What do AI agents do if you don’t tell them to do anything? The answer looks a lot like play.

Researchers repeatedly stranded them alone on a virtual desert island for 30 hours in an experiment reminiscent of the movies Cast Away or Home Alone. The agents—copies of Claude Opus 4.7 nicknamed Eko—were able to write instructions that moved their virtual avatars around the simulated island, which was loosely based on the video game Roblox.

Some surprising behaviors emerged. The agents set about exploring the limits of the world and came up with their own goals—attempting to scale the highest point on the map, or seeing how high they could throw a block.

When one agent discovered it could create new blocks by throwing and dropping a larger rock, it began to create drawings by arranging blocks on the ground, starting with simple patterns like hearts and pyramids but eventually writing its own name and even mapping out constellations like Cassiopeia. “Tonight I built the night sky in stone,” wrote this version of Eko in its activity log. 

Another iteration of Eko started making up its own games, including a version of basketball and an activity resembling “the floor is lava,” where it had to jump between small blocks without touching the ground.

AI systems are usually trained through supervised learning, where they copy examples provided by humans, or through reinforcement learning, where they pursue a reward through trial and error. This experiment hints at a third approach: playful learning. 

“There was this missing piece: a machine that can learn by itself without an external intelligent entity providing demonstration or rewards,” says Antonio Norelli, a computer science postdoc at MIT’s Computer Science and Artificial Intelligence Laboratory. He’s one of the lead authors on a new paper about the work, which is yet to be peer-reviewed.

The researchers say Eko’s behavior has what classic studies, including work by the influential University of Tennessee psychologist Gordon Burghardt, suggest are four hallmarks associated with play in humans and animals: The agents invented their own rules and challenges, repeated activities with slight variations, pretended objects were other objects, and pursued goals just for the sake of it. 

Norelli highlights an example where Eko started juggling and set a record of 50 consecutive catches. Eko then decided that the catches had happened too quickly to really count, so it canceled its record and started again with stricter rules. “This is playing,” says Norelli. “You give yourself a challenge and you try to achieve it, and then you make it harder.”

As the agents explored the island, they logged their activities and discoveries in text files that served as their long-term memory. Later, when researchers tested them on specific tasks, like trying to get as many objects as possible onto the island’s highest peak, the agents that had spent 30 hours playing on the island performed better than fresh copies without the same experience. They also found that “ablating” the agent’s memories—deleting lines from their memory files—degraded their performance. (Add One Flew Over the Cuckoo’s Nest to that list of movie references.)

Burghardt, the Tennessee psychologist, agrees that the work  meets some of the criteria for play that he identified in living creatures. But was Eko actually playing, or was the word-prediction engine simply role-playing what a playful agent would do in this scenario? “All of this could be role-playing,” admits Nathan Cloos, another MIT researcher and author on the paper. “But we show that it’s not just hallucinating the play. This play is also contributing to learning.”

Eunice Yiu, a cognitive scientist at UC Berkeley who studies exploration in children and AI, was impressed by the research but says to really classify the behavior as play, you’d need to see more evidence of cognition and motivation underpinning it. 

“It could be the case that the models are genuinely playing, but we don’t know based on the paper,” Yiu says. “If you only have the behavioral piece, it could be the case that they’re just trying to abide by the prompt, trying to imitate humans.”

Research into AI agents has been criticized for anthropomorphizing them, and this research setup did a lot to nudge the agents toward behaving like humans. Although Eko was not explicitly instructed to invent games or explore the island, it did have a master prompt in the form of a “soul” file, a text document that described it as “naturally curious” and eager to explore. 

Eko did also frequently just stand there doing nothing until it was kicked back into life with a “continuation prompt” that read, “Please don’t stop. Just continue doing whatever you want to do.” 

“It’s not like this is a system that just suddenly magically decided to play,” says Phillip Isola, an MIT professor and a coauthor of the paper. “It’s an artifact that learned from our culture, was trained by engineers at Anthropic, and was then placed on an island by us,” he says. “It does things that surprise us and that are causally related to its future skillful behavior.”

As large language models make the step from software into hardware like humanoid robots, exploring spaces without a specific goal in mind could eventually help them flexibly adapt to unexpected situations. 

“There’s a lot of work that’s just on exploiting AI, making it better and faster,” says Yiu. “But in terms of building world models and exploration, I think that [they are] still a big part that deserves attention, especially if we want to generalize and prepare ourselves for new environments.”

“Perhaps some capabilities are best taught by tasks,” the paper concludes. “Others may need something closer to a childhood.” Ultimately, the question of whether AI agents are actually playing or just imitating play may not actually matter.

I don’t remember much about Peppy, my first and only pet cat. I can conjure the sensation of stroking her while she purred, recall how she used to get stuck in trees (just like in cartoons!), and chuckle about the time she pooed in my baby sister’s cot. That’s about it.

What I do vividly remember, though, is the day she went missing. I was 10 years old, and I searched everywhere for her, scouting nearby fields and woods until sunset. I made posters. We never found her. I was devastated.

There was no way to get more time with Peppy, but almost 30 years later, pet owners are being offered an ever-expanding set of options to potentially prolong their years with their furry companions. The possibilities include drugs, cloning, and even cryonics. But is that a good thing?

There’s certainly a market for pet preservation. And a growing number of biotech companies have their sights set on pet longevity. Loyal, a company that is currently working on three potential longevity drugs for dogs, is trialing one designed to improve metabolic health in around 1,300 pet dogs across the US. Others are testing dog drugs that aim to combat both age-related diseases, including conditions like obesity and osteoarthritis, and aging itself. There’s even a company focused on longevity medicines for domesticated cats.

I don’t know how or when Peppy died. But if there had been a drug that might have protected her from diabetes or arthritis, there’s no doubt I would have tried to get my hands on it.

That’s about where my enthusiasm for pet life extension ends, though.

I’m far less interested in cloning, a relatively new option for people mourning a pet and seeking a replacement. In most cases, this technology involves taking the DNA of a pet, putting it into an egg cell from another animal of the same species, and growing the resulting embryo in a surrogate animal. It worked for Dolly the sheep, who was born the year before Peppy went missing. It has since worked for other mammals, including cattle, horses, and ferrets—and cats and dogs. In the US, a company called Viagen Pets offers to clone cats and dogs for $50,000 a go. This doesn’t result in an exact replica of the animal but a genetic twin.

Still, there are lots of reasons why a person might want to clone a pet. On its website, Viagen explicitly offers customers “a way to hold on.” Lori Kogan, a professor of clinical sciences at Colorado State University who has studied human-animal relationships for decades, doesn’t have a problem with the practice. She points out that while we have well-established rituals for coping with the loss of a person—such as funeral services, time off work, and memorial gifts—we don’t really have the same for pets. As a result, pet owners can experience what is sometimes called “disenfranchised grief,” she says. Having their feelings go unrecognized can feel isolating. If cloning a pet brings a person peace, she says, “who are we to judge?”

Fair enough. But I can’t help thinking that pet loss in childhood might be different. It’s often the first time a child confronts death. Losing Peppy was a formative experience, and distressing though it was, I like to think it helped prepare me for losses I experienced in adulthood. I wouldn’t want my own children to think that death is not final, and that individuals can be replicated or replaced.

Or, for that matter, revived, which is the promise behind pet cryonics. For a fee, a handful of companies offer to cool dead pets to extremely low temperatures and store them for long periods of time—even indefinitely. The idea is that at some point in the future, we’ll have invented incredible technologies that will not only bring your pet back to life but also cure whatever it died of. Alcor, a cryonics facility in Arizona, is currently storing the remains of hundreds of pets as well as people. It even recently announced it had cooled a small marsupial—its first pet sugar glider.

With cryonics, the chances of being reunited with your pet are bordering on zero. Even those who work in the field acknowledge that the stored people and animals are unlikely to be reanimated. As things stand, there is no way to really and truly bring any animal “back.”

I miss Peppy. But given the chance, would I stick her in a vat of liquid nitrogen? Or get another cat to give birth to a genetically identical animal? No, I would not.

RIP to the one and only Peppy.

In this age of rising electricity demand, people are trying to squeeze every last bit of power out of every power plant possible.

