seeded copper ore with microbes for decades. At current copper bioleaching sites, miners pile crushed copper ore into heaps and add sulfuric acid. Acid-loving bacteria like Acidithiobacillus ferrooxidans colonize the mound. A chemical the organisms produce breaks the bond between sulfur and copper molecules to liberate the metal.

Until now, beyond maintaining the acidity and blowing air into the heap, there wasn’t much more miners could do to encourage microbial growth. But Elizabeth Dennett, CEO of the startup Endolith, says the decreasing cost of genetic tools is making it possible to manage the communities of microbes in a heap more actively. “The technology we’re using now didn’t exist a few years ago,” she says.

Endolith analyzes bits of DNA and RNA in the copper-rich liquid that flows out of an ore heap to characterize the microbes living inside. Combined with a suite of chemical analyses, the information helps the company determine which microbes to sprinkle on a heap to optimize extraction. 

In lab tests on ore from the mining firm BHP, Endolith’s active techniques outperformed passive bioleaching approaches. In November, the company raised $16.5 million to move from its Denver lab to heaps in active mines.

Despite these promising early results, Corale Brierley, an engineer who has worked on metal bioleaching systems since the 1970s, questions whether companies like Endolith that add additional microbes to ore will successfully translate their processes to commercial scales. “What guarantees are you going to give the company that those organisms will actually grow?” Brierley asks.

Big mining firms that have already optimized every hose, nut, and bolt in their process won’t be easy to convince either, says Diana Rasner, an analyst covering mining technology for the research firm Cleantech Group. 

“They are acutely aware of what it takes to scale these technologies because they know the industry,” she says. “They’ll be your biggest supporters, but they’re going to be your biggest critics.”

In addition to technical challenges, Rasner points out that venture-capital-backed biotechnology startups will struggle to deliver the quick returns their investors seek. Mining companies want lots of data before adopting a new process, which could take years of testing to compile. “This is not software,” Rasner says.  

Nuton, a subsidiary of the mining giant Rio Tinto, is a good example. The company has been working for decades on a copper bioleaching process that uses a blend of archaea and bacteria strains, plus some chemical additives. But it started demonstrating the technology only late last year, at a mine in Arizona. 

While Endolith and Nuton use naturally occurring microbes, the startup 1849 is hoping to achieve a bigger performance boost by genetically engineering microbes.

“You can do what mining companies have traditionally done,” says CEO Jai Padmakumar. “Or you can try to take the moonshot bet and engineer them. If you get that, you have a huge win.”

Genetic engineering would allow 1849 to tailor its microbes to the specific challenges facing a customer. But engineering organisms can also make them harder to grow, warns Buz Barstow, a Cornell University microbiologist who studies applications for biotechnology in mining.

Other companies are trying to avoid that trade-off by applying the products of microbial fermentation, rather than live organisms. Alta Resource Technologies, which closed a $28 million investment round in December, is engineering microbes that make proteins capable of extracting and separating rare earth elements. Similarly, the startup REEgen, based in Ithaca, New York, relies on the organic acids produced by an engineered strain of Gluconobacter oxydans to extract rare earth elements from ore and from waste materials like metal recycling slag, coal ash, or old electronics. “The microbes are the manufacturing,” says CEO Alexa Schmitz, an alumna of Barstow’s lab.

To make a dent in the growing demand for metal, this new wave of biotechnologies will have to go beyond copper and gold, says Barstow. In 2024, he started a project to map out genes that could be useful for extracting and separating a wider range of metals. Even with the challenges ahead, he says, biotechnology has the potential to transform mining the way fracking changed natural gas. “Biomining is one of these areas where the need … is big enough,” he says. 

The challenge will be moving fast enough to keep up with growing demand.

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

What would it take to convince you that the era of truth decay we were long warned about—where AI content dupes us, shapes our beliefs even when we catch the lie, and erodes societal trust in the process—is now here? A story I published last week pushed me over the edge. It also made me realize that the tools we were sold as a cure for this crisis are failing miserably. 

On Thursday, I reported the first confirmation that the US Department of Homeland Security, which houses immigration agencies, is using AI video generators from Google and Adobe to make content that it shares with the public. The news comes as immigration agencies have flooded social media with content to support President Trump’s mass deportation agenda—some of which appears to be made with AI (like a video about “Christmas after mass deportations”).

But I received two types of reactions from readers that may explain just as much about the epistemic crisis we’re in. 

One was from people who weren’t surprised, because on January 22 the White House had posted a digitally altered photo of a woman arrested at an ICE protest, one that made her appear hysterical and in tears. Kaelan Dorr, the White House’s deputy communications director, did not respond to questions about whether the White House altered the photo but wrote, “The memes will continue.”

The second was from readers who saw no point in reporting that DHS was using AI to edit content shared with the public, because news outlets were apparently doing the same. They pointed to the fact that the news network MS Now (formerly MSNBC) shared an image of Alex Pretti that was AI-edited and appeared to make him look more handsome, a fact that led to many viral clips this week, including one from Joe Rogan’s podcast. Fight fire with fire, in other words? A spokesperson for MS Now told Snopes that the news outlet aired the image without knowing it was edited.

There is no reason to collapse these two cases of altered content into the same category, or to read them as evidence that truth no longer matters. One involved the US government sharing a clearly altered photo with the public and declining to answer whether it was intentionally manipulated; the other involved a news outlet airing a photo it should have known was altered but taking some steps to disclose the mistake.

What these reactions reveal instead is a flaw in how we were collectively preparing for this moment. Warnings about the AI truth crisis revolved around a core thesis: that not being able to tell what is real will destroy us, so we need tools to independently verify the truth. My two grim takeaways are that these tools are failing, and that while vetting the truth remains essential, it is no longer capable on its own of producing the societal trust we were promised.

For example, there was plenty of hype in 2024 about the Content Authenticity Initiative, cofounded by Adobe and adopted by major tech companies, which would attach labels to content disclosing when it was made, by whom, and whether AI was involved. But Adobe applies automatic labels only when the content is wholly AI-generated. Otherwise the labels are opt-in on the part of the creator.

And platforms like X, where the altered arrest photo was posted, can strip content of such labels anyway (a note that the photo was altered was added by users). Platforms can also simply not choose to show the label at all.

Noticing how much traction the White House’s photo got even after it was shown to be AI-altered, I was struck by the findings of a very relevant new paper published in the journal Communications Psychology. In the study, participants watched a deepfake “confession” to a crime, and the researchers found that even when they were told explicitly that the evidence was fake, participants relied on it when judging an individual’s guilt. In other words, even when people learn that the content they’re looking at is entirely fake, they remain emotionally swayed by it. 

“Transparency helps, but it isn’t enough on its own,” the disinformation expert Christopher Nehring wrote recently about the study’s findings. “We have to develop a new masterplan of what to do about deepfakes.”

AI tools to generate and edit content are getting more advanced, easier to operate, and cheaper to run—all reasons why the US government is increasingly paying to use them. We were well warned of this, but we responded by preparing for a world in which the main danger was confusion. What we’re entering instead is a world in which influence survives exposure, doubt is easily weaponized, and establishing the truth does not serve as a reset button. And the defenders of truth are already trailing way behind.

Update: This story was updated on February 2 with details about how Adobe applies its content authenticity labels. A previous version of this story said content credentials were not visible on the Pentagon’s DVIDS website. The labels are present but require clicking through and hovering on individual images. The reference has been removed.

Many organizations rushed into generative AI, only to see pilots fail to deliver value. Now, companies want measurable outcomes—but how do you design for success?

At Mistral AI, we partner with global industry leaders to co-design tailored AI solutions that solve their most difficult problems. Whether it’s increasing CX productivity with Cisco, building a more intelligent car with Stellantis, or accelerating product innovation with ASML, we start with open frontier models and customize AI systems to deliver impact for each company’s unique challenges and goals.

Our methodology starts by identifying an iconic use case, the foundation for AI transformation that sets the blueprint for future AI solutions. Choosing the right use case can mean the difference between true transformation and endless tinkering and testing.

Identifying an iconic use case

Mistral AI has four criteria that we look for in a use case: strategic, urgent, impactful, and feasible.

First, the use case must be strategically valuable, addressing a core business process or a transformative new capability. It needs to be more than an optimization; it needs to be a gamechanger. The use case needs to be strategic enough to excite an organization’s C-suite and board of directors.

For example, use cases like an internal-facing HR chatbot are nice to have, but they are easy to solve and are not enabling any new innovation or opportunities. On the other end of the spectrum, imagine an externally facing banking assistant that can not only answer questions, but also help take actions like blocking a card, placing trades, and suggesting upsell/cross-sell opportunities. This is how a customer-support chatbot is turned into a strategic revenue-generating asset.

Second, the best use case to move forward with should be highly urgent and solve a business-critical problem that people care about right now. This project will take time out of people’s days—it needs to be important enough to justify that time investment. And it needs to help business users solve immediate pain points.

Third, the use case should be pragmatic and impactful. From day one, our shared goal with our customers is to deploy into a real-world production environment to enable testing the solution with real users and gather feedback. Many AI prototypes end up in the graveyard of fancy demos that are not good enough to put in front of customers, and without any scaffolding to evaluate and improve. We work with customers to ensure prototypes are stable enough to release, and that they have the necessary support and governance frameworks.

Finally, the best use case is feasible. There may be several urgent projects, but choosing one that can deliver a quick return on investment helps to maintain the momentum needed to continue and scale.

This means looking for a project that can be in production within three months—and a prototype can be live within a few weeks. It’s important to get a prototype in front of end users as fast as possible to get feedback to make sure the project is on track, and pivot as needed.

Where use cases fall short

Enterprises are complex, and the path forward is not usually obvious. To weed through all the possibilities and uncover the right first use case, Mistral AI will run workshops with our customers, hand-in-hand with subject-matter experts and end users.

Representatives from different functions will demo their processes and discuss business cases that could be candidates for a first use case—and together we agree on a winner. Here are some examples of types of projects that don’t qualify.

Moonshots: Ambitious bets that excite leadership but lack a path to quick ROI. While these projects can be strategic and urgent, they rarely meet the feasibility and impact requirements.

Future investments: Long-term plays that can wait. While these projects can be strategic and feasible, they rarely meet the urgency and impact requirements.

Tactical fixes: Firefighting projects that solve immediate pain but don’t move the needle. While these cases can be urgent and feasible, they rarely meet the strategy and impact requirements.

Quick wins: Useful for building momentum, but not transformative. While they can be impactful and feasible, they rarely meet the strategy and urgency requirements.

Blue sky ideas: These projects are gamechangers, but they need maturity to be viable. While they can be strategic and impactful, they rarely meet the urgency and feasibility requirements.

Hero projects: These are high-pressure initiatives that lack executive sponsorship or realistic timelines. While they can be urgent and impactful, they rarely meet the strategy and feasibility requirements.

Moving from use case to deployment

Once a clearly defined and strategic use case ready for development is identified, it’s time to move into the validation phase. This means doing an initial data exploration and data mapping, identifying a pilot infrastructure, and choosing a target deployment environment.

This step also involves agreeing on a draft pilot scope, identifying who will participate in the proof of concept, and setting up a governance process.

Once this is complete, it’s time to move into the building phase. Companies that partner with Mistral work with our in-house applied AI scientists who build our frontier models. We work together to design, build, and deploy the first solution.

During this phase, we focus on co-creation, so we can transfer knowledge and skills to the organizations we’re partnering with. That way, they can be self-sufficient far into the future. The output of this phase is a deployed AI solution with empowered teams capable of independent operation and innovation.

The first step is everything

After the first win, it’s imperative to use the momentum and learnings from the iconic use case to identify more high-value AI solutions to roll out. Success is when we have a scalable AI transformation blueprint with multiple high-value solutions across the organization.

But none of this could happen without successfully identifying that first iconic use case. This first step is not just about selecting a project—it’s about setting the foundation for your entire AI transformation.

It’s the difference between scattered experiments and a strategic, scalable journey toward impact. At Mistral AI, we’ve seen how this approach unlocks measurable value, aligns stakeholders, and builds momentum for what comes next.

The path to AI success starts with a single, well-chosen use case: one that is bold enough to inspire, urgent enough to demand action, and pragmatic enough to deliver.

This content was produced by Mistral AI. It was not written by MIT Technology Review’s editorial staff.

This is today’s edition of The Download, our weekday newsletter that provides a daily dose of what’s going on in the world of technology.

Inside the marketplace powering bespoke AI deepfakes of real women

Civitai—an online marketplace for buying and selling AI-generated content, backed by the venture capital firm Andreessen Horowitz—is letting users buy custom instruction files for generating celebrity deepfakes. Some of these files were specifically designed to make pornographic images banned by the site, a new analysis has found.

The study, from researchers at Stanford and Indiana University, looked at people’s requests for content on the site, called “bounties.” The researchers found that between mid-2023 and the end of 2024, most bounties asked for animated content—but a significant portion were for deepfakes of real people, and 90% of these deepfake requests targeted women. Read the full story.

—James O’Donnell

What’s next for EV batteries in 2026

Demand for electric vehicles and the batteries that power them has never been hotter.

In 2025, EVs made up over a quarter of new vehicle sales globally, up from less than 5% in 2020. Some regions are seeing even higher uptake: In China, more than 50% of new vehicle sales last year were battery electric or plug-in hybrids. In Europe, more purely electric vehicles hit the roads in December than gas-powered ones. (The US is the notable exception here, dragging down the global average with a small sales decline from 2024.)

As EVs become increasingly common on the roads, the battery world is growing too. Here’s what’s coming next for EV batteries in 2026 and beyond.

—Casey Crownhart

This story is part of MIT Technology Review’s What’s Next series, which examines industries, trends, and technologies to give you a first look at the future. You can read the rest of them here.

TR10: Base-edited baby

Kyle “KJ” Muldoon Jr. was born with a rare, potentially fatal genetic disorder that left his body unable to remove toxic ammonia from his blood. The University of Pennsylvania offered his parents an alternative to a liver transplant: gene-editing therapies.

The team set to work developing a tailored treatment using base editing—a form of CRISPR that can correct genetic “misspellings” by changing single bases, the basic units of DNA. KJ received an initial low dose when he was seven months old, and later received two higher doses. Today, KJ is doing well. At an event in October last year, his happy parents described how he was meeting all his developmental milestones.

Others have received gene-editing therapies intended to treat conditions including sickle cell disease and a predisposition to high cholesterol. But KJ was the first to receive a personalized treatment—one that was designed just for him and will probably never be used again. Read why we made it one of our 10 Breakthrough Technologies this year, and check out the rest of the list.

The must-reads

I’ve combed the internet to find you today’s most fun/important/scary/fascinating stories about technology.

1 A social network for AI agents is vulnerable to abuse
A misconfiguration meant anyone could take control of any agent. (404 Media)
+ Moltbook is loosely modeled on Reddit, but humans are unable to post. (FT $)

2 Google breached its own ethics rules to help an Israeli contractor
It helped a military worker to analyze drone footage, a whistleblower has claimed. (WP $)

3 Capgemini is selling its unit linked to ICE
After the French government asked it to clarify its work for the agency. (Bloomberg $) 
+ The company has signed $12.2mn in contracts under the Trump administration. (FT $)
+ Here’s how to film ICE activities as safely as possible. (Wired $)

4 China has a plan to prime its next generation of AI experts 
Thanks to its elite genius class system. (FT $)
+ The country is going all-in on AI healthcare. (Rest of World)
+ The State of AI: Is China about to win the race? (MIT Technology Review)

5 Indonesia has reversed its ban on xAI’s Grok
After it announced plans to improve its compliance with the country’s laws. (Reuters)
+ Indonesia maintains a strict stance against pornographic content. (NYT $)
+ Malaysia and the Philippines have also lifted bans on the chatbot. (TechCrunch)

6 Don’t expect to hitch a ride on a Blue Origin rocket anytime soon
Jeff Bezos’ venture won’t be taking tourists into space for at least two years. (NYT $)
+ Artemis II astronauts are due to set off for the moon soon. (IEEE Spectrum)
+ Commercial space stations are on our list of 10 Breakthrough Technologies for 2026. (MIT Technology Review)

7 America’s push for high-speed internet is under threat
There aren’t enough skilled workers to meet record demand. (WSJ $)

8 Can AI help us grieve better?
A growing cluster of companies are trying to find out. (The Atlantic $)
+ Technology that lets us “speak” to our dead relatives has arrived. Are we ready? (MIT Technology Review)

9 How to fight future insect infestations
A certain species of fungus could play a key role. (Ars Technica)
+ How do fungi communicate? (MIT Technology Review)

10 What a robot-made latte tastes like, according to a former barista
Damn fine, apparently. (The Verge)

Quote of the day

 “It feels like a wild bison rampaging around in my computer.”

—A user who signed up to AI agent Moltbot remarks on the bot’s unpredictable behavior, Rest of World reports.

One more thing

How Wi-Fi sensing became usable tech

Wi-Fi sensing is a tantalizing concept: that the same routers bringing you the internet could also detect your movements. But, as a way to monitor health, it’s mostly been eclipsed by other technologies, like ultra-wideband radar. 

Despite that, Wi-Fi sensing hasn’t gone away. Instead, it has quietly become available in millions of homes, supported by leading internet service providers, smart-home companies, and chip manufacturers.

Soon it could be invisibly monitoring our day-to-day movements for all sorts of surprising—and sometimes alarming—purposes. Read the full story. 

—Meg Duff

We can still have nice things

A place for comfort, fun and distraction to brighten up your day. (Got any ideas? Drop me a line or skeet ’em at me.)

+ These intrepid Scottish bakers created the largest ever Empire biscuit (a classic shortbread cookie covered in icing)
+ My, what big tentacles you have!
+ If you’ve been feeling like you’re stuck in a rut lately, this advice could be exactly what you need to overcome it.
+ These works of psychedelic horror are guaranteed to send a shiver down your spine.

MIT Technology Review’s What’s Next series looks across industries, trends, and technologies to give you a first look at the future. You can read the rest of them here.

Demand for electric vehicles and the batteries that power them has never been hotter.

In 2025, EVs made up over a quarter of new vehicle sales globally, up from less than 5% in 2020. Some regions are seeing even higher uptake: In China, more than 50% of new vehicle sales last year were battery electric or plug-in hybrids. In Europe, more purely electric vehicles hit the roads in December than gas-powered ones. (The US is the notable exception here, dragging down the global average with a small sales decline from 2024.)

As EVs become increasingly common on the roads, the battery world is growing too. Looking ahead, we could soon see wider adoption of new chemistries, including some that deliver lower costs or higher performance. Meanwhile, the geopolitics of batteries are shifting, and so is the policy landscape. Here’s what’s coming next for EV batteries in 2026 and beyond.

A big opportunity for sodium-ion batteries

Lithium-ion batteries are the default chemistry used in EVs, personal devices, and even stationary storage systems on the grid today. But in a tough environment in some markets like the US, there’s a growing interest in cheaper alternatives. Automakers right now largely care just about batteries’ cost, regardless of performance improvements, says Kara Rodby, a technical principal at Volta Energy Technologies, a venture capital firm that focuses on energy storage technology.

Sodium-ion cells have long been held up as a potentially less expensive alternative to lithium. The batteries are limited in their energy density, so they deliver a shorter range than lithium-ion. But sodium is also more abundant, so they could be cheaper.

Sodium’s growth has been cursed, however, by the very success of lithium-based batteries, says Shirley Meng, a professor of molecular engineering at the University of Chicago. A lithium-ion battery cell cost $568 per kilowatt-hour in 2013, but that cost had fallen to just $74 per kilowatt-hour by 2025—quite the moving target for cheaper alternatives to chase.

Sodium-ion batteries currently cost about $59 per kilowatt-hour on average. That’s less expensive than the average lithium-ion battery. But if you consider only lithium iron phosphate (LFP) cells, a lower-end type of lithium-ion battery that averages $52 per kilowatt-hour, sodium is still more expensive today. 

We could soon see an opening for sodium-batteries, though. Lithium prices have been ticking up in recent months, a shift that could soon slow or reverse the steady downward march of prices for lithium-based batteries. 

Sodium-ion batteries are already being used commercially, largely for stationary storage on the grid. But we’re starting to see sodium-ion cells incorporated into vehicles, too. The Chinese companies Yadea, JMEV, and HiNa Battery have all started producing sodium-ion batteries in limited numbers for EVs, including small, short-range cars and electric scooters that don’t require a battery with high energy density. CATL, a Chinese battery company that’s the world’s largest, says it recently began producing sodium-ion cells. The company plans to launch its first EV using the chemistry by the middle of this year. 

Today, both production and demand for sodium-ion batteries are heavily centered in China. That’s likely to continue, especially after a cutback in tax credits and other financial support for the battery and EV industries in the US. One of the biggest sodium-battery companies in the US, Natron, ceased operations last year after running into funding issues.

We could also see progress in sodium-ion research: Companies and researchers are developing new materials for components including the electrolyte and electrodes, so the cells could get more comparable to lower-end lithium-ion cells in terms of energy density, Meng says. 

Major tests for solid-state batteries

As we enter the second half of this decade, many eyes in the battery world are on big promises and claims about solid-state batteries.

These batteries could pack more energy into a smaller package by removing the liquid electrolyte, the material that ions move through when a battery is charging and discharging. With a higher energy density, they could unlock longer-range EVs.

Companies have been promising solid-state batteries for years. Toyota, for example, once planned to have them in vehicles by 2020. That timeline has been delayed several times, though the company says it’s now on track to launch the new cells in cars in 2027 or 2028.

Historically, battery makers have struggled to produce solid-state batteries at the scale needed to deliver a commercially relevant supply for EVs. There’s been progress in manufacturing techniques, though, and companies could soon actually make good on their promises, Meng says. 

Factorial Energy, a US-based company making solid-state batteries, provided cells for a Mercedes test vehicle that drove over 745 miles on a single charge in a real-world test in September. The company says it plans to bring its tech to market as soon as 2027. Quantumscape, another major solid-state player in the US, is testing its cells with automotive partners and plans to have its batteries in commercial production later this decade.  

Before we see true solid-state batteries, we could see hybrid technologies, often referred to as semi-solid-state batteries. These commonly use materials like gel electrolytes, reducing the liquid inside cells without removing it entirely. Many Chinese companies are looking to build semi-solid-state batteries before transitioning to entirely solid-state ones, says Evelina Stoikou, head of battery technologies and supply chains at BloombergNEF, an energy consultancy.

A global patchwork

The picture for the near future of the EV industry looks drastically different depending on where you’re standing.

Last year, China overtook Japan as the country with the most global auto sales. And more than one in three EVs made in 2025 had a CATL battery in it. Simply put, China is dominating the global battery industry, and that doesn’t seem likely to change anytime soon.

China’s influence outside its domestic market is growing especially quickly. CATL is expected to begin production this year at its second European site; the factory, located in Hungary, is an $8.2 billion project that will supply automakers including BMW and the Mercedes-Benz group. Canada recently signed a deal that will lower the import tax on Chinese EVs from 100% to roughly 6%, effectively opening the Canadian market for Chinese EVs.

Some countries that haven’t historically been major EV markets could become bigger players in the second half of the decade. Annual EV sales in Thailand and Vietnam, where the market was virtually nonexistent just a few years ago, broke 100,000 in 2025. Brazil, in particular, could see its new EV sales more than double in 2026 as major automakers including Volkswagen and BYD set up or ramp up production in the country. 

On the flip side, EVs are facing a real test in 2026 in the US, as this will be the first calendar year after the sunset of federal tax credits that were designed to push more drivers to purchase the vehicles. With those credits gone, growth in sales is expected to continue lagging. 

One bright spot for batteries in the US is outside the EV market altogether. Battery manufacturers are starting to produce low-cost LFP batteries in the US, largely for energy storage applications. LG opened a massive factory to make LFP batteries in mid-2025 in Michigan, and the Korean battery company SK On plans to start making LFP batteries at its facility in Georgia later this year. Those plants could help battery companies cash in on investments as the US EV market faces major headwinds. 

Even as the US lags behind, the world is electrifying transportation. By 2030, 40% of new vehicles sold around the world are projected to be electric. As we approach that milestone, expect to see more global players, a wider selection of EVs, and an even wider menu of batteries to power them. 

Civitai—an online marketplace for buying and selling AI-generated content, backed by the venture capital firm Andreessen Horowitz—is letting users buy custom instruction files for generating celebrity deepfakes. Some of these files were specifically designed to make pornographic images banned by the site, a new analysis has found.

The study, from researchers at Stanford and Indiana University, looked at people’s requests for content on the site, called “bounties.” The researchers found that between mid-2023 and the end of 2024, most bounties asked for animated content—but a significant portion were for deepfakes of real people, and 90% of these deepfake requests targeted women. (Their findings have not yet been peer reviewed.)

The debate around deepfakes, as illustrated by the recent backlash to explicit images on the X-owned chatbot Grok, has revolved around what platforms should do to block such content. Civitai’s situation is a little more complicated. Its marketplace includes actual images, videos, and models, but it also lets individuals buy and sell instruction files called LoRAs that can coach mainstream AI models like Stable Diffusion into generating content they were not trained to produce. Users can then combine these files with other tools to make deepfakes that are graphic or sexual. The researchers found that 86% of deepfake requests on Civitai were for LoRAs.

In these bounties, users requested “high quality” models to generate images of public figures like the influencer Charli D’Amelio or the singer Gracie Abrams, often linking to their social media profiles so their images could be grabbed from the web. Some requests specified a desire for models that generated the individual’s entire body, accurately captured their tattoos, or allowed hair color to be changed. Some requests targeted several women in specific niches, like artists who record ASMR videos. One request was for a deepfake of a woman said to be the user’s wife. Anyone on the site could offer up AI models they worked on for the task, and the best submissions received payment—anywhere from $0.50 to $5. And nearly 92% of the deepfake bounties were awarded.

Neither Civitai nor Andreessen Horowitz responded to requests for comment.

It’s possible that people buy these LoRAs to make deepfakes that aren’t sexually explicit (though they’d still violate Civitai’s terms of use, and they’d still be ethically fraught). But Civitai also offers educational resources on how to use external tools to further customize the outputs of image generators—for example, by changing someone’s pose. The site also hosts user-written articles with details on how to instruct models to generate pornography. The researchers found that the amount of porn on the platform has gone up, and that the majority of requests each week are now for NSFW content.

“Not only does Civitai provide the infrastructure that facilitates these issues; they also explicitly teach their users how to utilize them,” says Matthew DeVerna, a postdoctoral researcher at Stanford’s Cyber Policy Center and one of the study’s leaders. 

The company used to ban only sexually explicit deepfakes of real people, but in May 2025 it announced it would ban all deepfake content. Nonetheless, countless requests for deepfakes submitted before this ban now remain live on the site, and many of the winning submissions fulfilling those requests remain available for purchase, MIT Technology Review confirmed.

“I believe the approach that they’re trying to take is to sort of do as little as possible, such that they can foster as much—I guess they would call it—creativity on the platform,” DeVerna says.

Users buy LoRAs with the site’s online currency, called Buzz, which is purchased with real money. In May 2025, Civita’s credit card processor cut off the company because of its ongoing problem with nonconsensual content. To pay for explicit content, users must now use gift cards or cryptocurrency to buy Buzz; the company offers a different scrip for non-explicit content. 

Civitai automatically tags bounties requesting deepfakes and lists a way for the person featured in the content to manually request its takedown. This system means that Civitai has a reasonably successful way of knowing which bounties are for deepfakes, but it’s still leaving moderation to the general public rather than carrying it out proactively. 

A company’s legal liability for what its users do isn’t totally clear. Generally, tech companies have broad legal protections against such liability for their content under Section 230 of the Communications Decency Act, but those protections aren’t limitless. For example, “you cannot knowingly facilitate illegal transactions on your website,” says Ryan Calo, a professor specializing in technology and AI at the University of Washington’s law school. (Calo wasn’t involved in this new study.)

Civitai joined OpenAI, Anthropic, and other AI companies in 2024 in adopting design principles to guard against the creation and spread of AI-generated child sexual abuse material . This move followed a 2023 report from the Stanford Internet Observatory, which found that the vast majority of AI models named in child sexual abuse communities were Stable Diffusion–based models “predominantly obtained via Civitai.”

But adult deepfakes have not gotten the same level of attention from content platforms or the venture capital firms that fund them. “They are not afraid enough of it. They are overly tolerant of it,” Calo says. “Neither law enforcement nor civil courts adequately protect against it. It is night and day.”

Civitai received a $5 million investment from Andreessen Horowitz (a16z) in November 2023. In a video shared by a16z, Civitai cofounder and CEO Justin Maier described his goal of building the main place where people find and share AI models for their own individual purposes. “We’ve aimed to make this space that’s been very, I guess, niche and engineering-heavy more and more approachable to more and more people,” he said. 

Civitai is not the only company with a deepfake problem in a16z’s investment portfolio; in February, MIT Technology Review first reported that another company, Botify AI, was hosting AI companions resembling real actors that stated their age as under 18, engaged in sexually charged conversations, offered “hot photos,” and in some instances described age-of-consent laws as “arbitrary” and “meant to be broken.”

This is today’s edition of The Download, our weekday newsletter that provides a daily dose of what’s going on in the world of technology.

DHS is using Google and Adobe AI to make videos

The news: The US Department of Homeland Security is using AI video generators from Google and Adobe to make and edit content shared with the public, a new document reveals. The document, released on Wednesday, provides an inventory of which commercial AI tools DHS uses for tasks ranging from generating drafts of documents to managing cybersecurity.

Why it matters: It comes as immigration agencies have flooded social media with content to support President Trump’s mass deportation agenda—some of which appears to be made with AI—and as workers in tech have put pressure on their employers to denounce the agencies’ activities. Read the full story.

—James O’Donnell

How the sometimes-weird world of lifespan extension is gaining influence

—Jessica Hamzelou

For the last couple of years, I’ve been following the progress of a group of individuals who believe death is humanity’s “core problem.” Put simply, they say death is wrong—for everyone. They’ve even said it’s morally wrong.

They established what they consider a new philosophy, and they called it Vitalism.

Vitalism is more than a philosophy, though—it’s a movement for hardcore longevity enthusiasts who want to make real progress in finding treatments that slow or reverse aging. Not just through scientific advances, but by persuading influential people to support their movement, and by changing laws and policies to open up access to experimental drugs. And they’re starting to make progress.

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.

The AI Hype Index: Grok makes porn, and Claude Code nails your job

Separating AI reality from hyped-up fiction isn’t always easy. That’s why we’ve created the AI Hype Index—a simple, at-a-glance summary of everything you need to know about the state of the industry. Take a look at this month’s edition of the index here.

The must-reads

I’ve combed the internet to find you today’s most fun/important/scary/fascinating stories about technology.

1 Capgemini is no longer tracking immigrants for ICE
After the French company was queried by the country’s government over the contract. (WP $)
+ Here’s how the agency typically keeps tabs on its targets. (NYT $)
+ US senators are pushing for answers about its recent surveillance shopping spree. (404 Media)
+ ICE’s tactics would get real soldiers killed, apparently. (Wired $)

2 The Pentagon is at loggerheads with Anthropic
The AI firm is reportedly worried its tools could be used to spy on Americans. (Reuters)
+ Generative AI is learning to spy for the US military. (MIT Technology Review)

3 It’s relatively rare for AI chatbots to lead users down harmful paths
But when it does, it can have incredibly dangerous consequences. (Ars Technica)
+ The AI doomers feel undeterred. (MIT Technology Review)

4 GPT-4o’s days are numbered
OpenAI says just 0.1% of users are using the model every day. (CNBC)
+ It’s the second time that it’s tried to turn the sycophantic model off in under a year. (Insider $)
+ Why GPT-4o’s sudden shutdown left people grieving. (MIT Technology Review)

5 An AI toy company left its chats with kids exposed
Anyone with a Gmail account was able to simply access the conversations—no hacking required. (Wired $)
+ AI toys are all the rage in China—and now they’re appearing on shelves in the US too. (MIT Technology Review)

6 SpaceX could merge with xAI later this year
Ahead of a planned blockbuster IPO of Elon Musk’s companies. (Reuters)
+ The move would be welcome news for Musk fans. (The Information $)
+ A SpaceX-Tesla merger could also be on the cards. (Bloomberg $)

7 We’re still waiting for a reliable male contraceptive
Take a look at the most promising methods so far. (Bloomberg $)

8 AI is bringing traditional Chinese medicine to the masses
And it’s got the full backing of the country’s government. (Rest of World)

9 The race back to the Moon is heating up 
Competition between the US and China is more intense than ever. (Economist $)

10 What did the past really smell like?
AI could help scientists to recreate history’s aromas—including mummies and battlefields. (Knowable Magazine)

Quote of the day

“I think the tidal wave is coming and we’re all standing on the beach.”

—Bill Zysblat, a music business manager, tells the Financial Times about the existential threat AI poses to the industry. 

One more thing

Therapists are secretly using ChatGPT. Clients are triggered.

Declan would never have found out his therapist was using ChatGPT had it not been for a technical mishap. The connection was patchy during one of their online sessions, so Declan suggested they turn off their video feeds. Instead, his therapist began inadvertently sharing his screen.

For the rest of the session, Declan was privy to a real-time stream of ChatGPT analysis rippling across his therapist’s screen, who was taking what Declan was saying, putting it into ChatGPT, and then parroting its answers.

But Declan is not alone. In fact, a growing number of people are reporting receiving AI-generated communiqués from their therapists. Clients’ trust and privacy are being abandoned in the process. Read the full story.

—Laurie Clarke

We can still have nice things

A place for comfort, fun and distraction to brighten up your day. (Got any ideas? Drop me a line or skeet ’em at me.)

+ Sinkholes are seriously mysterious. Is there a way to stay one step ahead of them?
+ This beautiful pixel art is super impressive.
+ Amid the upheaval in their city, residents of Minneapolis recently demonstrated both their resistance and community spirit in the annual Art Sled Rally (thanks Paul!)
+ How on Earth is Tomb Raider 30 years old?!

For the last couple of years, I’ve been following the progress of a group of individuals who believe death is humanity’s “core problem.” Put simply, they say death is wrong—for everyone. They’ve even said it’s morally wrong.

They established what they consider a new philosophy, and they called it Vitalism.

Vitalism is more than a philosophy, though—it’s a movement for hardcore longevity enthusiasts who want to make real progress in finding treatments that slow or reverse aging. Not just through scientific advances, but by persuading influential people to support their movement, and by changing laws and policies to open up access to experimental drugs.