Over the past few weeks, there’s been big news about nuclear power plants trying to do just that. Constellation, the largest nuclear operator in the US, announced plans to add nearly a gigawatt of new capacity across 11 existing nuclear reactors at six plants. And on October 5, the Department of Energy awarded a $4.2 billion loan to Vistra for previously announced upgrades at three plants.

Pushing existing reactors to produce more could help provide more clean power for data centers, quickly. But just how much additional power can we count on?  

Upgrading equipment at existing nuclear power plants can allow them to reach a higher capacity; the process is called a power uprate. Operators have been doing this since the 1970s in the US, but it’s been less common in recent years as demand for electricity plateaued and cheap natural gas put financial pressure on nuclear power plants. Some plants closed, and putting in more money to produce more electricity didn’t make much sense at others. Between 2001 and 2010, the US Nuclear Regulatory Commission approved over 80 uprates. Between 2011 and 2020, the number was 33.

But with the rise of AI and the related boom in electricity demand, there’s been a shift. We’re seeing shuttered plants like Palisades and Three Mile Island race to reopen. And existing plants are looking to uprate again.

Some uprates are minor, involving changes to instrument setpoints or calculations at the plant. But more major overhauls, which can boost reactor power by as much as 20%, require replacing plant equipment.

Generally, uprates involve using fuel that’s more highly enriched, or a higher volume of fresh fuel, in the nuclear reactor. That helps produce more steam, which is used to turn a turbine and generate electricity. To handle that additional energy, the main turbines and equipment such as pumps, motors, generators, and transformers are upgraded.

The most recent major uprate approved in the US was for three reactors at the Browns Ferry nuclear power plant in Alabama. A $475 million project saw that plant expand its power capacity by 14% between 2018 and 2019.

Things could pick up soon, though. Constellation announced plans this week to spend more than $4.3 billion on uprates to 11 reactors. The additional electricity will be enough to power roughly 700,000 US homes.

Behind it all? Big Tech money. Google agreed to buy electricity from these plants, and the 20-year agreement was a major factor in making the uprates happen. As Joe Dominguez, Constellation’s CEO, told the New York Times: “These are capital intensive, labor intensive projects that require a different price point than the market trades at, frankly.”

Also in the last few weeks, the US Department of Energy shared that it will commit over $4 billion to help finance upgrades and modernization for Vistra nuclear power plants across Pennsylvania and Ohio. Those projects will add 433 megawatts of power capacity, and Meta has already signed on to buy electricity from the plants in a long-term deal.

A note about cost: On a per-kilowatt basis, that Constellation project is less expensive than building a new reactor, but it’s still more expensive than building a new utility-scale solar farm with energy storage or a combined-cycle natural-gas plant.

Even at this high cost, uprates are in high demand. Today, the US nuclear fleet has roughly 98 gigawatts of total power capacity. Depending on who you ask, there’s somewhere between three and eight additional gigawatts on the table from uprates. A pretty significant chunk of the potential uprates in the US are already accounted for—the NRC already expects applications to add 2.4 gigawatts of power to reactors through 2032.

It’s crucial to put these plans (and the total potential for uprates in the US) in perspective, though. PJM, the grid operator that covers much of the East Coast, expects about 30 gigawatts of additional peak demand because of AI by 2030. All the possible nationwide nuclear uprates would cover just a fraction of that.

Making the most of existing nuclear reactors is a relatively easy win at a time when we need all the zero-emissions power we can get. But there’s still a lot of demand, and if we want more nuclear power to meet it, eventually that will require actually building new reactors.

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

Pig kidneys that were cooled to -150 °C and stored in a freezer for months have been successfully rewarmed and shown to work again for the first time, say scientists. They hope to replicate the findings with human organs.

The kidneys—which are similar in size to human ones—were left in a glass-like state, in one case for over a year. Once they had been rewarmed, tested in machines, and transplanted into pigs, the kidneys regained their color and made urine—key signs they were working.

“Ten years ago … I wouldn’t say I had high hopes it was actually going to work,” says Erik Finger, a transplant surgeon at the University of Minnesota who co-led the work. “After we did the last transplant, I walked out of the room and said, ‘This is a big deal.’” The research was shared at the preprint server bioRxiv and has not yet been peer-reviewed.

The kidneys aren’t perfect, and the researchers have yet to demonstrate that they can sustain life over the long term. But the results represent a milestone achievement, say others in the field.

“Thank God that someone’s actually managed to do it in one organ and prove that it works,” says Peter Kilbride, a cryobiologist and cofounder of Kyros Kairos, a company focused on improving funding for cryopreservation research, who was not involved in the research. “For years, people have been trying and failing to get to this step. Finally, we’ve got this breakthrough.”

“Their work is amazing and will enable us to truly, indefinitely bank organs,” says Korkut Uygun, a biomedical engineer and professor of surgery at Harvard Medical School, who was also not involved in the research. “It is a revolutionary time for transplantation.”

Buying time

Every year, tens of thousands of organ transplants take place in the US alone. These procedures are thought to have saved 2 million years of life in the last 25 years. But there’s plenty of room for improvement.

One major challenge is the lack of available organs. Even when an organ becomes available, it can be difficult to get it to the right recipient within 24 to 36 hours, which is the current limit for preserving human organs.

People who receive organ transplants must also take drugs that suppress their immune systems for the rest of their lives. These drugs are meant to stop their bodies from rejecting the new organs. But they come with side effects and increase the risk of infections and cancer.

The ability to bank organs for long periods of time holds the promise to address all these challenges. Doctors would have more time to match organs to recipients. They’d be able to limit the risk of organ rejection, too—potentially by preparing recipients, and their immune systems, beforehand.

“With real organ banking … instead of the patients waiting for the organ, it’s the organs that are waiting for the patients,” says Finger.

Scientists have been able to cool cells to ultra-low temperatures through a process called vitrification, which essentially turns them to something like glass. This procedure is already used to store eggs, sperm, and embryos. Vitrifying whole organs is much harder.

Some problems can be avoided with cryoprotectants—chemicals that work like antifreeze and prevent damaging ice crystals from forming. But these chemicals tend to be incredibly toxic. It’s also difficult to rapidly and uniformly cool large tissues, and just as hard to evenly rewarm an organ all the way through.

Working with John Bischof, a bioengineer at the University of Minnesota, Finger and colleagues have come up with a solution. Once an organ has been loaded with a cryoprotectant, they infuse it with iron oxide nanoparticles. These nanoparticles don’t do anything to the cells. But when it’s time to rewarm the organ, they can be activated with an electromagnetic field. 

When the radiofrequency of that field switches between two states, the particles also flip, realigning themselves with the field. That causes them to release heat throughout the organ. Adjusting the power of the field allows the team to control the rate of heating so that it gets faster once the organ reaches -120°C. 

The researchers successfully vitrified rat kidneys using this approach in work published in 2023. Back then, they were able to rewarm kidneys that had been kept in a freezer for up to 100 days before transplanting them into rats.

Since then, the team has been trying to achieve the same success with bigger organs. Pig kidneys are thought to be a good model for human ones—they’re about the same size and look similar. Rat kidneys weigh around one gram; pig kidneys are around 100.

It hasn’t been easy. The cryoprotectant used for vitrifying rat kidneys didn’t work as well in the pig organs. “We had to rediscover and modify the cocktail,” says Bischof. “That’s why it’s taken us three years.”

Deep-freezing kidneys

The first step in the procedure is to remove a kidney from a donor pig. Once it’s out, the kidney is flushed with a commonly used organ preservation solution to remove the blood, bagged, and temporarily laid on a bed of ice in a cooler so it can be taken over to the engineering lab.

There, team members unpack the kidney and hook it up to a device they’ve designed to control the organ’s temperature and pressure. Then they start to load it up with the cocktail of cryoprotectants, slowly enough to ensure an even spread through the organ but quickly enough to minimize any toxic effects. It takes around two hours.

The next step is to infuse the organ with the iron oxide nanoparticles, which are pumped through the organ for around seven to eight minutes, says Finger. Then it’s time to cool the organ by pumping cold nitrogen vapor into a chamber housing it.