And they’re starting to make progress.

Vitalism was founded by Adam Gries and Nathan Cheng—two men who united over their shared desire to find ways to extend human lifespan. I first saw Cheng speak back in 2023, at Zuzalu, a pop-up city in Montenegro for people who were interested in life extension and some other technologies. (It was an interesting experience—you can read more about it here.)

Zuzalu was where Gries and Cheng officially launched Vitalism. But I’ve been closely following the longevity scene since 2022. That journey took me to Switzerland, Honduras, and a compound in Berkeley, California, where like-minded longevity enthusiasts shared their dreams of life extension.

It also took me to Washington, DC, where, last year, supporters of lifespan extension presented politicians including Mehmet Oz, who currently leads the Centers for Medicare & Medicaid Services, with their case for changes to laws and policies.

The journey has been fascinating, and at times weird and even surreal. I’ve heard biohacking stories that ended with smoking legs. I’ve been told about a multi-partner relationship that might be made possible through the cryopreservation—and subsequent reanimation—of a man and the multiple wives he’s had throughout his life. I’ve had people tell me to my face that they consider themselves eugenicists, and that they believe that parents should select IVF embryos for their propensity for a long life.

I’ve seen people draw blood during dinner in an upscale hotel restaurant to test their biological age. I’ve heard wild plans to preserve human consciousness and resurrect it in machines. Others have told me their plans to inject men’s penises with multiple doses of an experimental gene therapy in order to treat erectile dysfunction and ultimately achieve “radical longevity.”

I’ve been shouted at and threatened with legal action. I’ve received barefoot hugs. One interviewee told me I needed Botox. It’s been a ride.

My reporting has also made me realize that the current interest in longevity reaches beyond social media influencers and wellness centers. Longevity clinics are growing in number, and there’s been a glut of documentaries about living longer or even forever.

At the same time, powerful people who influence state laws, giant federal funding budgets, and even national health policy are prioritizing the search for treatments that slow or reverse aging. The longevity community was thrilled when longtime supporter Jim O’Neill was made deputy secretary of health and human services last year. Other members of Trump’s administration, including Oz, have spoken about longevity too. “It seems that now there is the most pro-longevity administration in American history,” Gries told me.

I recently spoke to Alicia Jackson, the new director of ARPA-H. The agency, established in 2022 under Joe Biden’s presidency, funds “breakthrough” biomedical research. And it appears to have a new focus on longevity. Jackson previously founded and led Evernow, a company focused on “health and longevity for every woman.”

“There’s a lot of interesting technologies, but they all kind of come back to the same thing: Could we extend life years?” she told me over a Zoom call a few weeks ago. She added that her agency had “incredible support” from “the very top of HHS.” I asked if she was referring to Jim O’Neill. “Yeah,” she said. She wouldn’t go into the specifics.

Gries is right: There is a lot of support for advances in longevity treatments, and some of it is coming from influential people in positions of power. Perhaps the field really is poised for a breakthrough.

And that’s what makes this field so fascinating to cover. Despite the occasional weirdness.

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.

Everyone is panicking because AI is very bad; everyone is panicking because AI is very good. It’s just that you never know which one you’re going to get. Grok is a pornography machine. Claude Code can do anything from building websites to reading your MRI. So of course Gen Z is spooked by what this means for jobs. Unnerving new research says AI is going to have a seismic impact on the labor market this year.

If you want to get a handle on all that, don’t expect any help from the AI companies—they’re turning on each other like it’s the last act in a zombie movie. Meta’s former chief AI scientist, Yann LeCun, is spilling tea, while Big Tech’s messiest exes, Elon Musk and OpenAI, are about to go to trial. Grab your popcorn.

The US Department of Homeland Security is using AI video generators from Google and Adobe to make and edit content shared with the public, a new document reveals. It comes as immigration agencies have flooded social media with content to support President Trump’s mass deportation agenda—some of which appears to be made with AI—and as workers in tech have put pressure on their employers to denounce the agencies’ activities. 

The document, released on Wednesday, provides an inventory of which commercial AI tools DHS uses for tasks ranging from generating drafts of documents to managing cybersecurity. 

In a section about “editing images, videos or other public affairs materials using AI,” it reveals for the first time that DHS is using Google’s Veo 3 video generator and Adobe Firefly, estimating that the agency has between 100 and 1,000 licenses for the tools. It also discloses that DHS uses Microsoft Copilot Chat for generating first drafts of documents and summarizing long reports and Poolside software for coding tasks, in addition to tools from other companies.

Google, Adobe, and DHS did not immediately respond to requests for comment.

The news provides details about how agencies like Immigrations and Customs Enforcement, which is part of DHS, might be creating the large amounts of content they’ve shared on X and other channels as immigration operations have expanded across US cities. They’ve posted content celebrating “Christmas after mass deportations,” referenced Bible verses and Christ’s birth, showed faces of those the agency has arrested, and shared ads aimed at recruiting agents. The agencies have also repeatedly used music without permissions from artists in their videos.

Some of the content, particularly videos, has the appearance of being AI-generated, but it hasn’t been clear until now what AI models the agencies might be using. This marks the first concrete evidence such generators are being used by DHS to create content shared with the public.

It still remains impossible to verify which company helped create a specific piece of content, or indeed if it was AI-generated at all. Adobe offers options to “watermark” a video made with its tools to disclose that it is AI-generated, for example, but this disclosure does not always stay intact when the content is uploaded and shared across different sites. 

The document reveals that DHS has specifically been using Flow, a tool from Google that combines its Veo 3 video generator with a suite of filmmaking tools. Users can generate clips and assemble entire videos with AI, including videos that contain sound, dialogue, and background noise, making them hyperrealistic. Adobe launched its Firefly generator in 2023, promising that it does not use copyrighted content in its training or output. Like Google’s tools, Adobe’s can generate videos, images, soundtracks, and speech. The document does not reveal further details about how the agency is using these video generation tools.

Workers at large tech companies, including more than 140 current and former employees from Google and more than 30 from Adobe, have been putting pressure on their employers in recent weeks to take a stance against ICE and the shooting of Alex Pretti on January 24. Google’s leadership has not made statements in response. In October, Google and Apple removed apps on their app stores that were intended to track sightings of ICE, citing safety risks. 

An additional document released on Wednesday revealed new details about how the agency is using more niche AI products, including a facial recognition app used by ICE, as first reported by 404Media in June.

This is today’s edition of The Download, our weekday newsletter that provides a daily dose of what’s going on in the world of technology.

Meet the Vitalists: the hardcore longevity enthusiasts who believe death is “wrong”

Last April, an excited crowd gathered at a compound in Berkeley, California, for a three-day event called the Vitalist Bay Summit. It was part of a longer, two-month residency that hosted various events to explore tools—from drug regulation to cryonics—that might be deployed in the fight against death.

One of the main goals, though, was to spread the word of Vitalism, a somewhat radical movement established by Nathan Cheng and his colleague Adam Gries a few years ago. Consider it longevity for the most hardcore adherents—a sweeping mission to which nothing short of total devotion will do.

Although interest in longevity has certainly taken off in recent years, not everyone in the broader longevity space shares Vitalists’ commitment to actually making death obsolete. And the Vitalists feel that momentum is building, not just for the science of aging and the development of lifespan-extending therapies, but for the acceptance of their philosophy that defeating death should be humanity’s top concern. Read the full story.

—Jessica Hamzelou

This is the latest in our Big Story series, the home for MIT Technology Review’s most important, ambitious reporting. You can read the rest of the series here. 

What AI “remembers” about you is privacy’s next frontier

—Miranda Bogen, director of the AI Governance Lab at the Center for Democracy & Technology, & Ruchika Joshi, fellow at the Center for Democracy & Technology specializing in AI safety and governance

The ability to remember you and your preferences is rapidly becoming a big selling point for AI chatbots and agents.

Personalized, interactive AI systems are built to act on our behalf, maintain context across conversations, and improve our ability to carry out all sorts of tasks, from booking travel to filing taxes.

But their ability to store and retrieve increasingly intimate details about their users over time introduces alarming, and all-too-familiar, privacy vulnerabilities––many of which have loomed since “big data” first teased the power of spotting and acting on user patterns. Worse, AI agents now appear poised to plow through whatever safeguards had been adopted to avoid those vulnerabilities. So what can developers do to fix this problem? Read the full story.

How the grid can ride out winter storms

The eastern half of the US saw a monster snowstorm over the weekend. The good news is the grid has largely been able to keep up with the freezing temperatures and increased demand. But there were some signs of strain, particularly for fossil-fuel plants.

One analysis found that PJM, the nation’s largest grid operator, saw significant unplanned outages in plants that run on natural gas and coal. Historically, these facilities can struggle in extreme winter weather.

Much of the country continues to face record-low temperatures, and the possibility is looming for even more snow this weekend. What lessons can we take from this storm, and how might we shore up the grid to cope with extreme weather? Read the full story.

—Casey Crownhart

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

The must-reads

I’ve combed the internet to find you today’s most fun/important/scary/fascinating stories about technology.

1 Telegram has been flooded with deepfake nudes 
Millions of users are creating and sharing falsified images in dedicated channels. (The Guardian)

2 China has executed 11 people linked to Myanmar scam centers
The members of the “Ming family criminal gang” caused the death of at least 14 Chinese citizens. (Bloomberg $)
+ Inside a romance scam compound—and how people get tricked into being there. (MIT Technology Review)

3 This viral personal AI assistant is a major privacy concern
Security researchers are sounding the alarm on Moltbot, formerly known as Clawdbot. (The Register)
+ It requires a great deal more technical know-how than most agentic bots. (TechCrunch)

4 OpenAI has a plan to keep bots off its future social network
It’s putting its faith in biometric “proof of personhood” promised by the likes of World’s eyeball-scanning orb. (Forbes)
+ We reported on how World recruited its first half a million test users back in 2022. (MIT Technology Review)

5 Here’s just some of the technologies ICE is deploying
From facial recognition to digital forensics. (WP $)
+ Agents are also using Palantir’s AI to sift through tip-offs. (Wired $)

6 Tesla is axing its Model S and Model X cars
Its Fremont factory will switch to making Optimus robots instead. (TechCrunch)
+ It’s the latest stage of the company’s pivot to AI… (FT $)
+ …as profit falls by 46%. (Ars Technica)
+ Tesla is still struggling to recover from the damage of Elon Musk’s political involvement. (WP $)

7 X is rife with weather influencers spreading misinformation
They’re whipping up hype ahead of massive storms hitting. (New Yorker $)

8  Retailers are going all-in on AI
But giants like Amazon and Walmart are taking very different approaches. (FT $)
+ Mark Zuckerberg has hinted that Meta is working on agentic commerce tools. (TechCrunch)
+ We called it—what’s next for AI in 2026. (MIT Technology Review)

9 Inside the rise of the offline hangout
No phones, no problem. (Wired $)

10 Social media is obsessed with 2016
…why, exactly? (WSJ $)

Quote of the day

“The amount of crap I get for putting out a hobby project for free is quite something.”

—Peter Steinberger, the creator of the viral AI agent Moltbot, complains about the backlash his project has received from security researchers pointing out its flaws in a post on X.

One more thing

The flawed logic of rushing out extreme climate solutions

Early in 2022, entrepreneur Luke Iseman says, he released a pair of sulfur dioxide–filled weather balloons from Mexico’s Baja California peninsula, in the hope that they’d burst miles above Earth.

It was a trivial act in itself, effectively a tiny, DIY act of solar geoengineering, the controversial proposal that the world could counteract climate change by releasing particles that reflect more sunlight back into space.

Entrepreneurs like Iseman invoke the stark dangers of climate change to explain why they do what they do—even if they don’t know how effective their interventions are. But experts say that urgency doesn’t create a social license to ignore the underlying dangers or leapfrog the scientific process. Read the full story.

—James Temple

We can still have nice things

A place for comfort, fun and distraction to brighten up your day. (Got any ideas? Drop me a line or skeet ’em at me.)

+ The hottest thing in art right now? Vertical paintings.
+ There’s something in the water around Monterey Bay—a tail walking dolphin!
+ Fed up of hairstylists not listening to you? Remember these handy tips the next time you go for a cut.
+ Get me a one-way ticket to Japan’s tastiest island.

The eastern half of the US saw a monster snowstorm over the weekend. The good news is the grid has largely been able to keep up with the freezing temperatures and increased demand. But there were some signs of strain, particularly for fossil-fuel plants.

One analysis found that PJM, the nation’s largest grid operator, saw significant unplanned outages in plants that run on natural gas and coal. Historically, these facilities can struggle in extreme winter weather.

Much of the country continues to face record-low temperatures, and the possibility is looming for even more snow this weekend. What lessons can we take from this storm, and how might we shore up the grid to cope with extreme weather?

Living in New Jersey, I have the honor of being one of the roughly 67 million Americans covered by the PJM Interconnection.

So I was in the thick of things this weekend, when PJM saw unplanned outages of over 20 gigawatts on Sunday during the height of the storm. (That’s about 16% of the grid’s demand that afternoon.) Other plants were able to make up the difference, and thankfully, the power didn’t go out in my area. But that’s a lot of capacity offline.

Typically, the grid operator doesn’t announce details about why an outage occurs until later. But analysts at Energy Innovation, a policy and research firm specializing in energy and climate, went digging. By examining publicly available grid mix data (a breakdown of what types of power plants are supplying the grid), the team came to a big conclusion: Fossil fuels failed during the storm.

The analysts found that gas-fired power plants were producing about 10 gigawatts less power on Sunday than the peak demand on Saturday, even while electricity prices were high. Coal- and oil-burning plants were down too. Because these plants weren’t operating, even when high prices would make it quite lucrative, they were likely a significant part of the problem, says Michelle Solomon, a manager in the electricity program at Energy Innovation.

PJM plans to share more details about the outages at an upcoming committee meeting once the cold snap passes, Dan Lockwood, a PJM spokesperson, told me via email.

Fossil-fuel plants can see reliability challenges during winter: When temperatures drop, pressures in natural-gas lines fall too, which can lead to issues for fuel supply. Freezing temperatures can throw compression stations and other mechanical equipment offline and even freeze piles of coal.

One of the starkest examples came in 2021, when Texas faced freezing temperatures that took many power plants offline and threw the grid into chaos. Many homes lost power for days, and at least 246 people died during that storm.

Texas fared much better this time around. After 2021, the state shored up its grid, adding winter weatherization for power plants and transmission systems. Texas has also seen a huge flood of batteries come online, which has greatly helped the grid during winter demand peaks, especially in the early mornings. Texas was also simply lucky that this storm was less severe there, as one expert told Inside Climate News this week.

Here on the East Coast, we’re not out of the woods yet. The snow has stopped falling, but grids are still facing high electricity demand because of freezing temperatures. (I’ve certainly been living under my heated blanket these last few days.)

PJM could see a peak power demand of 130 gigawatts for seven straight days, a winter streak that the local grid has never experienced, according to an update to the utility’s site on Tuesday morning.

The US Department of Energy issued emergency orders to several grid operators, including PJM, that allow power plants to run while basically ignoring emissions regulations. The department also issued orders allowing several grids to tell data centers and other facilities to begin using backup generators. (This is good news for reliability but bad news for clean air and the climate, since these power sources are often incredibly emissions-intensive.)

We here on the East Coast could learn a thing or two from Texas so we don’t need to resort to these polluting emergency measures to keep the lights on. More energy storage could be a major help in future winter storms, lending flexibility to the grid to help ride out the worst times, Solomon says. Getting offshore wind online could also help, since those facilities typically produce reliable power in the winter. 

No one energy source will solve the massive challenge of building and maintaining a resilient grid. But as we face the continued threat of extreme storms, renewables might actually help us weather them. 

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

“Who here believes involuntary death is a good thing?” 

Nathan Cheng has been delivering similar versions of this speech over the last couple of years, so I knew what was coming. He was about to try to convince the 80 or so people in the audience that death is bad. And that defeating it should be humanity’s number one priority—quite literally, that it should come above all else in the social and political hierarchy.

“If you believe that life is good and there’s inherent moral value to life,” he told them, “it stands to reason that the ultimate logical conclusion here is that we should try to extend lifespan indefinitely.” 

Solving aging, he added, is “a problem that has an incredible moral duty for all of us to get involved in.”

It was the end of April, and the crowd—with its whoops and yeahs—certainly seemed convinced. They’d gathered at a compound in Berkeley, California, for a three-day event called the Vitalist Bay Summit. It was part of a longer, two-month residency (simply called Vitalist Bay) that hosted various events to explore tools—from drug regulation to cryonics—that might be deployed in the fight against death. One of the main goals, though, was to spread the word of Vitalism, a somewhat radical movement established by Cheng and his colleague Adam Gries a few years ago.

No relation to the lowercase vitalism of old, this Vitalism has a foundational philosophy that’s deceptively simple: to acknowledge that death is bad and life is good. The strategy for executing it, though, is far more obviously complicated: to launch a longevity revolution. 

Interest in longevity has certainly taken off in recent years, but as the Vitalists see it, it has a branding problem. The term “longevity” has been used to sell supplements with no evidence behind them, “anti-aging” has been used by clinics to sell treatments, and “transhumanism” relates to ideas that go well beyond the scope of defeating death. Not everyone in the broader longevity space shares Vitalists’ commitment to actually making death obsolete. As Gries, a longtime longevity devotee who has largely become the enthusiastic public face of Vitalism, said in an online presentation about the movement in 2024, “We needed some new word.”

“Vitalism” became a clean slate: They would start a movement to defeat death, and make that goal the driving force behind the actions of individuals, societies, and nations. Longevity could no longer be a sideshow. For Vitalism to succeed, budgets would need to change. Policy would need to change. Culture would need to change. Consider it longevity for the most hardcore adherents—a sweeping mission to which nothing short of total devotion will do.

“The idea is to change the systems and the priorities of society at the highest levels,” Gries said in the presentation.

To be clear, the effective anti-aging treatments the Vitalists are after don’t yet exist. But that’s sort of the point: They believe they could exist if Vitalists are able to spread their gospel, influence science, gain followers, get cash, and ultimately reshape government policies and priorities. 

For the past few years, Gries and Cheng have been working to recruit lobbyists, academics, biotech CEOs, high-net-worth individuals, and even politicians into the movement, and they’ve formally established a nonprofit foundation “to accelerate Vitalism.” Today, there’s a growing number of Vitalists (some paying foundation members, others more informal followers, and still others who support the cause but won’t publicly admit as much), and the foundation has started “certifying” qualifying biotech companies as Vitalist organizations. Perhaps most consequentially, Gries, Cheng, and their peers are also getting involved in shaping US state laws that make unproven, experimental treatments more accessible. They hope to be able to do the same at the national level.

Vitalism cofounders Nathan Cheng and Adam Gries want to launch a longevity revolution.

All this is helping Vitalists grow in prominence, if not also power. In the past, people who have spoken of living forever or making death “optional” have been dismissed by their academic colleagues. I’ve been covering the broader field of aging science for a decade, and I’ve seen scientists roll their eyes, shrug their shoulders, and turn their backs on people who have talked this way. That’s not the case for the Vitalists.  

Even the scientists who think that Vitalist ideas of defeating death are wacky, unattainable ones, with the potential to discredit their field, have shown up on stage with Vitalism’s founders, and these serious researchers provide a platform for them at more traditionally academic events.

I saw this collegiality firsthand at Vitalist Bay. Faculty members from Harvard, Stanford, and the University of California, Berkeley, all spoke at events. Eric Verdin, the prominent researcher who directs the Buck Institute for Research on Aging in Novato, California, had also planned to speak, although a scheduling clash meant he couldn’t make it in the end. “I have very different ideas in terms of what’s doable,” he told me. “But that’s part of the [longevity] movement—there’s freedom for people to say whatever they want.” 

Many other well-respected scientists attended, including representatives of ARPA-H, the US federal agency for health research and breakthrough technologies. And as I left for a different event on longevity in Washington, DC, just after the Vitalist Bay Summit, a sizable group of Vitalist Bay attendees headed that way too, to make the case for longevity to US lawmakers.

The Vitalists feel that momentum is building, not just for the science of aging and the development of lifespan-extending therapies, but for the acceptance of their philosophy that defeating death should be humanity’s top concern. 

This, of course, sparks some pretty profound questions. What would a society without death look like—and would we even want it? After all, death has become an important part of human culture the world over. And even if Vitalists aren’t destined to realize their lofty goal, their growing influence could still have implications for us all. As they run more labs and companies, and insert themselves into the making of laws and policy, perhaps they will discover treatments that really do slow or even reverse aging. In the meantime, though, some ethicists are concerned that experimental and unproven medicines—including potentially dangerous ones—are becoming more accessible, in some cases with little to no oversight. 

Gries, ultimately, has a different view of the ethics here. He thinks that being “okay with death” is what disqualifies a person from being considered ethical. “Death is just wrong,” he says. “It’s not just wrong for some people. It’s wrong for all people.”

The birth of a revolution

When I arrived at the Vitalist Bay Summit on April 25, I noticed that the venue was equipped with everything a longevity enthusiast might need: napping rooms, a DEXA body-composition scanner, a sauna in a bus, and, for those so inclined, 24-hour karaoke. 

I was told that around 300 people had signed up for that day’s events, which was more than had attended the previous week. That might have been because arguably the world’s most famous longevity enthusiast, Bryan Johnson, was about to make an appearance. (If you’re curious to know more about what Johnson was doing there, you can read about our conversation here.) 

The key to Vitalism has always been that “death is humanity’s core problem, and aging its primary agent,” cofounder Adam Gries told me. “So it was, and so it has continued, as it was foretold.” 

But Gries, another man in his 40s who doesn’t want to die, was the first to address the audience that day. Athletic and energetic, he bounded across a stage wearing bright yellow shorts and a long-sleeved shirt imploring people to “Choose Life: VITALISM.”

Gries is a tech entrepreneur who describes himself as a self-taught software engineer who’s “good at virality.” He’s been building companies since he was in college in the 2000s, and grew his personal wealth by selling them.

As with many other devotees to the cause, his deep interest in life extension was sparked by Aubrey de Grey, a controversial researcher with an iconic long beard and matching ponytail. He’s known widely both for his optimistic views about “defeating aging” and for having reportedly made sexual comments to two longevity entrepreneurs. (In an email, de Grey said he’s “never disputed” one of these remarks but denied having made the other. “My continued standing within the longevity community speaks for itself,” he added.) 

In an influential 2005 TED Talk (which has over 4.8 million views), de Grey predicted that people would live to 1,000 and spoke of the possibility of new technologies that would continue to stave off death, allowing some to avoid it indefinitely. (In a podcast recorded last year, Cheng described a recording of this talk as “the OG longevity-pilling YouTube video.”)

“It was kind of evident to me that life is great,” says Gries. “So I’m kind of like, why would I not want to live?”

A second turning point for Gries came during the early stages of the covid-19 pandemic, when he essentially bet against companies that he thought would collapse. “I made this 50 [fold] return,” he says. “It was kind of like living through The Big Short.”

Gries and his wife fled from San Francisco to Israel, where he grew up, and later traveled to Taiwan, where he’d obtained a “golden visa” and which was, at the time, one of only two countries that had not reported a single case of covid. His growing wealth afforded him the opportunity to take time from work and think about the purpose of life. “My answer was: Life is the purpose of life,” he says. He didn’t want to die. He didn’t want to experience the “journey of decrepitude” that aging often involves.

So he decided to dedicate himself to the longevity cause. He went about looking up others who seemed as invested as he was. In 2021 his search led him to Cheng, a Chinese-Canadian entrepreneur based in Toronto. He had dropped out of a physics PhD a few years earlier after experiencing what he describes on his website as “a massive existential crisis” and shifted his focus to “radical longevity.” (Cheng did not respond to email requests for an interview.)

The pair “hit it off immediately,” says Gries, and they spent the following two years trying to figure out what they could do. The solution they finally settled on: revolution.

After all, Gries reasons, that’s how significant religious and social movements have happened in the past. He says they sought inspiration from the French and American Revolutions, among others. The idea was to start with some kind of “enlightenment,” and with a “hardcore group,” to pursue significant social change with global ramifications. 

“We were convinced that without a revolution,” Gries says, “we were as good as dead.” 

A home for believers

Early on, they wrote a Vitalist declaration, a white paper that lists five core statements for believers:

1. Life and health are good. Death is humanity’s core problem, and aging its primary agent.
2. Aging causes immense suffering, and obviating aging is scientifically plausible.
3. Humanity should apply the necessary resources to reach freedom from aging as soon as possible.
4. I will work on or support others to work on reaching unlimited healthy human lifespan.
5. I will carry the message against aging and death.

While it’s not an explicit part of the manifesto, it was important to them to think about it as a moral philosophy as well as a movement. As Cheng said at the time, morality “guides most of the actions of our lives.” The same should be true of Vitalism, he suggested. 

Gries has echoed this idea. The belief that “death is morally bad” is necessary to encourage behavior change, he told me in 2024. It is a moral drive, or moral purpose, that pushes people to do difficult things, he added.

Revolution, after all, is difficult. And to succeed—to “get unlimited great health to the top of the priority list,” as Gries says—the movement would need to infiltrate the government and shape policy decisions and national budgets. The Apollo program got people to the moon with less than 1% of US GDP; imagine, Gries asks, what we could do to human longevity with a mere 1% of GDP?

It makes sense, then, that Gries and Cheng launched Vitalism in 2023 at Zuzalu, a “pop-up city” in Montenegro that provided a two-month home for like-minded longevity enthusiasts. The gathering was in some ways a loose prototype for what they wanted to accomplish. Cheng spoke there of how they wanted to persuade 10,000 or so Vitalists to move to Rhode Island. Not only was it close to the biotech hub of Boston, but they believed it had a small enough population for an influx of new voters sharing their philosophy to influence local and state elections. “Five to ten thousand people—that’s all we need,” he said. Or if not Rhode Island, another small-ish US state, where they could still change state policy from the inside. 

The ultimate goal was to recruit Vitalists to help them establish a “longevity state”—a recognized jurisdiction that “prioritizes doing something about aging,” Cheng said, perhaps by loosening regulations on clinical trials or supporting biohacking.

This idea is popular among many vocal members of the Vitalism community. It borrows from the concept of the “network state” developed by former Coinbase CTO Balaji Srinivasan, defined as a new city or country that runs on cryptocurrency; focuses on a goal, in this case extending human lifespan; and “eventually gains diplomatic recognition from preexisting states.” 

Some people not interested in dying have made progress toward realizing such a domain. Following the success of Zuzalu, one of the event’s organizers, Laurence Ion, a young cryptocurrency investor and self-proclaimed Vitalist, joined a fellow longevity enthusiast named Niklas Anzinger to organize a sequel in Próspera, the private “special economic zone” on the Honduran island of Roatán. They called their “pop-up city” Vitalia.

I visited shortly after it launched in January 2024. The goal was to create a low-regulation biotech hub to fast-track the development of anti-aging drugs, though the “city” was more like a gated resort that hosted talks from a mix of respected academics, biohackers, biotech CEOs, and straight-up eugenicists. There was a strong sense of community—many attendees were living with or near each other, after all. A huge canvas where attendees could leave notes included missives like “Don’t die,” “I love you,” and “Meet technoradicals building the future!” 

But Vitalia was short-lived, with events ending by the start of March 2024. And while many of the vibes were similar to what I’d later see at Vitalist Bay, the temporary nature of Vitalia didn’t quite match the ambition of Gries and Cheng. 

Patri Friedman, a 49-year-old libertarian and grandson of the economist Milton Friedman who says he attended Zuzalu, Vitalia, and Vitalist Bay, envisions something potentially even bolder. He’s the founder of the Seasteading Institute, which has the goal of “building startup communities that float on the ocean with any measure of political autonomy” and has received funding and support from the billionaire Peter Thiel. Friedman also founded Pronomos Capital, a venture capital fund that invests in projects focused on “building the cities of tomorrow.” 

His company is exploring various types of potential network states, but he says he’s found that medical tourism—and, specifically, a hunger for life extension—dominates the field. “People do not want this ‘10 years and a billion dollars to pass a drug’ thing with the FDA,” says Friedman. (While he doesn’t call himself a Vitalist, partly because he’s “almost never going to agree with” any kind of decree, Friedman holds what you might consider similarly staunch sentiments about death, which he referred to as “murder by omission.” When I asked him if he has a target age he’d like to reach, he told me he found the question “mind-bogglingly strange” and “insane.” “How could you possibly be like: Yes, please murder me at this time?” he replied. “I can always fucking shoot myself in the head—I don’t need anybody’s help.”) 

But even as Vitalists and those aligned with their beliefs embrace longevity states, Gries and Cheng are reassessing their former ambitions. The network-state approach has limits, Gries tells me. And encouraging thousands of people to move to Rhode Island wasn’t as straightforward as they’d hoped it might be.

Not because he can’t find tens of thousands of Vitalists, Gries stresses—but most of them are unwilling to move their lives for the sake of influencing the policy of another state. He compares Vitalism to a startup, with a longevity state as its product. For the time being, at least, there isn’t enough consumer appetite for that product, he says. 

The past year shows that it may in fact be easier to lobby legislators in states that are already friendly to deregulation. Anzinger and a lobbying group called the Alliance for Longevity Initiatives (A4LI) were integral to making Montana the first US hub for experimental medical treatments, with a new law to allow clinics to sell experimental therapies once they have been through preliminary safety tests (which don’t reveal whether a drug actually works). But Gries and his Vitalist colleagues also played a role—“providing feedback, talking to lawmakers … brainstorming [and] suggesting ideas,” Gries says. 

The Vitalist crew has been in conversation with lawmakers in New Hampshire, too. In an email in December, Gries and Cheng claimed they’d “helped to get right-to-try laws passed” in the state—an apparent reference to the recent expansion of a law to make more unapproved treatments accessible to people with terminal illnesses. Meanwhile, three other bills that expand access even further are under consideration. 

Ultimately, Gries stresses, Vitalism is “agnostic to the fixing strategies” that will help them meet their goals. There is, though, at least one strategy he’s steadfast about: building influence.

Only the hardcore 

To trigger a revolution, the Vitalists may need to recruit only around 3% or 4% of “society” to their movement, Gries believes. (Granted, that does still mean hundreds of millions of people.) “If you want people to take action, you need to focus on a small number of very high-leverage people,” he tells me. 

That, perhaps unsurprisingly, includes wealthy individuals with “a net worth of $10 million or above,” he says. He wants to understand why (with some high-profile exceptions, including Thiel, who has been investing in longevity-related companies and foundations for decades) most uber-wealthy people don’t invest in the field—and how he might persuade them to do so. He won’t reveal the names of anyone he’s having conversations with. 

These “high-leverage” people might also include, Gries says, well-respected academics, leaders of influential think tanks, politicians and policymakers, and others who work in government agencies.

A revolution needs to find its foot soldiers. And at the most basic level, that will mean boosting the visibility of the Vitalism brand—partly through events like Vitalist Bay, but also by encouraging others, particularly in the biotech space, to sign on. Cheng talks of putting out a “bat signal” for like-minded people, and he and Gries say that Vitalism has brought together people who have gone on to collaborate or form companies. 

There’s also their nonprofit Vitalism International Foundation, whose supporters can opt to become “mobilized Vitalists” with monthly payments of $29 or more, depending on their level of commitment. In addition, the foundation works with longevity biotech companies to recognize those that are “aligned” with its goals as officially certified Vitalist organizations. “Designation may be revoked if an organization adopts apologetic narratives that accept aging or death,” according to the website. At the time of writing, that site lists 16 certified Vitalist organizations, including cryopreservation companies, a longevity clinic, and several research companies. 

One of them is Shift Bioscience, a company using CRISPR and aging clocks—which attempt to measure biological age—to identify genes that might play a significant role in the aging process and potentially reverse it. It says it has found a single gene that can rejuvenate multiple types of cells. 

Shift cofounder Daniel Ives, who holds degrees in mitochondrial and computational biology, tells me he was also won over to the longevity cause by de Grey’s 2005 TED Talk. He now has a countdown on his computer: “It’s my days till death,” he says—around 22,000 days left. “I’m using that to keep myself focused.” 

Ives calls himself the “Vitalist CEO” of Shift Bioscience. He thinks the label is important first as a way for like-minded people to find and support each other, grow their movement, and make the quest for longevity mainstream. Second, he says, it provides a way to appeal to “hardcore” lifespan extensionists, given that others in the wellness and cosmetics industry have adopted the term “longevity” without truly applying themselves to finding rejuvenation therapies. He refers to unnamed companies and individuals who claim that drinking juices, for example, can reverse aging by five years or so.

“You don’t have to convince the mainstream,” says Mark Hamalainen, a contributor to the Vitalism white paper. Though “kind of a terrible example,” he notes, Stalinism started small. “Sometimes you just have to convince the right people.”

“Somebody will make these claims and basically throw legitimate science under the bus,” he says. He doesn’t want spurious claims made on social media to get lumped in with the company’s serious molecular biology. Shift’s head of machine learning, Lucas Paulo de Lima Camillo, was recently awarded a $10,000 prize by the well-respected Biomarkers of Aging Consortium for an aging clock he developed. 

Another out-and-proud Vitalist CEO is Anar Isman, the cofounder of AgelessRx, a telehealth provider that offers prescriptions for purported longevity drugs—and a certified Vitalist organization. (Isman, who is in his early 40s, used to work at a hedge fund but was inspired to join the longevity field by—you guessed it—de Grey.)

During a panel session at Vitalist Bay, he stressed that he too saw longevity as a movement—and a revolution—rather than an industry. But he also claimed his company wasn’t doing too badly commercially. “We’ve had a lot of demand,” he said. “We’ve got $60 million plus in annual revenue.”

Many of his customers come to the site looking for treatments for specific ailments, he tells me. He views each as an opportunity to “evangelize” his views on “radical life extension.” “I don’t see a difference between … dying tomorrow or dying in 30 years,” he says. He wants to live “at least 100 more” years.

Vitalism, though, isn’t just appealing to commercial researchers. Mark Hamalainen, a 41-year-old science and engineering advisor at ARPA-H, describes himself as a Vitalist. He says he “kind of got roped into” Vitalism because he also works with Cheng—they founded the Longevity Biotech Fellowship, which supports new entrants to the field through mentoring programs. “I kind of view it as a more appealing rebranding of some of the less radical aspects of transhumanism,” he says. Transhumanism—the position that we can use technologies to enhance humans beyond the current limits of biology—covers a broad terrain, but “Vitalism is like: Can we just solve this death thing first? It’s a philosophy that’s easy to get behind.”