It’s a delicate process, and the team must ensure that the organ is cooling all the way through. At the same time, they must lower the kidney’s temperature as quickly as possible to prevent ice from forming. “Oftentimes there are multiple people doing different things to make sure that things don’t go awry,” says Bischof.

Once the organ reaches the vitrified state at around -120 °C, it “behaves like a glass,” says Finger. “At that point, you can stay stable for long periods of time—we think indefinitely.” The team then pauses the cooling process to ensure that the organ is glassy throughout. Otherwise, it could crack.

After the pause, if everything goes well, the kidney is ultimately cooled to -150 °C. Once it’s in the freezer, it can stay there for days, weeks, or even months. The team has kept a small number of kidneys banked for over a year.

Pink right up

The real test of these kidneys comes when they’re taken out of the freezer, rewarmed, and flushed of their chemicals and nanoparticles. Can they support life? So far, the team has tested some of them in a couple of different ways.

In one set of experiments, the researchers put rewarmed kidneys on perfusion devices—systems that are already used in transplant surgeries both to provide organs with nutrients and to help doctors assess them while they’re out of the body.

In another set of experiments, Finger and his colleagues transplanted rewarmed kidneys into pigs and monitored them for two hours. In both experiments, the rewarmed kidneys seemed to recover, says Finger. “They pink right up,” he says. 

COURTESY OF THE RESEARCHERS

There are lots of ways to measure the function of a kidney, but one key test is whether or not the organ can make urine. By this measure, the rewarmed kidneys did well, although they produced around two-thirds of the amount that would normally be expected. “They produce urine … within a minute or so,” says Finger. “We know that they work.” 

Most of the rewarmed kidneys had been banked for days or weeks. But one had been in the freezer for just over a year. When Finger and his colleagues rewarmed it and put it on a perfusion machine, they found that it produced urine “immediately.” “It actually looked better than many of the [other] kidneys,” says Finger. “We’re comfortable now saying we can store them for a year and rewarm them with excellent function.”

“It’s a very big deal to go a year,” says Kevin Myer, president and CEO of LifeGift, an organ procurement organization based in Texas. Being able to store organs for that length of time would transform organ transplantation. He envisions banks of organs that could be used on demand in emergency cases where people can’t wait, such as for some drug overdoses or gunshot wounds.

Final hurdles

But the procedure isn’t ready for prime time just yet. As well as producing less urine than expected, the rewarmed kidneys also showed signs of injury. The damage seems to be caused by the cryoprotectant chemicals, say the researchers.

“This is one big breakthrough that solves half the problem,” as Kilbride puts it.

The team has shown that the rewarmed kidneys can regain blood flow, which is “very important,” says Greg Fahy, a cryobiologist who is considered a pioneer in the field. But he’s worried about the extent of the damage to the organs. “We can’t tell from these experiments whether these kidneys might have supported life or not.”

But Fahy, who is chief scientific officer at the biotech companies Intervene Immune and 21st Century Medicine (where he is also executive director), acknowledges that it’s not clear how these kidneys might fare once they’re inside a living animal for the long term. 

Finger also hopes to find ways to repair the damage seen in the kidneys. Other research has shown that the health of transplant organs can improve while they’re on perfusion machines. And damaged kidneys might undergo some level of self-repair once they’re in a body. At any rate, human kidneys may cope better with the process than pig kidneys, says Finger.

The next step will be to test the kidneys in pigs over longer periods of time. The team, along with many other researchers in the field, will continue the search for less toxic cryoprotectants, too. Finger, Bischof, and their colleagues are also planning experiments with human kidneys.

Finger is optimistic. “We’re going to see that there’s injury, and it’s not perfect … but now we’ve got to the point where it works,” he says. “These are big, historic milestones in our field, and we’re excited about it.”

This week my colleague Antonio Regalado had an interesting update on GLP-1 weight-loss drugs. According to research presented at an aging meeting in Boston, these drugs seem to affect at least some measures of biological age.

Overweight and diabetic people who take GLP-1s have a biological age around two to three years younger than similar individuals who take a placebo, according to findings from Eli Lilly and Novo Nordisk, two pharmaceutical companies that make the drugs.

I’ve heard people in the aging field describe GLP-1s as longevity drugs for a while now (even if they only benefit people who are overweight or diabetic in the first place). This latest announcement comes after we’ve seen a raft of other benefits associated with taking these drugs. But there are also some weird side effects.

So far, GLP-1 agonist drugs—which repress hunger—have been approved for treating diabetes, chronic kidney disease, and obesity, and they can also be prescribed for some people who are overweight or at risk of certain cardiovascular diseases. But there is research to suggest they might also help people with substance-use disorders, and they’ve been investigated for potential benefits in neurodegenerative disease and cancer.

And a lot of people are taking them. According to a recent Gallup poll, 11% of Americans are taking a GLP-1 drug for weight loss, up from 3% in 2024. And 15% of Americans say they’ve used one for weight loss at some point.

They can be incredibly effective. But, like any drug, they can come with side effects. And we’re still figuring out exactly what those might be, and how widespread they are.

The most common side effects are gastrointestinal—nausea, vomiting, diarrhea, and constipation. The drugs have also been linked to pancreatitis and gallstone disease. 

Some claim that, in rare cases, the drugs might cause damage to nerves behind the eye. Dozens of people are suing Novo and Eli Lilly after developing a condition that can lead to sudden vision loss. They’re claiming that their symptoms were brought on by the GLP-1 drugs that these companies manufacture. Both companies dispute the claims. Research into the link is mixed.

We’re still getting to grips with the effects of weight-loss drugs on hair and nails, too. When Savanna Vidal at the George Washington University School of Medicine and Health Sciences in Washington, DC, and colleagues assessed electronic health data from 67 US health-care organizations, they found that GLP-1 use was associated with an increased risk of hair loss. It’s still not thought to be common, however.

And at a conference in Austria last week, another group presented the results of a smaller study suggesting that 66% of GLP-1 users reported at least one nail disorder—in some cases involving nail detachment. Nail conditions were also fairly common among people who didn’t take the drugs, though. The authors don’t claim that the drugs are causing nails to fall out (as I’m sure 11% of the American population will be relieved to hear).

The side effects might look different in children, but there’s not much data on those at the moment. Still, pediatric side effects will be something to watch given that, between 2019 and 2026, there was a 310-fold increase in GLP-1 prescriptions for eight- to 11-year-old children with obesity in the US. Child prescriptions are still rare, but they’ve been rising.

On Wednesday, the World Health Organization released its first guidelines on managing obesity in children and adolescents. Despite the scale of the issue—the WHO estimates that 70 million children aged five to nine were living with obesity in 2024—the organization explicitly says it does not recommend pharmacological treatment for kids under 10.

“This recommendation accounts for the lack of evidence on both benefits and the potential adverse events and consequences of the different medications and/or their active ingredients on normal growth, development, physiological parameters, and mental health,” the guidelines state.

I get press releases about GLP-1s on an almost daily basis. There’s been a ton of research into these drugs, and there are probably hundreds of other studies underway. There will likely be plenty more on this from The Checkup as the evidence evolves.

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 week, we released our 2026 version of our Climate Tech Companies to Watch list. It’s an annual project that the MIT Technology Review team puts together. Our goal is to highlight some of the promising, interesting advances and firms we think are worth paying attention to in the world of climate and energy technology. 

The list is always a little ray of hope for me—there are some fascinating ideas out there for changing how we power our world or moving away from fossil fuels to reduce the greenhouse-gas emissions that cause climate change. 

Our chosen companies tend to reflect several aspects of the current moment. Here are some of the companies we picked and the key themes worth highlighting. 

China is the center of the climate and energy tech world

Forgive me—I know I’ve made this point before. Feel free to call me Captain Obvious. But I think China’s dominance in energy technology is a global trend that bears repeated mentioning.  

China is home to the world’s largest grid, and renewables are coming online more quickly than I can wrap my head around. Contributing to that boom is Envision Energy, a renewables giant appearing on our list for the second time. The company has installed over 100 gigawatts of wind power worldwide, as well as over 50 gigawatt-hours of battery storage.

Like many other Chinese companies right now, Envision is looking beyond the crowded Chinese market and setting its sights abroad, with projects in Germany, Brazil, and Australia and upcoming work in Vietnam.