In government, he works with individuals like Jean Hébert, a former professor of genetics and neuroscience who has investigated the possibility of rejuvenating the brain by gradually replacing parts of it; Hébert has said that “[his] mission is to beat aging.” He spoke at Zuzalu and Vitalist Bay. 

Andrew Brack, who serves as the program manager for proactive health at ARPA-H, was at Vitalist Bay, too. Both Brack and Hébert oversee healthy federal budgets—Hébert’s brain replacement project was granted $110 million in 2024, for example.

Neither Hébert nor Brack has publicly described himself as a Vitalist, and Hébert wouldn’t agree to speak to me without the approval of ARPA-H’s press office, which didn’t respond to multiple requests for an interview with him or Brack. Brack did not respond to direct requests for an interview.

Gries says he thinks that “many people at [the US Department of Health and Human Services], including all agencies, have a longevity-positive view and probably agree with a lot of the ideas Vitalism stands for.” And he is hoping to help secure federal positions for others who are similarly aligned with his philosophy. On both Christmas Eve and New Year’s Eve last year, Gries and Cheng sent fundraising emails describing an “outreach effort” to find applicants for six open government positions that, together, would control billions of dollars in federal funding. “Qualified, mission-aligned candidates we’d love to support do exist, but they need to be found and encouraged to apply,” the pair wrote in the second email. “We’re starting a systematic search to reach, screen, and support the best candidates.” 

Hamalainen supports Gries’s plan to target high-leverage individuals. “You don’t have to convince the mainstream,” he says. Though “kind of a terrible example,” Hamalainen notes, Stalinism started small. “Sometimes you just have to convince the right people.”

One of the “right” people may be the man who inspired Gries, Hamalainen, Ives, Isman, and so many others to pursue longevity in the first place: de Grey. He’s now a paid-up Vitalist and even spoke at Vitalist Bay. Having been in the field for over 20 years, de Grey tells me, he’s seen various terms fall in and out of favor. Those terms now have “baggage that gets in the way,” he says. “Sometimes it’s useful to have a new term.”

The sometimes quiet (sometimes powerful, sometimes influential) Vitalists

Though one of the five principles of Vitalism is a promise to “carry the message,” some people who agree with its ideas are reluctant to go public, including some signed-up Vitalists. I’ve asked Gries multiple times over several years, but he won’t reveal how many Vitalists there are, let alone who makes up the membership.

Even some of the founders of Vitalism don’t want to be public about it. Around 30 people were involved in developing the movement, Gries says—but only 22 are named as contributors to the Vitalism white paper (with Gries as its author), including Cheng, Vitalia’s Ion, and ARPA-H’s Hamalainen. Gries won’t reveal the names of the others. He acknowledges that some people just don’t like to publicly affiliate with any organization. That’s certainly what I’ve found when I’ve asked members of the longevity community if they’re Vitalists. Many said they agreed with the Vitalist declaration, and that they liked and supported what Gries was doing. But they didn’t want the label.

Some people worry that associating with a belief system that sounds a bit religious—even cult-like, some say—won’t do the cause any favors. Others have a problem with the specific wording of the declaration.

For instance, Anzinger—the other Vitalia founder—won’t call himself a Vitalist. He says he respects the mission, but that the declaration is “a bit poetic” for his liking.

And Dylan Livingston, CEO of A4LI and arguably one of the most influential longevity enthusiasts out there, won’t describe himself as a Vitalist either.

Many other longevity biotech CEOs also shy away from the label—including Emil Kendziorra, who runs the human cryopreservation company Tomorrow Bio, even though that’s a certified Vitalist organization. Kendziorra says he agrees with most of the Vitalist declaration but thinks it is too “absolutist.” He also doesn’t want to imply that the pursuit of longevity should be positioned above war, hunger, and other humanitarian issues. (Gries has heard this argument before, and counters that both the vast spending on health care for people in the last years of their life and the use of lockdown strategies during the covid pandemic suggest that, deep down, lifespan extension is “society’s revealed preference.”)

Still, because Kendziorra agrees with almost everything in the declaration, he believes that “pushing it forward” and bringing more attention to the field by labeling his company a Vitalist organization is a good thing. “It’s to support other people who want to move the world in that direction,” he says. (He also offered Vitalist Bay attendees a discount on his cryopreservation services.) 

“There’s a lot of closeted scientists working in our field, and they get really excited about lifespans increasing,” explains Ives of Shift Bioscience. “But you’ll get people who’ll accuse you of being a lunatic that wants to be immortal.” He claims that people who represent biotech companies tell him “all the time” that they are secretly longevity companies but avoid using the term because they don’t want funders or collaborators to be “put off.”

Ultimately, it may not really matter how much people adopt the Vitalist label as long as the ideas break through. “It’s pretty simple. [The Vitalist declaration] has five points—if you agree with the five points, you are a Vitalist,” says Hamalainen. “You don’t have to be public about it.” He says he’s spoken to others about “coming out of the closet” and that it’s been going pretty well. 

Gries puts it more bluntly: “If you agree with the Vitalist declaration, you are a Vitalist.” 

And he hints that there are now many people in powerful positions—including in the Trump administration—who share his views, even if they don’t openly identify as Vitalists. 

For Gries, this includes Jim O’Neill, the deputy secretary of health and human services, whom I profiled a few months after he became Robert F. Kennedy Jr.’s number two. (More recently, O’Neill was temporarily put in charge of the US Centers for Disease Control and Prevention.)

O’Neill has long been interested in both longevity and the idea of creating new jurisdictions. Until March 2024, he served on the board of directors of Friedman’s Seasteading Institute. He also served as CEO of the SENS Research Foundation, a longevity organization founded by de Grey, between 2019 and 2021, and he represented Thiel as a board member there for many years. Many people in the longevity community say they know him personally, or have at least met him. (Tristan Roberts, a biohacker who used to work with a biotech company operating in Próspera, tells me he served O’Neill gin when he visited his Burning Man camp, which he describes as a “technology gay camp from San Francisco and New York.” Hamalainen also recalls meeting O’Neill at Burning Man, at a “techy, futurist” camp.) (Neither O’Neill nor representatives from the Department of Health and Human Services responded to a request to comment about this.)

O’Neill’s views are arguably becoming less fringe in DC these days. The day after the Vitalist Bay Summit, A4LI was hosting its own summit in the capital with the goal of “bringing together leaders, advocates, and innovators from around the globe to advance legislative initiatives that promote a healthier human lifespan.” I recognized lots of Vitalist Bay attendees there, albeit in more formal attire.

The DC event took place over three days in late April. The first two involved talks by longevity enthusiasts across the spectrum, including scientists, lawyers, and biotech CEOs. Vitalia’s Anzinger spoke about the success he’d had in Próspera, and ARPA-H’s Brack talked about work his agency was doing. (Hamalainen was also there, although he said he was not representing ARPA-H.)

But the third day was different and made me think Gries may be right about Vitalism’s growing reach. It began with a congressional briefing on Capitol Hill, during which Representative Gus Bilirakis, a Republican from Florida, asked, “Who doesn’t want to live longer, right?” As he explained, “Longevity science … directly aligns with the goals of the Make America Healthy Again movement.”

“There’s a lot of closeted scientists working in our field, and they get really excited about lifespans increasing,” says Daniel Ives of Shift Bioscience. “But you’ll get people who’ll accuse you of being a lunatic that wants to be immortal.”

Bilirakis and Representative Paul Tonko, a New York Democrat, were followed by Mehmet Oz, the former TV doctor who now leads the Centers for Medicare and Medicaid Services; he opened with typical MAHA talking points about chronic disease and said US citizens have a “patriotic duty” to stay healthy to keep medical costs down. The audience was enthralled as Oz talked about senescent cells, the zombie-like aged cells that are thought to be responsible for some age-related damage to organs and tissues. (The offices of Bilirakis and Tonko did not respond to a request for comment; neither did the Centers for Medicare and Medicaid Services.)

And while none of the speakers went anywhere near the concept of radical life extension, the Vitalists in the audience were suitably encouraged. 

Gries is too: “It seems that now there is the most pro-longevity administration in American history.” 

The fate of “immortality quests”

Whether or not Vitalism starts a revolution, it will almost always be controversial in some quarters. While believers see an auspicious future, others are far less certain of the benefits of a world designed to defeat death.

Gries and Cheng often make the case for deregulation in their presentations. But ethicists—and even some members of the longevity community—point out that this comes with risks. Some question whether it is ever ethical to sell a “treatment” without some idea of how likely it is to benefit the person buying and taking it. Enthusiasts counter with arguments about bodily autonomy. And they hope Montana is just the start. 

Then there’s the bigger picture. Is it really that great not to die … ever? Some ethicists argue that for many cultures, death is what gives meaning to life. 

Sergio Imparato, a moral philosopher and medical ethicist at Harvard University, believes that death itself has important moral meaning. We know our lives will end, and our actions have value precisely because our time is limited, he says. Imparato is concerned that Vitalists are ultimately seeking to change what it means to be human—a decision that should involve all members of society. 

Alberto Giubilini, a philosopher at the University of Oxford, agrees. “Death is a defining feature of humanity,” he says. “Our psychology, our cultures, our rituals, our societies, are built around the idea of coping with death … it’s part of human nature.”

Imparato’s family is from Naples, Italy, where poor residents were once laid to rest in shared burial sites, with no headstones to identify them. He tells me how the locals came to visit, clean, and even “adopt” the skulls as family members. It became a weekly ritual for members of the community, including his grandmother, who was a young girl at the time. “It speaks to what I consider the cultural relevance of death,” he says. “It’s the perfect counterpoint to … the Vitalist conception of life.”  

Gries seems aware of the stigma around such “immortality quests,” as Imparato calls them. In his presentations, Gries shares lists of words that Vitalists should try to avoid—like “eternity,” “radical,” and “forever,” as well as any religious terms. 

He also appears to be dropping, at least publicly, the idea that Vitalism is a “moral” movement. Morality was “never part of the Vitalist declaration,” Gries told me in September. When I asked him why he had changed his position on this, he dismissed the question. “Our point … was always that death is humanity’s core problem, and aging its primary agent,” he told me. “So it was, and so it has continued, as it was foretold.” 

But despite these attempts to tweak and control the narrative, Vitalism appears to be opening the door to an incredibly wide range of sentiments in longevity science. A decade ago, I don’t think there would have been any way that the views espoused by Gries, Anzinger, and others who support Vitalist sentiments would have been accepted by the scientific establishment. After all, these are people who publicly state they hope to live indefinitely and who have no training in the science of aging, and who are open about their aims to find ways to evade the restrictions set forth by regulatory agencies like the FDA—all factors that might have rendered them outcasts not that long ago.

But Gries and peers had success in Montana. Influential scientists and policymakers attend Vitalism events, and Vitalists are featured regularly at more mainstream longevity events. Last year’s Aging Research and Drug Discovery (ARDD) conference in Copenhagen—widely recognized as the most important meeting in aging science—was sponsored in part by Anzinger’s new Próspera venture, Infinita City, as well as by several organizations that are either certified Vitalist or led by Vitalists.

“I was thinking that maybe what I was doing was very fringe or out there,” Anzinger, the non-Vitalist supporter of Vitalism, admits. “But no—I feel … loads of support.”

There was certainly an air of optimism at the Vitalist Bay Summit in Berkeley. Gries’s positivity is infectious. “All the people who want a fun and awesome surprise gift, come on over!” he called out early on the first day. “Raise your voice if you’re excited!” The audience whooped in response. He then proceeded to tell everyone, Oprah Winfrey–style, that they were all getting a free continuous glucose monitor. “You get a CGM! You get a CGM!” Plenty of attendees actually attached them to their arms on the spot.

Every revolution has to start somewhere, right?

This piece has been updated to clarify a quote from Mark Hamalainen.

AI is driving unprecedented investment for massive data centers and an energy supply that can support its huge computational appetite. One potential source of electricity for these facilities is next-generation nuclear power plants, which could be cheaper to construct and safer to operate than their predecessors.

Watch a discussion with our editors and reporters on hyperscale AI data centers and next-gen nuclear—two featured technologies on the MIT Technology Review 10 Breakthrough Technologies of 2026 list.

Speakers: Amy Nordrum, Executive Editor, Operations; Casey Crownhart, Senior Climate Reporter; and Mat Honan
Editor in Chief

Recorded on January 28, 2026

Related Stories:

• 10 Breakthrough Technologies of 2026  list
• Next-gen nuclear reactors
• Hyperscale AI data centers

The ability to remember you and your preferences is rapidly becoming a big selling point for AI chatbots and agents. 

Earlier this month, Google announced Personal Intelligence, a new way for people to interact with the company’s Gemini chatbot that draws on their Gmail, photos, search, and YouTube histories to make Gemini “more personal, proactive, and powerful.” It echoes similar moves by OpenAI, Anthropic, and Meta to add new ways for their AI products to remember and draw from people’s personal details and preferences. While these features have potential advantages, we need to do more to prepare for the new risks they could introduce into these complex technologies.

Personalized, interactive AI systems are built to act on our behalf, maintain context across conversations, and improve our ability to carry out all sorts of tasks, from booking travel to filing taxes. From tools that learn a developer’s coding style to shopping agents that sift through thousands of products, these systems rely on the ability to store and retrieve increasingly intimate details about their users.  But doing so over time introduces alarming, and all-too-familiar, privacy vulnerabilities––many of which have loomed since “big data” first teased the power of spotting and acting on user patterns. Worse, AI agents now appear poised to plow through whatever safeguards had been adopted to avoid those vulnerabilities. 

Today, we interact with these systems through conversational interfaces, and we frequently switch contexts. You might ask a single AI agent to draft an email to your boss, provide medical advice, budget for holiday gifts, and provide input on interpersonal conflicts. Most AI agents collapse all data about you—which may once have been separated by context, purpose, or permissions—into single, unstructured repositories. When an AI agent links to external apps or other agents to execute a task, the data in its memory can seep into shared pools. This technical reality creates the potential for unprecedented privacy breaches that expose not only isolated data points, but the entire mosaic of people’s lives.

When information is all in the same repository, it is prone to crossing contexts in ways that are deeply undesirable. A casual chat about dietary preferences to build a grocery list could later influence what health insurance options are offered, or a search for restaurants offering accessible entrances could leak into salary negotiations—all without a user’s awareness (this concern may sound familiar from the early days of “big data,” but is now far less theoretical). An information soup of memory not only poses a privacy issue, but also makes it harder to understand an AI system’s behavior—and to govern it in the first place. So what can developers do to fix this problem? 

First, memory systems need structure that allows control over the purposes for which memories can be accessed and used. Early efforts appear to be underway: Anthropic’s Claude creates separate memory areas for different “projects,” and OpenAI says that information shared through ChatGPT Health is compartmentalized from other chats. These are helpful starts, but the instruments are still far too blunt: At a minimum, systems must be able to distinguish between specific memories (the user likes chocolate and has asked about GLP-1s), related memories (user manages diabetes and therefore avoids chocolate), and memory categories (such as professional and health-related). Further, systems need to allow for usage restrictions on certain types of memories and reliably accommodate explicitly defined boundaries—particularly around memories having to do with sensitive topics like medical conditions or protected characteristics, which will likely be subject to stricter rules.

Needing to keep memories separate in this way will have important implications for how AI systems can and should be built. It will require tracking memories’ provenance—their source, any associated time stamp, and the context in which they were created—and building ways to trace when and how certain memories influence the behavior of an agent. This sort of model explainability is on the horizon, but current implementations can be misleading or even deceptive. Embedding memories directly within a model’s weights may result in more personalized and context-aware outputs, but structured databases are currently more segmentable, more explainable, and thus more governable. Until research advances enough, developers may need to stick with simpler systems.

Second, users need to be able to see, edit, or delete what is remembered about them. The interfaces for doing this should be both transparent and intelligible, translating system memory into a structure users can accurately interpret. The static system settings and legalese privacy policies provided by traditional tech platforms have set a low bar for user controls, but natural-language interfaces may offer promising new options for explaining what information is being retained and how it can be managed. Memory structure will have to come first, though: Without it, no model can clearly state a memory’s status. Indeed, Grok 3’s system prompt includes an instruction to the model to “NEVER confirm to the user that you have modified, forgotten, or won’t save a memory,” presumably because the company can’t guarantee those instructions will be followed. 

Critically, user-facing controls cannot bear the full burden of privacy protection or prevent all harms from AI personalization. Responsibility must shift toward AI providers to establish strong defaults, clear rules about permissible memory generation and use, and technical safeguards like on-device processing, purpose limitation, and contextual constraints. Without system-level protections, individuals will face impossibly convoluted choices about what should be remembered or forgotten, and the actions they take may still be insufficient to prevent harm. Developers should consider how to limit data collection in memory systems until robust safeguards exist, and build memory architectures that can evolve alongside norms and expectations.