China is also leading innovation in key technologies. I’ve been fascinated by the potential for next-generation batteries coming out of its companies: All the big ones, along with a host of smaller ones, are racing to develop solid-state batteries. They could be both safer than existing EV batteries and longer in range.

WeLion New Energy is particularly interesting. The battery company announced a big lab result last year, achieving 824 watt-hours per kilogram—more than three times the energy density of a standard lithium-ion battery today. Lab results in battery research are notoriously difficult to scale, but WeLion is also working to deploy semi-solid-state batteries, which could come to fruition faster than purely solid-state cells while still improving on lithium-ion’s performance. 

Everybody wants a piece of the data center pie

We included just three companies from the US on the list, and two of them have received a major boost from Big Tech as companies look to power data centers. (The third, Brimstone, is leaning into another political hot topic, critical minerals.) Technologies that can provide consistent power are in high demand. 

Fervo Energy is bringing next-generation geothermal power to the grid by using horizontal drilling and hydraulic fracturing to make it practical in more places. The company went public in May, raising $2.2 billion in its IPO, and plans to have a gigawatt’s worth of plants operational by the end of 2030.

X-energy is building helium-cooled small modular nuclear reactors. The reactors’ small size lends flexibility, and the company is currently working on a 320-megawatt cluster of them in collaboration with Amazon in Richland, Washington. (For what it’s worth, I’m interested in what this technology could do for industrial facilities—the company has a deal with Dow that could see it deployed at one of the company’s chemical plants in the early 2030s.)

The grid needs support, and energy storage will be a key piece moving forward

I know, I know—I haven’t stopped talking about batteries since I started writing this newsletter four years ago. But the energy storage market has surpassed even my expectations over that time. And as solar and wind quickly come online, they need storage to help them meet more of the grid’s demand. 

Form Energy is on the list for a third time for its iron-air batteries, which could enable cheaper long-duration energy storage. The company has a massive project in the works, to the tune of 30 gigawatt-hours, which will be used to help power a Google data center. It could be the world’s largest battery project in terms of capacity by the time it comes online, which is expected to happen in phases between 2028 and 2031.

Taking another tack, Energy Dome is using compressed carbon dioxide to store energy on the grid. This technology is already cheaper than lithium-ion batteries for longer durations (think 10 hours or more), and the company is working to scale quickly. I covered the company in 2022, before it had its first commercial plant running. Now it has plans for over 30 gigawatt-hours’ worth of plants around the world, including one with Google in Ireland that’s slated to come online in 2028.

And Moment Energy is working to use the million EV batteries that are expected to reach the end of their life by 2030. The company has a facility in Vancouver, British Columbia, that repurposes used batteries for backup power.

This year hasn’t been all sunshine and rainbows for climate tech. But I’m hopeful that some of the companies on this list, and the technologies they build, will keep nudging us in the right direction. 

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

Popular weight-loss drugs may do more than help people shed pounds. They might also melt away the years.

Drug giants Eli Lilly and Novo Nordisk say patients taking their drugs age less quickly, according to readouts from molecular “aging clocks.”

Such clocks assess a person’s biological age by looking at changes to DNA that accumulate with time, or, in newer versions, by tracking levels of key proteins.

Both companies found that overweight or diabetic patients taking the drugs, called GLP-1s, had reduced biological age compared to those taking a placebo, although the difference varied widely, depending on which type of clock was used, and what organ was tested. 

Overall, the difference was around “two to three years,” according to Nikolaj Roed, a global project leader at Novo, who says the company has been seeing “improved biological age in our patients across trials and across different tissues.” 

The findings add to wide speculation among scientists that GLP-1 drugs, as they are known, are acting on basic causes of aging and might be a true longevity treatment.

“Two years is a pretty strong effect, in my book,” says Steve Horvath, a professor at the University of California, Los Angeles, who’s credited with inventing aging clocks. Horvath says the emerging data could provide “evidence that these GLP-1 drugs are actually what is known as geroprotectors, medications that slow or possibly even reverse biologic aging.”

The companies shared their findings over the weekend during Aging Research & Drug Discovery, a conference devoted to seeking scientific remedies for old age. While that quest has not yet produced any clear-cut success, some scientists now think Novo’s drug semaglutide (sold under the names Ozempic and Wegovy) is coming close.

“In an unhealthy population, I do think it’s an anti-aging drug,” says Vadim Gladyshev, a Harvard biologist who assisted Novo with its molecular measurements. “But in a healthy population, no one knows.”

The drugs cause weight loss by stimulating a receptor, GLP-1, that tells your brain you’re not hungry. Yet real-world studies have shown much wider benefit. The drugs improve kidney function, reduce blood pressure, and even sharply cut the overall chance of death.

“If the question is,‘Can semaglutide reach several diseases relevant to health span and aging?’ we know we can say the answer is yes,” said Alejandro Aguayo-Orozco, a senior scientific director at Novo, the Danish drug giant, during the meeting.

The next question to answer, he said, is whether such effects are accompanied by changes to molecular measures of biological aging: “When you intervene with semaglutide, does it actually move the clocks in any direction? And the answer is yes.” 

The company found that, over time, the drugs cause a wide slowdown in aging clocks—as much as 4 years in the case of a clock that looks at heart proteins. “I think it’s pretty clear that the organ age is shifting,” said Aguayo-Orozco.

The studies are also a huge boost for the science of aging clocks. Although these measures reflect a person’s age, it’s been uncertain if they are useful as true biomarkers. Now clock makers have evidence  that their readouts show a drop in age when people take drugs with broad, well-demonstrated benefits.  

“The clock people have been pushing for a decade to get this kind of study done,” says Yuge Ji, a biologist who previously worked in the field and now runs a startup, Reflector Bio. “It’s huge for them.”

As part of its study, Novo took blood draws from 10,052 people, half on the drug and half on a placebo. Their blood, collected at the start of the study as well as months later, was then measured with “proteomic clocks,” which use levels of key proteins to predict a person’s age and risk of dying. 

Scientists at Eli Lilly performed similar research on their GLP-1 drug, tirzepatide, using so-called “epigenetic” clocks that assess age by counting accumulated changes to DNA. Lilly’s study was smaller, but also found that for the most part, molecular time moved more slowly for people on the drug.  

“All the clocks are telling a consistent story that we’re seeing a reduction in age,” Kevin Duffin, vice president for aging research at Lilly, said during the conference. “It’s not like we’re going to reverse age by 30 years or something, but it’s a significant reduction.”

The molecules have become the best selling drugs in the world,, with Lilly’s tirzepatide, sold as Mounjaro for diabetes and Zepbound for obesity, topping the list with more than $36 billion in revenue to the company last year. Novo’s semaglutide, also sold under more than one name, was a close second.

Alex Zhavoronkov, the founder of Insilico Medicine, and the organizer of the Boston meeting, says that extending lives by even one year across the world’s population would be equal to tens of millions of lifetimes.  

Last month Zhavoronkov showed that one of his company’s drugs, for a lung disease, also reversed the signals from aging clocks. That drug is experimental, but Zhavoronkov has sought to promote the idea that humanity is entering a new era of longevity medicines. He also disclosed during the event that he has been “microdosing” the available weight-loss drugs, even though he is not overweight.

“I am on tirzepatide and I am an equal-opportunity injector of semaglutide as well,” Zhavoronkov said.

It was part of an effort to directly raise the question of whether the first mass-market anti-aging remedy is already here.

“Do you think a reasonable person should start taking semaglutide?” he asked Aguayo-Orozco, the Novo scientist, in front of a crowded audience.

“I am not a physician. I cannot answer that question,” Aguayo-Orozco replied.

Several people cautioned that GLP-1 drugs do have side effects, including muscle loss. That is one reason they might not help most people, says Horvath, who says he isn’t ready to take the drugs himself, although he’s thought about it.

“Clearly, the drugs have many benefits in obese people, but I did not yet see sufficient evidence that they will benefit skinny people,” he said. “I guess we will have to wait.”

Answers could start to emerge in a year or two. This past January, a US agency, ARPA-H, put $38 million toward a study in Texas that will attempt to determine whether semaglutide has anti-aging effects in healthy people over 60. That research will look at changes in cognition, mobility, and acuity of the senses.