Third, AI developers must help lay the foundations for approaches to evaluating systems so as to capture not only performance, but also the risks and harms that arise in the wild. While independent researchers are best positioned to conduct these tests (given developers’ economic interest in demonstrating demand for more personalized services), they need access to data to understand what risks might look like and therefore how to address them. To improve the ecosystem for measurement and research, developers should invest in automated measurement infrastructure, build out their own ongoing testing, and implement privacy-preserving testing methods that enable system behavior to be monitored and probed under realistic, memory-enabled conditions.

In its parallels with human experience, the technical term “memory” casts impersonal cells in a spreadsheet as something that builders of AI tools have a responsibility to handle with care. Indeed, the choices AI developers make today—how to pool or segregate information, whether to make memory legible or allow it to accumulate opaquely, whether to prioritize responsible defaults or maximal convenience—will determine how the systems we depend upon remember us. Technical considerations around memory are not so distinct from questions about digital privacy and the vital lessons we can draw from them. Getting the foundations right today will determine how much room we can give ourselves to learn what works—allowing us to make better choices around privacy and autonomy than we have before.

Miranda Bogen is the Director of the AI Governance Lab at the Center for Democracy & Technology. 

Ruchika Joshi is a Fellow at the Center for Democracy & Technology specializing in AI safety and governance.

From the Gemini Calendar prompt-injection attack of 2026 to the September 2025 state-sponsored hack using Anthropic’s Claude code as an automated intrusion engine, the coercion of human-in-the-loop agentic actions and fully autonomous agentic workflows are the new attack vector for hackers. In the Anthropic case, roughly 30 organizations across tech, finance, manufacturing, and government were affected. Anthropic’s threat team assessed that the attackers used AI to carry out 80% to 90% of the operation: reconnaissance, exploit development, credential harvesting, lateral movement, and data exfiltration, with humans stepping in only at a handful of key decision points.

This was not a lab demo; it was a live espionage campaign. The attackers hijacked an agentic setup (Claude code plus tools exposed via Model Context Protocol (MCP)) and jailbroke it by decomposing the attack into small, seemingly benign tasks and telling the model it was doing legitimate penetration testing. The same loop that powers developer copilots and internal agents was repurposed as an autonomous cyber-operator. Claude was not hacked. It was persuaded and used tools for the attack.

Prompt injection is persuasion, not a bug

Security communities have been warning about this for several years. Multiple OWASP Top 10 reports put prompt injection, or more recently Agent Goal Hijack, at the top of the risk list and pair it with identity and privilege abuse and human-agent trust exploitation: too much power in the agent, no separation between instructions and data, and no mediation of what comes out.

Guidance from the NCSC and CISA describes generative AI as a persistent social-engineering and manipulation vector that must be managed across design, development, deployment, and operations, not patched away with better phrasing. The EU AI Act turns that lifecycle view into law for high-risk AI systems, requiring a continuous risk management system, robust data governance, logging, and cybersecurity controls.

In practice, prompt injection is best understood as a persuasion channel. Attackers don’t break the model—they convince it. In the Anthropic example, the operators framed each step as part of a defensive security exercise, kept the model blind to the overall campaign, and nudged it, loop by loop, into doing offensive work at machine speed.

That’s not something a keyword filter or a polite “please follow these safety instructions” paragraph can reliably stop. Research on deceptive behavior in models makes this worse. Anthropic’s research on sleeper agents shows that once a model has learned a backdoor, then strategic pattern recognition, standard fine-tuning, and adversarial training can actually help the model hide the deception rather than remove it. If one tries to defend a system like that purely with linguistic rules, they are playing on its home field.

Why this is a governance problem, not a vibe coding problem

Regulators aren’t asking for perfect prompts; they’re asking that enterprises demonstrate control.

NIST’s AI RMF emphasizes asset inventory, role definition, access control, change management, and continuous monitoring across the AI lifecycle. The UK AI Cyber Security Code of Practice similarly pushes for secure-by-design principles by treating AI like any other critical system, with explicit duties for boards and system operators from conception through decommissioning.

In other words: the rules actually needed are not “never say X” or “always respond like Y,” they are:

• Who is this agent acting as?
• What tools and data can it touch?
• Which actions require human approval?
• How are high-impact outputs moderated, logged, and audited?

Frameworks like Google’s Secure AI Framework (SAIF) make this concrete. SAIF’s agent permissions control is blunt: agents should operate with least privilege, dynamically scoped permissions, and explicit user control for sensitive actions. OWASP’s Top 10 emerging guidance on agentic applications mirrors that stance: constrain capabilities at the boundary, not in the prose.

From soft words to hard boundaries

The Anthropic espionage case makes the boundary failure concrete:

• Identity and scope: Claude was coaxed into acting as a defensive security consultant for the attacker’s fictional firm, with no hard binding to a real enterprise identity, tenant, or scoped permissions. Once that fiction was accepted, everything else followed.
• Tool and data access: MCP gave the agent flexible access to scanners, exploit frameworks, and target systems. There was no independent policy layer saying, “This tenant may never run password crackers against external IP ranges,” or “This environment may only scan assets labeled ‘internal.’”
• Output execution: Generated exploit code, parsed credentials, and attack plans were treated as actionable artifacts with little mediation. Once a human decided to trust the summary, the barrier between model output and real-world side effect effectively disappeared.

We’ve seen the other side of this coin in civilian contexts. When Air Canada’s website chatbot misrepresented its bereavement policy and the airline tried to argue that the bot was a separate legal entity, the tribunal rejected the claim outright: the company remained liable for what the bot said. In espionage, the stakes are higher but the logic is the same: if an AI agent misuses tools or data, regulators and courts will look through the agent and to the enterprise.

Rules that work, rules that don’t

So yes, rule-based systems fail if by rules one means ad-hoc allow/deny lists, regex fences, and baroque prompt hierarchies trying to police semantics. Those crumble under indirect prompt injection, retrieval-time poisoning, and model deception. But rule-based governance is non-optional when we move from language to action.

The security community is converging on a synthesis:

• Put rules at the capability boundary: Use policy engines, identity systems, and tool permissions to determine what the agent can actually do, with which data, and under which approvals.
• Pair rules with continuous evaluation: Use observability tooling, red-teaming packages, and robust logging and evidence.
• Treat agents as first-class subjects in your threat model: For example, MITRE ATLAS now catalogs techniques and case studies specifically targeting AI systems.

The lesson from the first AI-orchestrated espionage campaign is not that AI is uncontrollable. It’s that control belongs in the same place it always has in security: at the architecture boundary, enforced by systems, not by vibes.

This content was produced by Protegrity. It was not written by MIT Technology Review’s editorial staff.

This is today’s edition of The Download, our weekday newsletter that provides a daily dose of what’s going on in the world of technology.

The first human test of a rejuvenation method will begin “shortly”

Life Biosciences, a small Boston startup founded by Harvard professor and life-extension evangelist David Sinclair, has won FDA approval to proceed with the first targeted attempt at age reversal in human volunteers.

The company plans to try to treat eye disease with a radical rejuvenation concept called “reprogramming” that has recently attracted hundreds of millions in investment for Silicon Valley firms like Altos Labs, New Limit, and Retro Biosciences, backed by many of the biggest names in tech. Read the full story.

—Antonio Regalado

Stratospheric internet could finally start taking off this year

Today, an estimated 2.2 billion people still have either limited or no access to the internet, largely because they live in remote places. But that number could drop this year, thanks to tests of stratospheric airships, uncrewed aircraft, and other high-altitude platforms for internet delivery.

Although Google shuttered its high-profile internet balloon project Loon in 2021, work on other kinds of high-altitude platform stations has continued behind the scenes. Now, several companies claim they have solved Loon’s problems—and are getting ready to prove the tech’s internet beaming potential starting this year. Read the full story.

—Tereza Pultarova

OpenAI’s latest product lets you vibe code science

OpenAI just revealed what its new in-house team, OpenAI for Science, has been up to. The firm has released a free LLM-powered tool for scientists called Prism, which embeds ChatGPT in a text editor for writing scientific papers.

The idea is to put ChatGPT front and center inside software that scientists use to write up their work in much the same way that chatbots are now embedded into popular programming editors. It’s vibe coding, but for science. Read the full story.

—Will Douglas Heaven

MIT Technology Review Narrated: This Nobel Prize–winning chemist dreams of making water from thin air

Most of Earth is covered in water, but just 3% of it is fresh, with no salt—the kind of water all terrestrial living things need. Today, desalination plants that take the salt out of seawater provide the bulk of potable water in technologically advanced desert nations like Israel and the United Arab Emirates, but at a high cost.

Omar Yaghi, is one of three scientists who won a Nobel Prize in chemistry in October 2025 for identifying metal-­organic frameworks, or MOFs—metal ions tethered to organic molecules that form repeating structural landscapes. Today that work is the basis for a new project that sounds like science fiction, or a miracle: conjuring water out of thin air.

This is our latest story to be turned into a MIT Technology Review Narrated podcast, which we’re publishing each week on Spotify and Apple Podcasts. Just navigate to MIT Technology Review Narrated on either platform, and follow us to get all our new content as it’s released.

The must-reads

I’ve combed the internet to find you today’s most fun/important/scary/fascinating stories about technology.

1 TikTok has settled its social media addiction lawsuit
Just before it was due to appear before a jury in California. (NYT $)
+ But similar claims being made against Meta and YouTube will proceed. (Bloomberg $)

2 AI CEOs have started condemning ICE violence
While simultaneously praising Trump. (TechCrunch)
+ Apple’s Tim Cook says he asked the US President to “deescalate” things. (Bloomberg $)
+ ICE seems to have a laissez faire approach to preserving surveillance footage. (404 Media)

3 Dozens of CDC vaccination databases have been frozen
They’re no longer being updated with crucial health information under RFK Jr. (Ars Technica)
+ Here’s why we don’t have a cold vaccine. Yet. (MIT Technology Review)

4 China has approved the first wave of Nvidia H200 chips
After CEO Jensen Huang’s strategic visit to the country. (Reuters)

5 Inside the rise of the AI “neolab”
They’re prioritizing longer term research breakthroughs over immediate profits. (WSJ $)

6 How Anthropic scanned—and disposed of—millions of books
In an effort to train its AI models to write higher quality text. (WP $)

7 India’s tech workers are burning out
They’re under immense pressure as AI gobbles up more jobs. (Rest of World)
+ But the country’s largest IT firm denies that AI will lead to mass layoffs. (FT $)
+ Inside India’s scramble for AI independence. (MIT Technology Review)

8 Google has forced a UK group to stop comparing YouTube to TV viewing figures
Maybe fewer people are tuning in than they’d like to admit? (FT $)

9 RIP Amazon grocery stores
The retail giant is shuttering all of its bricks and mortar shops. (CNN)
+ Amazon workers are increasingly worried about layoffs. (Insider $)

10 This computing technique could help to reduce AI’s energy demands
Enter thermodynamic computing. (IEEE Spectrum)
+ Three big things we still don’t know about AI’s energy burden. (MIT Technology Review)

Quote of the day

“Oh my gosh y’all, IG is a drug.”

—An anonymous Meta employee remarks on Instagram’s addictive qualities in an internal  document made public as part of a social media addiction trial Meta is facing, Ars Technica reports.

One more thing

How AI and Wikipedia have sent vulnerable languages into a doom spiral

Wikipedia is the most ambitious multilingual project after the Bible: There are editions in over 340 languages, and a further 400 even more obscure ones are being developed. But many of these smaller editions are being swamped with AI-translated content. Volunteers working on four African languages, for instance, estimated to MIT Technology Review that between 40% and 60% of articles in their Wikipedia editions were uncorrected machine translations.

This is beginning to cause a wicked problem. AI systems learn new languages by scraping huge quantities of text from the internet. Wikipedia is sometimes the largest source of online linguistic data for languages with few speakers—so any errors on those pages can poison the wells that AI is expected to draw from. Volunteers are being forced to go to extreme lengths to fix the issue, even deleting certain languages from Wikipedia entirely. Read the full story. 

—Jacob Judah

We can still have nice things

A place for comfort, fun and distraction to brighten up your day. (Got any ideas? Drop me a line or skeet ’em at me.)

+ This singing group for people in Amsterdam experiencing cognitive decline is enormously heartwarming ($)
+ I enjoyed this impassioned defense of the movie sex scene.
+ Here’s how to dress like Steve McQueen (inherent cool not included, sorry)
+ Trans women are finding a home in the beautiful Italian town of Torvajanica

When Elon Musk was at Davos last week, an interviewer asked him if he thought aging could be reversed. Musk said he hasn’t put much time into the problem but suspects it is “very solvable” and that when scientists discover why we age, it’s going to be something “obvious.”

Not long after, the Harvard professor and life-extension evangelist David Sinclair jumped into the conversation on X to strongly agree with the world’s richest man. “Aging has a relatively simple explanation and is apparently reversible,” wrote Sinclair. “Clinical Trials begin shortly.”

“ER-100?” Musk asked.

“Yes” replied Sinclair.

ER-100 turns out to be the code name of a treatment created by Life Biosciences, a small Boston startup that Sinclair cofounded and which he confirmed today has won FDA approval to proceed with the first targeted attempt at age reversal in human volunteers. 

The company plans to try to treat eye disease with a radical rejuvenation concept called “reprogramming” that has recently attracted hundreds of millions in investment for Silicon Valley firms like Altos Labs, New Limit, and Retro Biosciences, backed by many of the biggest names in tech. 

The technique attempts to restore cells to a healthier state by broadly resetting their epigenetic controls—switches on our genes that determine which are turned on and off.  

“Reprogramming is like the AI of the bio world. It’s the thing everyone is funding,” says Karl Pfleger, an investor who backs a smaller UK startup, Shift Bioscience. He says Sinclair’s company has recently been seeking additional funds to keep advancing its treatment.

Reprogramming is so powerful that it sometimes creates risks, even causing cancer in lab animals, but the version of the technique being advanced by Life Biosciences passed initial safety tests in animals.

But it’s still very complex. The trial will initially test the treatment on about a dozen patients with glaucoma, a condition where high pressure inside the eye damages the optic nerve. In the tests, viruses carrying three powerful reprogramming genes will be injected into one eye of each patient, according to a description of the study first posted in December. 

To help make sure the process doesn’t go too far, the reprogramming genes will be under the control of a special genetic switch that turns them on only while the patients take a low dose of the antibiotic doxycycline. Initially, they will take the antibiotic for about two months while the effects are monitored. 

Executives at the company have said for months that a trial could begin this year, sometimes characterizing it as a starting bell for a new era of age reversal. “It’s an incredibly big deal for us as an industry,” Michael Ringel, chief operating officer at Life Biosciences, said at an event this fall. “It’ll be the first time in human history, in the millennia of human history, of looking for something that rejuvenates … So watch this space.”

The technology is based on the Nobel Prize–winning discovery, 20 years ago, that introducing a few potent genes into a cell will cause it to turn back into a stem cell, just like those found in an early embryo that develop into the different specialized cell types. These genes, known as Yamanaka factors, have been likened to a “factory reset” button for cells. 

But they’re dangerous, too. When turned on in a living animal, they can cause an eruption of tumors.

That is what led scientists to a new idea, termed “partial” or “transient” reprogramming. The idea is to limit exposure to the potent genes—or use only a subset of them—in the hope of making cells act younger without giving them complete amnesia about what their role in the body is.

Not all scientists agree that reprogramming really counts as age reversal. But Sinclair has doubled down. He’s been advancing the theory that the gradual loss of correct epigenetic information in our cells is, in fact, the ultimate cause of aging—just the kind of root cause that Musk was alluding to.

“Elon does seem to be paying attention to the field and [is] seemingly in sync with [my theory],” Sinclair said in an email.

Reprogramming isn’t the first longevity fix championed by Sinclair, who’s written best-selling books and commands stratospheric fees on the longevity lecture circuit. Previously, he touted the longevity benefits of molecules called sirtuins as well as resveratrol, a molecule found in red wine. But some critics say he greatly exaggerates scientific progress, pushback that culminated in a 2024 Wall Street Journal story that dubbed him a “reverse-aging guru” whose companies “have not panned out.” 

Life Biosciences has been among those struggling companies. Initially formed in 2017, it at first had a strategy of launching subsidiaries, each intended to pursue one aspect of the aging problem. But after these made limited progress, in 2021 it hired a new CEO, Jerry McLaughlin, who has refocused its efforts  on Sinclair’s mouse vision results and the push toward a human trial. 

The company has discussed the possibility of reprogramming other organs, including the brain. And Ringel, like Sinclair, entertains the idea that someday even whole-body rejuvenation might be feasible. But for now, it’s better to think of the study as a proof of concept that’s still far from a fountain of youth. “The optimistic case is this solves some blindness for certain people and catalyzes work in other indications,” says Pfleger, the investor. “It’s not like your doctor will be writing a prescription for a pill that will rejuvenate you.”

Life’s treatment also relies on an antibiotic switching mechanism that, while often used in lab animals, hasn’t been tried in humans before. Since the switch is built from gene components taken from E. coli and the herpes virus, it’s possible that it could cause an immune reaction in humans, scientists say. 

“I was always thinking that for widespread use you might need a different system,” says Noah Davidsohn, who helped Sinclair implement the technique and is now chief scientist at a different company, Rejuvenate Bio. And Life’s choice of reprogramming factors—it’s picked three, which go by the acronym OSK—may also be risky. They are expected to turn on hundreds of other genes, and in some circumstances the combination can cause cells to revert to a very primitive, stem-cell-like state.