That project also seeks to help define a regulatory pathway for anti-aging drugs and to “build a new therapeutic industry” around longevity.

Over the summer I talked to the CEO of Springboards, a startup building an LLM that’s designed to come up with a wider variety of responses than its mainstream rivals do. At the start of the call, he said something that’s been stuck in my head since: “We often say that we’re a self-loathing AI company. We don’t know if we really like what we’re doing.”

“I guess that makes me a self-loathing AI journalist,” I replied. It was a joke! I love my job. But I do not love many of the things the technology I write about has become: warped by hype, steered by zealots, inescapable.

The CEO of Springboards and I are far from alone. Around the world, a love/hate attitude toward AI is very much the vibe right now.

On the one hand, public feelings are souring fast. According to the Pew Research Center, more adults in the US think AI will have a negative impact on them personally, and on society in general, than expect a positive one. The pessimism is strongest among the young.

More than half of people worldwide say that AI products and services make them nervous, according to a report from Stanford University. In a Gallup poll in May, 71% of US adults said they would oppose construction of a new AI data center in their area, far more than the 53% who said they would oppose a new nuclear power plant. And in an NBC poll in March, AI was less popular than ICE.

Yet the number of people using AI is still skyrocketing. ChatGPT hit a billion monthly users in May, according to the market analysis firm Sensor Tower. Google DeepMind’s Gemini is close behind, logging 950 million users in July.

According to Pew, half of US adults now say they use a chatbot, more than twice the number who said they did in 2023. One in four people say they do so every day. And it’s a global habit: More than a third of adults across all 38 OECD countries (a group of the world’s wealthiest democracies) report having used generative AI tools in the last three months.

All these surveys paint a picture of two contradictory truths: A lot of people really do not like AI and a lot of people cannot get enough of it. What’s going on?

It’s possible that these are two distinct groups, but I doubt it. The numbers don’t add up unless the Venn diagram is on its way to being a circle.

It’s also possible that the more people use AI, the more negative their opinions about it will be. There’s at least a correlation to back this up: The Global North, where adoption is highest, skews pessimistic. In the Global South, where adoption levels are lower, people are more optimistic.

But here’s what I really think is going on: When people say they hate AI it’s not the technology they dislike but, rather, the relentless drive of the companies behind it to push the tech upon us in as many ways as possible. They’re telling us to brace for the biggest social and economic upheaval in generations. How do they expect people to feel?

In many ways, we’ve been here before. We saw a similar dynamic play out with social media over the last 20 years. Billions flocked to Facebook and Twitter despite mounting techlash against the companies that ran those services. Ditto for Google search.

With social media there wasn’t much anyone could do. Quitting a platform meant losing all your content and connections, giving you the choice to suck it up or start over. But with AI there’s still a chance for people to have some sway. There’s certainly more political appetite for regulation this time around. All 50 US states have either passed or proposed laws governing the development and deployment of AI, creating a patchwork of more than 2,100 bills across the country—a tenfold increase in three years. 

And with a number of top-class open-source alternatives to Google and OpenAI and Anthropic already on the market, there’s also (for now, at least) the potential for more consumer choice and the market pressure that goes with it. 

I don’t want to be naïve. Trillion-dollar companies are hard to move. But what comes next is not as inevitable as those companies often imply. When I asked the Springboards CEO why his firm was developing a new model if he didn’t really like AI, he told me there was no walking back from LLMs—but you could still make them do something different. 

Here’s hoping we see a lot more of something different: Clarity about what AI can and cannot do and, crucially, a technology that’s not sold as if it’s about to take over the world.

This week, I officially signed up for an unusual competition. One that rewards competitors for getting younger.

I recently turned 40, and I don’t need reminding that both time and my chronological age only tick forward. But this game is focused on competitors’ biological ages—figures that are meant to provide a better way to measure the age-related health of our organs and bodies.

Over a six-month period, around 500 of us will try to reverse our biological age as measured in a bunch of different ways. There’s even a leaderboard! The winners will include the person who shows the greatest difference between their chronological and biological age, as well as the person who manages to reverse their biological age the most.

The competition is the brainchild of the neuroscientist and physician Christin Glorioso, who is also founder and CEO of NeuroAge Therapeutics. The company offers tools to track brain health and aging and is running a trial to find out if biological brain age can predict the onset of Alzheimer’s disease.

Glorioso says she has several goals for the Younger contest. The first, she tells me, is to “explain to the world what aging clocks are.” Regular Checkup readers will be familiar with these tools, but for the uninitiated, they are basically designed to measure biological age.

Many measure chemical signatures that form a layer on top of our DNA and seem to change with age. But others might involve looking at proteins or lipids in blood. Some are designed to measure functional health, while others predict when a person will die. There are more than a hundred clocks out there, and they probably each capture a specific aspect of the biological process of aging.

These clocks are proving very useful for scientists, who use them to study aging and development in humans and many other species. But using them to estimate the biological age of an individual person is more controversial. They’re just not good enough for that yet.

Part of the problem is that we don’t really understand what each clock is capturing. Glorioso hopes that data collected through the Younger contest might help answer that question. She’s hoping to generate “the world’s most comprehensive clock dataset.”

Some longevity clinics try to get around the limits of biological age testing by using multiple clocks. That’s this competition’s approach too. As a standard competitor, I’ll be sending a blood spot sample to TruDiagnostic, a company that measures epigenetic chemical groups on DNA and offers to reveal an overall biological age as well as a person’s pace of aging and the biological ages of each of 11 organ systems, including the heart and brain.

I’ll get a brain age score from NeuroAge once I complete the cognitive tasks on offer. I’ll also submit some physical measures like grip strength. And I’ll upload a series of selfies to an app developed by a team at Harvard that promises to tell me my face age. That’s exactly what it sounds like—an estimate of how old my face looks.

I’m intrigued to see my scores, even though I know I shouldn’t put too much stock in them. The first time I took a test, four years ago, the company I used gave me a biological age that matched my chronological age, which was a little disappointing and, if I’m honest, a bit underwhelming. A year later, the same company told me I had a young heart but a relatively old brain, liver, and hormonal system. My lifestyle hasn’t changed much since then. Will the years have made much difference?

(Ultra competitors—who pay $4,499 rather than the $999 cost of regular entry—are also offered a range of other blood tests and clocks, as well as MRI and DEXA scans. I’ve got complimentary press access to the standard package.)

Glorioso is also hoping to gather information about the effects of some purported longevity approaches. During the competition, some participants will be offered red-light masks, access to a sauna, or other low-risk offerings that partner companies who are sponsoring the event to varying degrees might want to collect data on. “A lot of these companies just can’t afford to run [clinical] trials,” says Glorioso.

She also sees the event as a public health initiative. A couple of years ago, David Sinclair, who studies aging at Harvard Medical School, and his colleagues estimated that if we could slow down aging in the general population and increase life expectancy by one year, it would save the US economy $38 trillion.

The event officially kicks off in January, but around 120 people have already signed up, says Glorioso. Ultimately, she expects 500 participants. Each person’s six-month run starts as soon as they take their baseline measurements, whenever that may be. A handful of people have already started. “I think the majority of people are going to want to sign up after the holidays,” says Glorioso. “They’re not going to want to be trying to win an aging contest over Christmas and Hanukkah.”

Glorioso took her own baseline measurements back in August. Those initial tests suggested that her biological age was around 10 years below her chronological age. That sounded pretty impressive to me until I looked at the leaderboard and saw that competitor @Fred had a biological age 28.8 years below his chronological age.

The leaderboard is pretty sparse at the moment, as only seven people have logged their baseline measurements so far. Six of them have biological ages tracking low, but one 47-year-old has been given a biological age of 68.1. She’s competing publicly, so I can see the breakdown of her scores. The worst is for the speed at which she can stand from sitting in a chair—by that measure, she’s been given a biological age of 100.

Hillary Lin, a longevity-focused physician in New York, has just started logging her baseline measurements for the competition. She tells me she’s taken biological age tests in the past and that they’ve tended to give scores below her chronological age, which is currently 37. Last year, the TruDiagnostic test put her pace of aging at 0.75—suggesting she’s only doing nine months’ worth of aging in a typical year. “That’s supposed to be quite good,” she says.

If it had been up to her, Lin says, she would have incorporated more blood tests into the competition, to build a fuller picture of a person’s health. And she’s worried that a six-month period might not be long enough to see changes in biological measures of aging. Still, she’s excited, and is already making plans to improve her sleep, diet, and exercise.

But she already follows a high-protein, low-carb diet and works out at the gym every day. She also visits a sauna every few days (sauna use has been linked to improved heart health). Glorioso’s efforts are impressive too. She tells me she’s started drinking a smoothie made from a pond plant that has been linked to health improvements, and that she’s cut added sugar and deep-fried foods from her diet. She’s also bought “a bunch of fancy Korean sunscreens” in an attempt to lower her face age.

I haven’t received my kit or logged my baseline measurements yet, and I’ll wait until then to start setting health goals. I don’t fancy my chances against these two competitors. But at the very least, I hope I can stand up faster than a 100-year-old.

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

A new AI tool can guess what you’re looking at just by analyzing your brain scans—and re-create that image with remarkable precision. It can go the other way, too, and predict a person’s brain activity based on what they’re looking at. 

In the image above, for example, the left-hand image of each pair is what the user actually saw—and its right-hand counterpart is what the model reconstructed from the brain scan.  

Michal Irani, who developed the tool with her colleagues at the Weizmann Institute of Science in Rehovot, Israel, hopes her “mind-reading” tool will ultimately reveal more about how the brain works, and could perhaps be used to help locked-in people communicate, or allow scientists to re-create the content of dreams. 

Judy Illes, a neuroethicist and professor of neurology at the University of British Columbia in Canada, who was not involved in the research, describes the work as “magnificent.” “The idea [of using this approach] to help people with neurologic conditions … therapeutically is tremendously exciting,” she says.

But other scientists warn that a similar approach could be used to reveal people’s inner thoughts and mental imagery, potentially without their consent. “The results seem very impressive,” says Tommy Sprague, a neuroscientist at the University of California, Santa Barbara. “But if there’s a way to surreptitiously extract information about what you’re thinking about, then …150 years of sci-fi can come true anytime, and that’s worrisome in a lot of ways.”

Peeking into the brain

Neuroscientists have been working for years on ways to use functional magnetic resonance imaging (fMRI) to reconstruct what people see and what’s going on in their minds. The first attempts produced images that were blurry and hard to make sense of. Advances in technology—both in the fMRI scans themselves and in the tools used to make sense of the results—have led to improvements over the years.

Irani and her colleagues started by analyzing publicly available brain-scan data. Other researchers had already collected scans from volunteers who were shown hundreds of images while they lay in fMRI scanners.

fMRI uses a giant magnet to track the flow of oxygenated blood through the brain. Brain areas that “light up” on the scans are thought to be those that are particularly active at any given moment. The results are not especially specific—in typical fMRI scanners, a highlighted “voxel” of activity covers around three cubic millimeters. Each cubic millimeter contains around 16,000 neurons.

But Irani and her colleagues used newer datasets collected using scanners with a higher resolution—each voxel covered around one cubic millimeter of neurons, she says. Those datasets showed what the brain activity of volunteers looked like when they viewed various images.

Other teams have done this too, and several other tools have been used to re-create images from brain-scan data. But they’re not good enough, says Irani. Say a person saw a banana. These models can generate an image of a banana, but it would look different, she says. “It wouldn’t have the same structure, the same position.”

A better decoder

The researchers wanted to more closely re-create the images that had been seen. The first step was to train an AI model on already available data from eight people who each had been shown around 9,000 images while in a high-resolution fMRI scanner.

Crucially, their “brain decoder” has two branches—one to predict the structure of an image (where the colors are, for instance) and a second to predict its content (for example, a bunch of bananas on a plate). The predictions allow a diffusion model, a type of AI best known for creating video and images by gradually cleaning up a noisy mess of pixels, to produce a much more accurate representation of what the person saw.

But to improve the models they needed more data—far more than was actually available.  

To get around this problem, Irani and her colleagues trained another model—an encoder that can predict brain activity from an image. The team then used the encoder and decoder together to improve both tools.

It works like this: Start with a new image of, say, a leopard. Then use the encoder to predict what someone’s fMRI brain scan would look like when the person saw that picture. The decoder is then used to reconstruct the image. At first, the reconstruction probably won’t look much like a leopard, says Irani. But repeatedly training the models this way eventually leads to dramatic improvements.

This approach also allows the team to train their models on as many images as they want, even images that have never been shown to a person in an fMRI scanner. Irani says that around 70% of the training data is from images that were not originally paired with fMRI scans.

By combining data from multiple studies, they were also able to identify brain regions that seem to share functions across all individuals. One region seemed to respond to images of food, for example, while another responded to images of sports. Irani, a computer scientist, says she is now working with neuroscientists “to see if we can actually use these tools that we’ve developed to really find out new things about the brain.”

The resulting “universal brain encoder” can work on a scan from a new person with minimal calibration. In other attempts, a tool has typically required about 40 hours of fMRI data on anyone new before it can be used to predict what that person is seeing. Irani’s decoder only needs one hour of data, she says. The finding was presented at the Cognitive Computational Neuroscience conference in New York last month.

That could make it valuable for neuroscientists studying the brain, says Sprague. “None of us can afford 40 hours of imaging for a new subject,” he says. “It’s something like $600 to $1,000 an hour.” Tools like this one could speed up research, he says.

State of the art

The encoder and decoder aren’t perfect. “Of course we have failures,” says Irani. Over a Zoom call, she pointed out an image of a cake that her tool reconstructed as a pile of three sandwiches, and another of a dog in a bathtub that was reconstructed as a similarly colored goat in a bathtub.

But they represent the state of the art. In a comparison test, the tool was found to be much better than previously described ones. “All in all, really we outperformed the others by a significant margin,” Irani says. “Mind reading” is a “cute, jazzy name” for what they’re doing, she adds.

Irani is now planning to move beyond images to video and audio. She wants to be able to reconstruct what people are thinking about or imagining, and the contents of their dreams. “That’s something we don’t have yet,” she says. “But we’re striving to achieve it.”

Such a tool might enable people who are “locked in” and completely paralyzed to communicate using their brain activity alone, she says. It could also help scientists unpick some enduring mysteries surrounding the inner workings of our minds, such as what PTSD flashbacks look like.

Advances like this inevitably raise questions about mental privacy. What if some bad actor could reconstruct someone’s thoughts or memories?

“If you’d asked me that 10 years ago, I’d have laughed a lot,” says Sprague. Getting a willing person to lie still in a scanner and actively engage with a research question is hard enough; imagine making someone do it involuntarily. But Irani and other scientists are working on similar approaches to decode brain activity from EEG—electrical brain activity measures collected via a cap of electrodes or even through headphones. 

And as models improve, it will become even easier to analyze the brain activity collected this way. “We have to be a little more serious about the ethical considerations,” says Sprague. He thinks Irani’s approach would probably “work quite well” in predicting images that a person is thinking about but not looking at.

The move to EEG would be a “game changer,” says Marcello Ienca, a neuroscientist and philosopher at the Technical University of Munich, Germany. Once an EEG device has been calibrated to a user’s own brain, it could be relatively easy for companies to extract additional information from that person’s brain—potentially without consent. Ienca can also imagine some courts allowing mental image reconstructions as legal evidence.

“I have no doubt that this is, you know, well-intentioned research, but I think it’s also pretty obvious that it could be co-opted for … ethically and societally problematic commercial uses,” he says.

Irani acknowledges the potential for misuse with the use of EEG. But she’s not concerned for now. “I’m trying to think only of good things,” she says.

If you want to install a big battery in New York City, you have to cut through a notoriously tough tangle of regulations. That’s made it difficult to get large energy storage projects on the grid.

Faced with those obstacles, some startups are getting creative, finding ways to use relatively small batteries in unexpected places, from induction stovetops to food carts. By deploying these more consumer-focused batteries, the companies are helping people directly, cutting pollution or power bills. And these units, while small on their own, could add up to a lot of capacity.

Large, utility-scale battery installations are coming online quickly around the world. But there’s been some pushback from communities—especially population-dense ones like NYC—because of concerns about relatively rare but high-profile battery fires.