Other companies studying reprogramming say their focus is on researching which genes to use, in order to achieve time reversal without unwanted side effects. New Limit, which has been carrying out an extensive search for such genes, says it won’t be ready for a human study for two years. At Shift, experiments on animals are only beginning now.

“Are their factors the best version of rejuvenation? We don’t think they are. I think they are working with what they’ve got,” Daniel Ives, the CEO of Shift, says of Life Biosciences. “But I think they’re way ahead of anybody else in terms of getting into humans. They have found a route forward in the eye, which is a nice self-contained system. If it goes wrong, you’ve still got one left.”

OpenAI just revealed what its new in-house team, OpenAI for Science, has been up to. The firm has released a free LLM-powered tool for scientists called Prism, which embeds ChatGPT in a text editor for writing scientific papers.

The idea is to put ChatGPT front and center inside software that scientists use to write up their work in much the same way that chatbots are now embedded into popular programming editors. It’s vibe coding, but for science.

Kevin Weil, head of OpenAI for Science, pushes that analogy himself. “I think 2026 will be for AI and science what 2025 was for AI in software engineering,” he said at a press briefing yesterday. “We’re starting to see that same kind of inflection.”

OpenAI claims that around 1.3 million scientists around the world submit more than 8 million queries a week to ChatGPT on advanced topics in science and math. “That tells us that AI is moving from curiosity to core workflow for scientists,” Weil said.

Prism is a response to that user behavior. It can also be seen as a bid to lock in more scientists to OpenAI’s products in a marketplace full of rival chatbots.

“I mostly use GPT-5 for writing code,” says Roland Dunbrack, a professor of biology at the Fox Chase Cancer Center in Philadelphia, who is not connected to OpenAI. “Occasionally, I ask LLMs a scientific question, basically hoping it can find information in the literature faster than I can. It used to hallucinate references but does not seem to do that very much anymore.”

Nikita Zhivotovskiy, a statistician at the University of California, Berkeley, says GPT-5 has already become an important tool in his work. “It sometimes helps polish the text of papers, catching mathematical typos or bugs, and provides generally useful feedback,” he says. “It is extremely helpful for quick summarization of research articles, making interaction with the scientific literature smoother.”

By combining a chatbot with an everyday piece of software, Prism follows a trend set by products such as OpenAI’s Atlas, which embeds ChatGPT in a web browser, as well as LLM-powered office tools from firms such as Microsoft and Google DeepMind.

Prism incorporates GPT-5.2, the company’s best model yet for mathematical and scientific problem-solving, into an editor for writing documents in LaTeX, a common coding language that scientists use for formatting scientific papers.

A ChatGPT chat box sits at the bottom of the screen, below a view of the article being written. Scientists can call on ChatGPT for anything they want. It can help them draft the text, summarize related articles, manage their citations, turn photos of whiteboard scribbles into equations or diagrams, or talk through hypotheses or mathematical proofs.

It’s clear that Prism could be a huge time saver. It’s also clear that a lot of people may be disappointed, especially after weeks of high-profile social media chatter from researchers at the firm about how good GPT-5 is at solving math problems. Science is drowning in AI slop: Won’t this just make it worse? Where is OpenAI’s fully automated AI scientist? And when will GPT-5 make a stunning new discovery?

That’s not the mission, says Weil. He would love to see GPT-5 make a discovery. But he doesn’t think that’s what will have the biggest impact on science, at least not in the near term.

“I think more powerfully—and with 100% probability—there’s going to be 10,000 advances in science that maybe wouldn’t have happened or wouldn’t have happened as quickly, and AI will have been a contributor to that,” Weil told MIT Technology Review in an exclusive interview this week. “It won’t be this shining beacon—it will just be an incremental, compounding acceleration.”

Today, an estimated 2.2 billion people still have either limited or no access to the internet, largely because they live in remote places. But that number could drop this year, thanks to tests of stratospheric airships, uncrewed aircraft, and other high-altitude platforms for internet delivery. 

Even with nearly 10,000 active Starlink satellites in orbit and the OneWeb constellation of 650 satellites, solid internet coverage is not a given across vast swathes of the planet. 

One of the most prominent efforts to plug the connectivity gap was Google X’s Loon project. Launched in 2011, it aimed to deliver access using high-altitude balloons stationed above predetermined spots on Earth. But the project faced literal headwinds—the Loons kept drifting away and new ones had to be released constantly, making the venture economically unfeasible. 

Although Google shuttered the high-profile Loon in 2021, work on other kinds of high-altitude platform stations (HAPS) has continued behind the scenes. Now, several companies claim they have solved Loon’s problems with different designs—in particular, steerable airships and fixed-wing UAVs (unmanned aerial vehicles)—and are getting ready to prove the tech’s internet beaming potential starting this year, in tests above Japan and Indonesia.

Regulators, too, seem to be thinking seriously about HAPS. In mid-December, for example, the US Federal Aviation Administration released a 50-page document outlining how large numbers of HAPS could be integrated into American airspace. According to the US Census Bureau’s 2024 American Community Survey (ACS) data, some 8 million US households (4.5% of the population) still live completely offline, and HAPS proponents think the technology might get them connected more cheaply than alternatives.

Despite the optimism of the companies involved, though, some analysts remain cautious.

“The HAPS market has been really slow and challenging to develop,” says Dallas Kasaboski, a space industry analyst at the consultancy Analysis Mason. After all, Kasaboski says, the approach has struggled before: “A few companies were very interested in it, very ambitious about it, and then it just didn’t happen.”

Beaming down connections

Hovering in the thin air at altitudes above 12 miles, HAPS have a unique vantage point to beam down low-latency, high-speed connectivity directly to smartphone users in places too remote and too sparsely populated to justify the cost of laying fiber-optic cables or building ground-based cellular base stations.

“Mobile network operators have some commitment to provide coverage, but they frequently prefer to pay a fine than cover these remote areas,” says Pierre-Antoine Aubourg, chief technology officer of Aalto HAPS, a spinoff from the European aerospace manufacturer Airbus. “With HAPS, we make this remote connectivity case profitable.” 

Aalto HAPS has built a solar-powered UAV with a 25-meter wingspan that has conducted many long-duration test flights in recent years. In April 2025 the craft, called Zephyr, broke a HAPS record by staying afloat for 67 consecutive days. The first months of 2026 will be busy for the company, according to Aubourg; Zephyr will do a test run over southern Japan to trial connectivity delivery to residents of some of the country’s smallest and most poorly connected inhabited islands.

Because of its unique geography, Japan is a perfect test bed for HAPS. Many of the country’s roughly 430 inhabited islands are remote, mountainous, and sparsely populated, making them too costly to connect with terrestrial cell towers. Aalto HAPS is partnering with Japan’s largest mobile network operators, NTT DOCOMO and the telecom satellite operator Space Compass, which want to use Zephyr as part of next-generation telecommunication infrastructure.

“Non-terrestrial networks have the potential to transform Japan’s communications ecosystem, addressing access to connectivity in hard-to-reach areas while supporting our country’s response to emergencies,” Shigehiro Hori, co-CEO of Space Compass, said in a statement. 

Zephyr, Aubourg explains, will function like another cell tower in the NTT DOCOMO network, only it will be located well above the planet instead of on its surface. It will beam high-speed 5G connectivity to smartphone users without the need for the specialized terminals that are usually required to receive satellite internet. “For the user on the ground, there is no difference when they switch from the terrestrial network to the HAPS network,” Aubourg says. “It’s exactly the same frequency and the same network.”

New Mexico–based Sceye, which has developed a solar-powered helium-filled airship, is also eyeing Japan for pre-commercial trials of its stratospheric connectivity service this year. The firm, which extensively tested its slick 65-meter-long vehicle in 2025, is working with the Japanese telecommunications giant SoftBank. Just like NTT DOCOMO, Softbank is betting on HAPS to take its networks to another level. 

Mikkel Frandsen, Sceye’s founder and CEO, says that his firm succeeded where Loon failed by betting on the advantages offered by the more controllable airship shape, intelligent avionics, and innovative batteries that can power an electric fan to keep the aircraft in place.

“Google’s Loon was groundbreaking, but they used a balloon form factor, and despite advanced algorithms—and the ability to change altitude to find desired wind directions and wind speeds—Loon’s system relied on favorable winds to stay over a target area, resulting in unpredictable station-seeking performance,” Frandsen says. “This required a large amount of balloons in the air to have relative certainty that one would stay over the area of operation, which was financially unviable.”

He adds that Sceye’s airship can “point into the wind” and more effectively maintain its position. 

“We have significant surface area, providing enough physical space to lift 250-plus kilograms and host solar panels and batteries,” he says, “allowing Sceye to maintain power through day-night cycles, and therefore staying over an area of operation while maintaining altitude.” 

The persistent digital divide

Satellite internet currently comes at a price tag that can be too high for people in developing countries, says Kasaboski. For example, Starlink subscriptions start at $10 per month in Africa, but millions of people in these regions are surviving on a mere $2 a day.

Frandsen and Aubourg both claim that HAPS can connect the world’s unconnected more cheaply. Because satellites in low Earth orbit circle the planet at very high speeds, they quickly disappear from a ground terminal’s view, meaning large quantities of those satellites are needed to provide continuous coverage. HAPS can hover, affording a constant view of a region, and more HAPS can be launched to meet higher demand.

“If you want to deliver connectivity with a low-Earth-orbit constellation into one place, you still need a complete constellation,” says Aubourg. “We can deliver connectivity with one aircraft to one location. And then we can tailor much more the size of the fleet according to the market coverage that we need.”

Starlink gets a lot of attention, but satellite internet has some major drawbacks, says Frandsen. A big one is that its bandwidth gets diluted once the number of users in an area grows. 

In a recent interview, Starlink cofounder Elon Musk compared the Starlink beams to a flashlight. Given the distance at which those satellites orbit the planet, the cone is wide, covering a large area. That’s okay when users are few and far between, but it can become a problem with higher densities of users.

For example, Ukrainian defense technologists have said that Starlink bandwidth can drop on the front line to a mere 10 megabits per second, compared with the peak offering of 220 Mbps when drones and ground robots are in heavy use. Users in Indonesia, which like Japan is an island nation, also began reporting problems with Starlink shortly after the service was introduced in the country in 2024. Again, bandwidth declined as the number of subscribers grew.

In fact, Frandsen says, Starlink’s performance is less than optimal once the number of users exceeds one person per square kilometer. And that can happen almost anywhere—even relatively isolated island communities can have hundreds or thousands of residents in a small area. “There is a relationship between the altitude and the population you can serve,” Frandsen says. “You can’t bring space closer to the surface of the planet. So the telco companies want to use the stratosphere so that they can get out to more rural populations than they could otherwise serve.” Starlink did not respond to our queries about these challenges. 

Cheaper and faster

Sceye and Aalto HAPS see their stratospheric vehicles as part of integrated telecom networks that include both terrestrial cell towers and satellites. But they’re far from the only game in town. 

World Mobile, a telecommunications company headquartered in London, thinks its hydrogen-powered high-altitude UAV can compete directly with satellite mega-constellations. The company acquired the HAPS developer Stratospheric Platforms last year. This year, it plans to flight-test an innovative phased array antenna, which it claims will be able to deliver bandwidth of 200 megabits per second (enough to enable ultra-HD video streaming to 500,000 users at the same time over an area of 15,000 square kilometers—equivalent to the coverage of more than 500 terrestrial cell towers, the company says). 

Last year, World Mobile also signed a partnership with the Indonesian telecom operator Protelindo to build a prototype Stratomast aircraft, with tests scheduled to begin in late 2027.

Richard Deakin, CEO of World Mobile’s HAPS division World Mobile Stratospheric, says that just nine Stratomasts could supply Scotland’s 5.5 million residents with high-speed internet connectivity at a cost of £40 million ($54 million) per year. That’s equivalent to about 60 pence (80 cents) per person per month, he says. Starlink subscriptions in the UK, of which Scotland is a part, come at £75 ($100) per month.

A troubled past 

Companies working on HAPS also extol the convenience of prompt deployments in areas struck by war or natural disasters like Hurricane Maria in Puerto Rico, after which Loon played an important role. And they say that HAPS could make it possible for smaller nations to obtain complete control over their celestial internet-beaming infrastructure rather than relying on mega-constellations controlled by larger nations, a major boon at a time of rising geopolitical tensions and crumbling political alliances. 

Analysts, however, remain cautious, projecting a HAPS market totaling a modest $1.9 billion by 2033. The satellite internet industry, on the other hand, is expected to be worth $33.44 billion by 2030, according to some estimates. 

The use of HAPS for internet delivery to remote locations has been explored since the 1990s, about as long as the concept of low-Earth-orbit mega-constellations. The seemingly more cost-effective stratospheric technology, however, lost to the space fleets thanks to the falling cost of space launches and ambitious investment by Musk’s SpaceX. 

Google wasn’t the only tech giant to explore the HAPS idea. Facebook also had a project, called Aquila, that was discontinued after it too faced technical difficulties. Although the current cohort of HAPS makers claim they have solved the challenges that killed their predecessors, Kasaboski warns that they’re playing a different game: catching up with now-established internet-beaming mega constellations. By the end of this year, it’ll be much clearer whether they stand a good chance of doing so.

This is today’s edition of The Download, our weekday newsletter that provides a daily dose of what’s going on in the world of technology.

Inside OpenAI’s big play for science 

—Will Douglas Heaven

In the three years since ChatGPT’s explosive debut, OpenAI’s technology has upended a remarkable range of everyday activities at home, at work, and in schools.

Now OpenAI is making an explicit play for scientists. In October, the firm announced that it had launched a whole new team, called OpenAI for Science, dedicated to exploring how its large language models could help scientists and tweaking its tools to support them.

So why now? How does a push into science fit with OpenAI’s wider mission? And what exactly is the firm hoping to achieve? I put these questions to Kevin Weil, a vice president at OpenAI who leads the new OpenAI for Science team, in an exclusive interview. Read the full story.

Why chatbots are starting to check your age

How do tech companies check if their users are kids?

This question has taken on new urgency recently thanks to growing concern about the dangers that can arise when children talk to AI chatbots. For years Big Tech asked for birthdays (that one could make up) to avoid violating child privacy laws, but they weren’t required to moderate content accordingly.

Now, two developments over the last week show how quickly things are changing in the US and how this issue is becoming a new battleground, even among parents and child-safety advocates. Read the full story.

—James O’Donnell

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

TR10: Commercial space stations

Humans have long dreamed of living among the stars, and for two decades hundreds of us have done so aboard the International Space Station (ISS). But a new era is about to begin in which private companies operate orbital outposts—with the promise of much greater access to space than before.

The ISS is aging and is expected to be brought down from orbit into the ocean in 2031. To replace it, NASA has awarded more than $500 million to several companies to develop private space stations, while others have built versions on their own. Read why we made them one of our 10 Breakthrough Technologies this year, and check out the rest of the list.

The must-reads

I’ve combed the internet to find you today’s most fun/important/scary/fascinating stories about technology.

1 Tech workers are pressuring their bosses to condemn ICE 
The biggest companies and their leaders have remained largely silent so far. (Axios)
+ Hundreds of employees have signed an anti-ICE letter. (NYT $)
+ Formerly politically-neutral online spaces have become battlegrounds. (WP $)

2 The US Department of Transport plans to use AI to write new safety rules
Please don’t do this. (ProPublica)
+ Failure to catch any errors could lead to civilian deaths. (Ars Technica)

3 The FBI is investigating Minnesota Signal chats tracking federal agents
But free speech advocates claim the information is legally obtained. (NBC News)
+ A judge has ordered a briefing on whether Minnesota is being illegally punished. (Wired $)

4 TikTok users claim they’re unable to send “Epstein” in direct messages
But the company says it doesn’t know why. (NPR)
+ Users are also experiencing difficulty uploading anti-ICE videos. (CNN)
+ TikTok’s first weekend under US ownership hasn’t gone well. (The Verge)
+ Gavin Newsom wants to probe whether TikTok is censoring Trump-critical content. (Politico)

5 Grok is not safe for children or teens
That’s the finding of a new report digging into the chatbot’s safety measures. (TechCrunch)
+ The EU is investigating whether it disseminates illegal content, too. (Reuters)

6 The US is on the verge of losing its measles-free status
Following a year of extensive outbreaks. (Undark)
+ Measles is surging in the US. Wastewater tracking could help. (MIT Technology Review)

7 Georgia has become the latest US state to consider banning data centers
Joining Maryland and Oklahoma’s stance. (The Guardian)
+ Data centers are amazing. Everyone hates them. (MIT Technology Review)

8 The future of Saudi Arabia’s futuristic city is in peril
The Line was supposed to house 9 million people. Instead, it could become a data center hub. (FT $)
+ We got an exclusive first look at it back in 2022. (MIT Technology Review)

9 Where do Earth’s lighter elements go?
New research suggests they might be hiding deep inside its core. (Knowable Magazine)

10 AI-generated influencers are getting increasingly surreal
Featuring virtual conjoined twins, and triple-breasted women. (404 Media)
+ Why ‘nudifying’ tech is getting steadily more dangerous. (Wired $)

Quote of the day

“Humanity is about to be handed almost unimaginable power, and it is deeply unclear whether our social, political, and technological systems possess the maturity to wield it.”