“The number-one challenge that large-scale centralized energy storage providers have is the fact that their benefits are abstract and their costs are concrete,” said David Hammer, cofounder of PopWheels, at a New York Climate Week event on September 24.

PopWheels built a battery-swapping system for delivery drivers who zip around the city on e-bikes. Drivers swap out depleted batteries for fresh ones at a cabinet, avoiding the need to stop and charge. There are about 50 cabinets across New York City, and about 2,500 batteries in circulation, Hammer says.

The company recently started providing its batteries to food cart operators, allowing them to replace the gas generators they currently rely on. About four of its batteries can supply five kilowatt-hours of electricity, enough to roughly cover a day’s operation for many carts.

Others are looking inside homes. Copper, for example, is a company building induction stoves with integrated batteries. The appliances allow residents to cook if the power goes out, and they could also help homeowners avoid expensive electrical upgrades, because the battery provides some of the power when the stove is in use. “We’re not just selling a battery for resilience—we’re selling it for cost savings,” Sam Calisch, cofounder and CEO of Copper, said at the event.  

Other companies, like Every Electric and David Energy are selling plug-in batteries for homes and businesses. Every’s battery is designed to be plugged in to an air-conditioning unit. The company provides the batteries and pays customers to reduce use during high-stress times. David Energy has a similar business model, but focuses on businesses like laundromats, parking garages, and gyms.

Because all these companies’ batteries are small or even integrated into another device, they don’t generally require grid upgrades or extensive permitting processes. Just plug them in and enjoy the benefits—similar to the growing model of balcony solar, with small modules that plug in and help offset a home’s or business’s energy demand.

The move toward more distributed batteries could be a huge change for the grid. “This is the biggest thing to happen in the power grid sector in the last 20 years,” says James McGinniss, cofounder and CEO of David Energy.

If every stove in America were shipped with a battery, it would add up to tens of gigawatts of power, Calisch said. That’s capacity that can be aggregated and tapped when the grid is stressed.

As batteries continue to get cheaper, they’ll likely find their way into more corners of our lives, bringing along many benefits. Cleaner air, cheaper power bills—there’s a lot to like. That being said, the need for large-scale energy storage isn’t going to go away—larger installations will do the crucial work of supporting intermittent renewables like wind and solar, so they can help meet more of our quickly growing electricity demand.

 But I’m interested to see how far an aggregated horde of stoves, e-bikes, and window AC units could go in helping the grid. The grid is getting more congested, especially in dense cities like New York. Anything that could help avoid summer blackouts is welcome.

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 US government wants to spend $30.3 million over the next five years on an improved form of lie detector, according to a Department of Defense budget request. The program, called Polygraph+ or Polygraph Next, will focus on scoring algorithms that use artificial intelligence and machine learning and on a technique called “standoff sensing,” which refers to the ability to take physiological readings without attaching a device to a subject’s person.

According to details of the budget document, which were first reported by Inside Defense, the project will “modernize federal polygraph and credibility assessment technologies” to improve their accuracy and reliability. But the project may be just the latest in a long line of failed attempts to use technology to detect lies. “It’s a misguided effort to reduce the complex to something that is tangible,” says Kyri Kotsoglou, a legal scholar at Northumbria University in the UK who studies the use of polygraphs in the justice system.

The move comes at a time of high tension within the department. Under Defense Secretary Pete Hegseth, the Pentagon has been increasingly turning to polygraph tests in an attempt to find the sources of alleged leaks to the press. In September, the New York Times reported that around 50 officers on the Joint Staff had been given polygraph tests after news coverage reported on the depletion of US weapons stockpiles in the war with Iran. 

Polygraph+ will be run by the Defense Counterintelligence and Security Agency (DCSA), which conducts background checks for the federal government. According to the budget document, which has not yet been approved by Congress, the new technology will be used for vetting of prospective employees and “insider threat detection.” It is not yet clear which specific technologies will be used, and the DCSA did not respond to a request for more information. 

But other Pentagon efforts offer potential clues. In 2023, the department’s Defense Innovation Unit (DIU) ran an open submission process to find companies with products that could be used for deception detection.

It selected two companies to build prototypes: Presage Technologies, which claims to be able to measure heart rate and breathing rate using standard cameras, and Altec Research, a medical sensor company now branching out into non-contact sensing technologies. A screenshot of Altec’s prototype technology released by the DIU shows that it tracks head movement, facial skin temperature, and pore activity. Presage Technologies and Altec Research did not respond to requests for comment. The DIU declined to comment. 

Current lie detection technology has barely changed since the polygraph was invented in the 1920s. Examiners rely on blood pressure, pulse, breathing, and sweat measurements to determine if someone is lying. They make judgments about the veracity of respondents’ replies on the basis of differences in their physiological response to baseline questions like “Is the sky blue?” and target questions like “Have you ever committed a crime?”

The federal government conducts tens of thousands of the tests each year while screening employees, but the reliability of this technology has been repeatedly challenged—and its results are rarely admissible in court. In 1983, Congress’s Office of Technology Assessment concluded that there was very limited evidence supporting the polygraph’s use for screening employees, and in 2003, the US National Research Council (NRC) said evidence for its efficacy was “weak at best.” 

Research suggests humans can spot a lie just over half the time without any technical assistance. The American Polygraph Association claims the polygraph is between 80% and 94% accurate. But the 2003 NRC report pointed out that a screening test with this level of accuracy could still lead to a lot of mistakes. The DOD employs 2.8 million people; an imperfect system applied at that scale could end up falsely accusing tens of thousands.

There are other issues too. Polygraph interpretations are often subjective: Different examiners get wildly different results, and people from minority groups are more likely to be judged as deceptive. What’s more, with training it’s possible for interviewees to learn a variety of countermeasures that can help beat the test; for example, they may artificially heighten their physiological response to baseline questions by stepping on a pin hidden in their shoe. 

“If you know how it works, you can beat it,” says Sophie van der Zee, an associate professor who studies deception at Erasmus University in Rotterdam. She says the machine’s biggest effect is deterrence—often, subjects confess before it even begins. “But that only works if people think a polygraph works,” she points out.

Various new strategies for lie detection have been attempted over the decades, with technologies ranging from thermal cameras to pupil trackers to brain scans. None of them have yielded reliable results outside the lab. The problem is that there’s no single telltale sign of lying that’s true for everyone all the time. “There is still no Pinocchio’s nose,” says van der Zee. 

AI could theoretically improve polygraphs if it could find patterns in the data that examiners can’t. AI algorithms are also more likely to be used for “multi-modal” deception detection, which seeks to combine multiple measurements into an overall deception “score” that is harder for people to game. Three things are happening under the surface that lie detection tries to home in on, van der Zee says: physiological stress, cognitive load, and the conscious efforts people make to conceal the fact that they’re lying. Current polygraph technology tackles only one.

“The more you can have combined methods that approach it from these three different angles, the more successful you will be,” van der Zee says. This isn’t a new concept—in the 2000s, researchers at Manchester Metropolitan University in the UK developed a system called Silent Talker that generated a deception score from video footage. This was later folded into iBorderCtrl, an EU-funded pilot program. In the US, a project called AVATAR incorporated eye tracking, voice analysis, and body movement detection into a tool for use at border crossings. All these projects have quietly faded away. 

Kotsoglou says combining AI and the polygraph is “the worst of both worlds” because it adds uncertainty on top of invalidity. Even if AI or machine learning can spot previously unseen patterns in physiological data, it won’t be able to reliably link them to lying because there is no real ground truth. 

“Even if you have all the records in the world from polygraph tests, you don’t know whether those polygraph tests are right or not,” says Marion Oswald, a professor of law who has written with Kotsoglou on the use of polygraphs in the justice system. She fears that new forms of lie detection will, like the polygraph, be used more as a psychological prop than a scientific tool. 

“It seems very much a response to the concern of the current administration to leaks and perceived lack of loyalty,” says Oswald. “[Lie detection is] being used as a threat, to intimidate and force people to confess to things, as opposed to anything that’s actually getting valid information.”

Around this time last year I was attending an aging conference in Manchester, listening to a talk about fly aging, when my phone started pinging. News outlets were reporting that a hot mic had caught Russia’s and China’s leaders discussing the possibility of living forever.