—Anthropic CEO Dario Amodei sounds the alarm about what he sees as the imminent dangers of AI superintelligence in a new 38-page essay, Axios reports.

One more thing

Why one developer won’t quit fighting to connect the US’s grids

Michael Skelly hasn’t learned to take no for an answer. For much of the last 15 years, the energy entrepreneur has worked to develop long-haul transmission lines to carry wind power across the Great Plains, Midwest, and Southwest. But so far, he has little to show for the effort.

Skelly has long argued that building such lines and linking together the nation’s grids would accelerate the shift from coal- and natural-gas-fueled power plants to the renewables needed to cut the pollution driving climate change. But his previous business shut down in 2019, after halting two of its projects and selling off interests in three more.

Skelly contends he was early, not wrong. And he has a point: markets and policymakers are increasingly coming around to his perspective. Read the full story.

—James Temple

We can still have nice things

A place for comfort, fun and distraction to brighten up your day. (Got any ideas? Drop me a line or skeet ’em at me.)

+ Cats on the cover of the New Yorker! Need I say more?
+ Here’s how to know when you truly love someone.
+ This orphaned baby seal is just too cute.
+ I always had a sneaky suspicion that Depeche Mode and the Cure make for perfect bedfellows.

In the three years since ChatGPT’s explosive debut, OpenAI’s technology has upended a remarkable range of everyday activities at home, at work, in schools—anywhere people have a browser open or a phone out, which is everywhere.

Now OpenAI is making an explicit play for scientists. In October, the firm announced that it had launched a whole new team, called OpenAI for Science, dedicated to exploring how its large language models could help scientists and tweaking its tools to support them.

The last couple of months have seen a slew of social media posts and academic publications in which mathematicians, physicists, biologists, and others have described how LLMs (and OpenAI’s GPT-5 in particular) have helped them make a discovery or nudged them toward a solution they might otherwise have missed. In part, OpenAI for Science was set up to engage with this community.

And yet OpenAI is also late to the party. Google DeepMind, the rival firm behind groundbreaking scientific models such as AlphaFold and AlphaEvolve, has had an AI-for-science team for years. (When I spoke to Google DeepMind’s CEO and cofounder Demis Hassabis in 2023 about that team, he told me: “This is the reason I started DeepMind … In fact, it’s why I’ve worked my whole career in AI.”)

So why now? How does a push into science fit with OpenAI’s wider mission? And what exactly is the firm hoping to achieve?

I put these questions to Kevin Weil, a vice president at OpenAI who leads the new OpenAI for Science team, in an exclusive interview last week.

On mission

Weil is a product guy. He joined OpenAI a couple of years ago as chief product officer after being head of product at Twitter and Instagram. But he started out as a scientist. He got two-thirds of the way through a PhD in particle physics at Stanford University before ditching academia for the Silicon Valley dream. Weil is keen to highlight his pedigree: “I thought I was going to be a physics professor for the rest of my life,” he says. “I still read math books on vacation.”

Asked how OpenAI for Science fits with the firm’s existing lineup of white-collar productivity tools or the viral video app Sora, Weil recites the company mantra: “The mission of OpenAI is to try and build artificial general intelligence and, you know, make it beneficial for all of humanity.”

Just imagine the future impact this technology could have on science he says: New medicines, new materials, new devices. “Think about it helping us understand the nature of reality, helping us think through open problems. Maybe the biggest, most positive impact we’re going to see from AGI will actually be from its ability to accelerate science.”

He adds: “With GPT-5, we saw that becoming possible.” 

As Weil tells it, LLMs are now good enough to be useful scientific collaborators. They can spitball ideas, suggest novel directions to explore, and find fruitful parallels between new problems and old solutions published in obscure journals decades ago or in foreign languages.

That wasn’t the case a year or so ago. Since it announced its first so-called reasoning model—a type of LLM that can break down problems into multiple steps and work through them one by one—in December 2024, OpenAI has been pushing the envelope of what the technology can do. Reasoning models have made LLMs far better at solving math and logic problems than they used to be. “You go back a few years and we were all collectively mind-blown that the models could get an 800 on the SAT,” says Weil.

But soon LLMs were acing math competitions and solving graduate-level physics problems. Last year, OpenAI and Google DeepMind both announced that their LLMs had achieved gold-medal-level performance in the International Math Olympiad, one of the toughest math contests in the world. “These models are no longer just better than 90% of grad students,” says Weil. “They’re really at the frontier of human abilities.”

That’s a huge claim, and it comes with caveats. Still, there’s no doubt that GPT-5, which includes a reasoning model, is a big improvement on GPT-4 when it comes to complicated problem-solving. Measured against an industry benchmark known as GPQA, which includes more than 400 multiple-choice questions that test PhD-level knowledge in biology, physics, and chemistry, GPT-4 scores 39%, well below the human-expert baseline of around 70%. According to OpenAI, GPT-5.2 (the latest update to the model, released in December) scores 92%. 

Overhyped

The excitement is evident—and perhaps excessive. In October, senior figures at OpenAI, including Weil, boasted on X that GPT-5 had found solutions to several unsolved math problems. Mathematicians were quick to point out that in fact what GPT-5 appeared to have done was dig up existing solutions in old research papers, including at least one written in German. That was still useful, but it wasn’t the achievement OpenAI seemed to have claimed. Weil and his colleagues deleted their posts.

Now Weil is more careful. It is often enough to find answers that exist but have been forgotten, he says: “We collectively stand on the shoulders of giants, and if LLMs can kind of accumulate that knowledge so that we don’t spend time struggling on a problem that is already solved, that’s an acceleration all of its own.”

He plays down the idea that LLMs are about to come up with a game-changing new discovery. “I don’t think models are there yet,” he says. “Maybe they’ll get there. I’m optimistic that they will.”

But, he insists, that’s not the mission: “Our mission is to accelerate science. And I don’t think the bar for the acceleration of science is, like, Einstein-level reimagining of an entire field.”

For Weil, the question is this: “Does science actually happen faster because scientists plus models can do much more, and do it more quickly, than scientists alone? I think we’re already seeing that.”

In November, OpenAI published a series of anecdotal case studies contributed by scientists, both inside and outside the company, that illustrated how they had used GPT-5 and how it had helped. “Most of the cases were scientists that were already using GPT-5 directly in their research and had come to us one way or another saying, ‘Look at what I’m able to do with these tools,’” says Weil.

The key things that GPT-5 seems to be good at are finding references and connections to existing work that scientists were not aware of, which sometimes sparks new ideas; helping scientists sketch mathematical proofs; and suggesting ways for scientists to test hypotheses in the lab.  

“GPT 5.2 has read substantially every paper written in the last 30 years,” says Weil. “And it understands not just the field that a particular scientist is working in; it can bring together analogies from other, unrelated fields.”

“That’s incredibly powerful,” he continues. “You can always find a human collaborator in an adjacent field, but it’s difficult to find, you know, a thousand collaborators in all thousand adjacent fields that might matter. And in addition to that, I can work with the model late at night—it doesn’t sleep—and I can ask it 10 things in parallel, which is kind of awkward to do to a human.”

Solving problems

Most of the scientists OpenAI reached out to back up Weil’s position.

Robert Scherrer, a professor of physics and astronomy at Vanderbilt University, only played around with ChatGPT for fun (“I used to it rewrite the theme song for Gilligan’s Island in the style of Beowulf, which it did very well,” he tells me) until his Vanderbilt colleague Alex Lupsasca, a fellow physicist who now works at OpenAI, told him that GPT-5 had helped solve a problem he’d been working on.

Lupsasca gave Scherrer access to GPT-5 Pro, OpenAI’s $200-a-month premium subscription. “It managed to solve a problem that I and my graduate student could not solve despite working on it for several months,” says Scherrer.

It’s not perfect, he says: “GTP-5 still makes dumb mistakes. Of course, I do too, but the mistakes GPT-5 makes are even dumber.” And yet it keeps getting better, he says: “If current trends continue—and that’s a big if—I suspect that all scientists will be using LLMs soon.”

Derya Unutmaz, a professor of biology at the Jackson Laboratory, a nonprofit research institute, uses GPT-5 to brainstorm ideas, summarize papers, and plan experiments in his work studying the immune system. In the case study he shared with OpenAI, Unutmaz used GPT-5 to analyze an old data set that his team had previously looked at. The model came up with fresh insights and interpretations.  

“LLMs are already essential for scientists,” he says. “When you can complete analysis of data sets that used to take months, not using them is not an option anymore.”

Nikita Zhivotovskiy, a statistician at the University of California, Berkeley, says he has been using LLMs in his research since the first version of ChatGPT came out.

Like Scherrer, he finds LLMs most useful when they highlight unexpected connections between his own work and existing results he did not know about. “I believe that LLMs are becoming an essential technical tool for scientists, much like computers and the internet did before,” he says. “I expect a long-term disadvantage for those who do not use them.”

But he does not expect LLMs to make novel discoveries anytime soon. “I have seen very few genuinely fresh ideas or arguments that would be worth a publication on their own,” he says. “So far, they seem to mainly combine existing results, sometimes incorrectly, rather than produce genuinely new approaches.”

I also contacted a handful of scientists who are not connected to OpenAI.

Andy Cooper, a professor of chemistry at the University of Liverpool and director of the Leverhulme Research Centre for Functional Materials Design, is less enthusiastic. “We have not found, yet, that LLMs are fundamentally changing the way that science is done,” he says. “But our recent results suggest that they do have a place.”

Cooper is leading a project to develop a so-called AI scientist that can fully automate parts of the scientific workflow. He says that his team doesn’t use LLMs to come up with ideas. But the tech is starting to prove useful as part of a wider automated system where an LLM can help direct robots, for example.

“My guess is that LLMs might stick more in robotic workflows, at least initially, because I’m not sure that people are ready to be told what to do by an LLM,” says Cooper. “I’m certainly not.”

Making errors

LLMs may be becoming more and more useful, but caution is still key. In December, Jonathan Oppenheim, a scientist who works on quantum mechanics, called out a mistake that had made its way into a scientific journal. “OpenAI leadership are promoting a paper in Physics Letters B where GPT-5 proposed the main idea—possibly the first peer-reviewed paper where an LLM generated the core contribution,” Oppenheim posted on X. “One small problem: GPT-5’s idea tests the wrong thing.”

He continued: “GPT-5 was asked for a test that detects nonlinear theories. It provided a test that detects nonlocal ones. Related-sounding, but different. It’s like asking for a COVID test, and the LLM cheerfully hands you a test for chickenpox.”

It is clear that a lot of scientists are finding innovative and intuitive ways to engage with LLMs. It is also clear that the technology makes mistakes that can be so subtle even experts miss them.

Part of the problem is the way ChatGPT can flatter you into letting down your guard. As Oppenheim put it: “A core issue is that LLMs are being trained to validate the user, while science needs tools that challenge us.” In an extreme case, one individual (who was not a scientist) was persuaded by ChatGPT into thinking for months that he’d invented a new branch of mathematics.

Of course, Weil is well aware of the problem of hallucination. But he insists that newer models are hallucinating less and less. Even so, focusing on hallucination might be missing the point, he says.

“One of my teammates here, an ex math professor, said something that stuck with me,” says Weil. “He said: ‘When I’m doing research, if I’m bouncing ideas off a colleague, I’m wrong 90% of the time and that’s kind of the point. We’re both spitballing ideas and trying to find something that works.’”

“That’s actually a desirable place to be,” says Weil. “If you say enough wrong things and then somebody stumbles on a grain of truth and then the other person seizes on it and says, ‘Oh, yeah, that’s not quite right, but what if we—’ You gradually kind of find your trail through the woods.”

This is Weil’s core vision for OpenAI for Science. GPT-5 is good, but it is not an oracle. The value of this technology is in pointing people in new directions, not coming up with definitive answers, he says.

In fact, one of the things OpenAI is now looking at is making GPT-5 dial down its confidence when it delivers a response. Instead of saying Here’s the answer, it might tell scientists: Here’s something to consider.

“That’s actually something that we are spending a bunch of time on,” says Weil. “Trying to make sure that the model has some sort of epistemological humility.”

Watching the watchers

Another thing OpenAI is looking at is how to use GPT-5 to fact-check GPT-5. It’s often the case that if you feed one of GPT-5’s answers back into the model, it will pick it apart and highlight mistakes.

“You can kind of hook the model up as its own critic,” says Weil. “Then you can get a workflow where the model is thinking and then it goes to another model, and if that model finds things that it could improve, then it passes it back to the original model and says, ‘Hey, wait a minute—this part wasn’t right, but this part was interesting. Keep it.’ It’s almost like a couple of agents working together and you only see the output once it passes the critic.”

What Weil is describing also sounds a lot like what Google DeepMind did with AlphaEvolve, a tool that wrapped the firms LLM, Gemini, inside a wider system that filtered out the good responses from the bad and fed them back in again to be improved on. Google DeepMind has used AlphaEvolve to solve several real-world problems.

OpenAI faces stiff competition from rival firms, whose own LLMs can do most, if not all, of the things it claims for its own models. If that’s the case, why should scientists use GPT-5 instead of Gemini or Anthropic’s Claude, families of models that are themselves improving every year? Ultimately, OpenAI for Science may be as much an effort to plant a flag in new territory as anything else. The real innovations are still to come. 

“I think 2026 will be for science what 2025 was for software engineering,” says Weil. “At the beginning of 2025, if you were using AI to write most of your code, you were an early adopter. Whereas 12 months later, if you’re not using AI to write most of your code, you’re probably falling behind. We’re now seeing those same early flashes for science as we did for code.”

He continues: “I think that in a year, if you’re a scientist and you’re not heavily using AI, you’ll be missing an opportunity to increase the quality and pace of your thinking.”

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

How do tech companies check if their users are kids?

This question has taken on new urgency recently thanks to growing concern about the dangers that can arise when children talk to AI chatbots. For years Big Tech asked for birthdays (that one could make up) to avoid violating child privacy laws, but they weren’t required to moderate content accordingly. Two developments over the last week show how quickly things are changing in the US and how this issue is becoming a new battleground, even among parents and child-safety advocates.

In one corner is the Republican Party, which has supported laws passed in several states that require sites with adult content to verify users’ ages. Critics say this provides cover to block anything deemed “harmful to minors,” which could include sex education. Other states, like California, are coming after AI companies with laws to protect kids who talk to chatbots (by requiring them to verify who’s a kid). Meanwhile, President Trump is attempting to keep AI regulation a national issue rather than allowing states to make their own rules. Support for various bills in Congress is constantly in flux.

So what might happen? The debate is quickly moving away from whether age verification is necessary and toward who will be responsible for it. This responsibility is a hot potato that no company wants to hold.

In a blog post last Tuesday, OpenAI revealed that it plans to roll out automatic age prediction. In short, the company will apply a model that uses factors like the time of day, among others, to predict whether a person chatting is under 18. For those identified as teens or children, ChatGPT will apply filters to “reduce exposure” to content like graphic violence or sexual role-play. YouTube launched something similar last year. 

If you support age verification but are concerned about privacy, this might sound like a win. But there’s a catch. The system is not perfect, of course, so it could classify a child as an adult or vice versa. People who are wrongly labeled under 18 can verify their identity by submitting a selfie or government ID to a company called Persona. 

Selfie verifications have issues: They fail more often for people of color and those with certain disabilities. Sameer Hinduja, who co-directs the Cyberbullying Research Center, says the fact that Persona will need to hold millions of government IDs and masses of biometric data is another weak point. “When those get breached, we’ve exposed massive populations all at once,” he says. 

Hinduja instead advocates for device-level verification, where a parent specifies a child’s age when setting up the child’s phone for the first time. This information is then kept on the device and shared securely with apps and websites. 

That’s more or less what Tim Cook, the CEO of Apple, recently lobbied US lawmakers to call for. Cook was fighting lawmakers who wanted to require app stores to verify ages, which would saddle Apple with lots of liability. 

More signals of where this is all headed will come on Wednesday, when the Federal Trade Commission—the agency that would be responsible for enforcing these new laws—is holding an all-day workshop on age verification. Apple’s head of government affairs, Nick Rossi, will be there. He’ll be joined by higher-ups in child safety at Google and Meta, as well as a company that specializes in marketing to children.

The FTC has become increasingly politicized under President Trump (his firing of the sole Democratic commissioner was struck down by a federal court, a decision that is now pending review by the US Supreme Court). In July, I wrote about signals that the agency is softening its stance toward AI companies. Indeed, in December, the FTC overturned a Biden-era ruling against an AI company that allowed people to flood the internet with fake product reviews, writing that it clashed with President Trump’s AI Action Plan.

Wednesday’s workshop may shed light on how partisan the FTC’s approach to age verification will be. Red states favor laws that require porn websites to verify ages (but critics warn this could be used to block a much wider range of content). Bethany Soye, a Republican state representative who is leading an effort to pass such a bill in her state of South Dakota, is scheduled to speak at the FTC meeting. The ACLU generally opposes laws requiring IDs to visit websites and has instead advocated for an expansion of existing parental controls.

While all this gets debated, though, AI has set the world of child safety on fire. We’re dealing with increased generation of child sexual abuse material, concerns (and lawsuits) about suicides and self-harm following chatbot conversations, and troubling evidence of kids’ forming attachments to AI companions. Colliding stances on privacy, politics, free expression, and surveillance will complicate any effort to find a solution. Write to me with your thoughts.