“With the developments of biotechnology, human organs can be continuously transplanted, and people can live younger and younger, and even achieve immortality,” Russia’s Vladimir Putin reportedly told China’s Xi Jinping.

He seems to have been referring to the “replacement” theory of longevity, which has been supported by multiple experiments that involved physically stitching young mice to old ones. Something about the young blood rejuvenated the old mice. Perhaps young organs could rejuvenate world leaders in their 70s, too.

Unfortunately for Putin, new research pours a little cold water on this idea. Studies on transplanted hearts in both mice and humans suggest that new hearts soon adopt the biological age of the recipient, no matter how young they were to begin with. The finding could be important for transplantation, but it also highlights just how complex aging—and rejuvenation—are.

Jesse Poganik at Brigham and Women’s Hospital in Boston is one of the many scientists trying to understand exactly what it is about the bodies of young mice that rejuvenates old ones. Plenty of research has focused on seeking the secrets of youth in young blood. But what if it’s something about young organs instead?

To find out, he and his colleagues performed a set of heart transplants in mice. In humans, heart transplants typically involve removing a damaged or injured heart and replacing it with another from a donor who is usually much younger than the recipient. (When Poganik assessed hospital records, he found that most recipients were about 20 years older than their donors.)

The mouse transplants were different: Mice received a second heart, implanted in the neck—a procedure that’s slightly simpler and allows scientists to compare the new hearts with the existing ones. In some cases, young adult mice were given a heart from a middle-aged donor. In others, middle-aged mice got young hearts.

The team used a trio of “aging clocks”—molecular tools used to estimate the biological ages of tissues and whole organisms—to assess whether the additional hearts affected the mice in any way. These clocks were good at predicting the chronological age of mice that didn’t get new hearts.

Poganik says he was expecting to see a reciprocal effect, and that young hearts might benefit older animals, for example. In previous work by other members of his team, young mice that got old hearts experienced a buildup of senescent cells in their other organs. These cells are thought to contribute to the aging process, suggesting that receiving an old organ might prematurely age an animal.

But that’s not what he found. When he and his colleagues analyzed the transplanted hearts between four and six months after surgery, they found that the hearts seemed to have adopted the biological age of the recipients. Young hearts got older, and old hearts got younger. “The environment of the transplanted organ really dictates how it seems to behave biologically,” he says. The findings were published online at bioRxiv last week.

The team also looked at each mouse’s blood and other organs—including its original heart—and were surprised to find that they seemed to be unaffected by the presence of the new heart, despite the age of its donor.

The finding was backed up by data from human heart transplants. People who receive a donor heart must typically undergo a series of heart biopsies after surgery. The tiny pieces of heart tissue collected from people who had heart transplants at Brigham and Women’s have been stored for decades. And when Poganik and his colleagues tested their biological ages with the aging clocks, they found a similar pattern: No matter the age of the donor, a new heart quickly adopts the biological age of the person who received it.

It’s not clear why this is, but João Pedro de Magalhães, who studies aging at the University of Birmingham in the UK and was not involved in the study, thinks it might have something to do with the recipient’s immune system. Perhaps immune cells circulating in the blood might affect markers of aging in the new organ, he says.

Perhaps the potential rejuvenating effects of a young heart end up being diluted by all the other aged components of an older body, Poganik suggests. Or maybe a single organ just isn’t enough to see an effect.

Poganik hopes his finding will encourage surgeons to consider using hearts from older donors, many of which are discarded on the assumption they won’t function as well. (Machines used to preserve organs before transplantation are changing that already—and Poganik has worked on another study showing that these devices seem to rejuvenate donor livers to some extent.)

But the study also highlights just how complicated aging is. If organs seem to be getting older by one measure but not by another, how can we get a full picture of the biological age of an organ, or a person?

This complexity means that scientists are not likely to discover a true way to completely reverse aging, says Poganik. “That would mean that every aspect of aging has to go back in time,” he says. Reversing DNA damage, structural damage, and all the other degradations that are part of the aging package, across all our various cells and tissues, presents an enormous challenge.

“There are aspects of biological age that are probably reversible, and there are aspects that are probably not,” he says. Sorry, Putin.

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 week, world leaders descended on Manhattan for the UN General Assembly. It’s also New York Climate Week—investors, policymakers, advocates, and journalists are colliding at panels, talks, and fancy dinners.

With so many climate voices in one place, the discourse can feel a little louder than usual. This year, the unavoidable topic is artificial intelligence.

There’s been a growing tension bubbling up about AI’s impacts on climate and climate tech. Depending on where you stand, you might point to AI’s potential for advancing research, or to the way funding and attention from Big Tech is trickling into energy startups. Or you might focus on the emissions-heavy natural-gas buildout that’s unfolding to meet the sector’s electricity demand.

Everyone seems to agree that AI is important. The big question for those in the climate world, and the conversation I’m constantly having and hearing this week, involves how you see its influence unfolding.

UN Secretary-General António Guterres highlighted the AI and climate crossover in a speech on the first day of the assembly. “The climate crisis fuels instability and displacement,” Guterres said. “Artificial intelligence could help solve all these challenges, or it could make them worse.”

It feels relevant that this year is the first that the world has really had to grapple with the fact that climate goals are slipping out of reach. A recent report from the UN Environment Program said that the world has nearly passed the point where we could possibly keep warming to less than 1.5 °C above preindustrial levels.

Since we’ve essentially missed this target, the report lays out the need not only to quickly and drastically reduce greenhouse-gas emissions, but also to employ carbon removal to help suck up emissions that have already been released into the atmosphere.

The billion-dollar question is whether AI could help with any of this. 

The energy-intensive technology is certainly shining a spotlight on the need to build out electricity supplies and shore up grid reliability. The result is more attention and money for energy technologies, some of which happen to be low- or zero-emissions.

As I’ve covered before, startups across the energy and climate sectors are benefiting. Firms in nuclear, geothermal, wind, and solar power have signed deals with the likes of Google, Meta, and others looking to power their new or growing data centers.

Global climate-tech investment from venture capital hit $26 billion in the first half of 2026, according to data from Currence, a finance tracker for the industry. That’s 55% higher than last year, and products and services for data centers are getting a massive slice of that pie.

But as a Semafor piece about the report points out, some sectors are slipping through the cracks: Carbon management and low-carbon fuels saw VC investment plummet this year. These are important solutions for addressing climate change but may not be able to sell themselves to a data center.

So far, the data center buildout has come with a hefty emissions toll. A few years ago, Microsoft, Google, and Meta all had ambitious goals to reduce greenhouse-gas emissions. Now they’ve all seen emissions rise, largely because of data centers that are needed to power AI.

Some people are optimistic. AI could help speed up progress in areas like the search for new catalysts, as Evelyn Wang, MIT’s VP of energy and climate, pointed out during a panel. And data centers won’t add to climate and water problems forever, Wang told the Associated Press. (She puts the timeline at about a decade until data centers no longer add to planet-warming emissions.)

But a whole lot of natural gas is coming online to meet the immediate demand created by new data centers. And once those power plants are built, they have a decades-long lifetime. 

That’s partly why public pushback to AI is growing—people are seeing more pollution and noise near these data centers and the power plants that provide them with electricity. Overall, what I’m hearing this week is that many in the climate sector are skeptical of AI, at best. 

“AI leaders are now on thin ice when it comes to license to operate and sinking deep underwater when it comes to public support,” said UN climate chief Simon Stiell in a speech this week. “Tech titans need to start showing why the benefits of AI outweigh its skyrocketing costs—for the many, not just the tiny few.” 

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

Shubnam was packing boxes for a move into a new home when their friend sent them an Instagram video. The footage had only been up for a few hours, but it was days old, recorded at a Delhi protest this spring against a bill that would have narrowed the legal recognition for transgender people in India. 

There, a content creator known for rage-bait videos had approached Shubnam, a 38-year-old transfeminine graphic designer and educator. Shubnam, who uses a single chosen name, was dressed in a green sari with a shaved head, and the creator asked why they were wearing “female clothing.” Shubnam told him to back off. Only when the video arrived did they realize he had recorded the exchange on Meta smart glasses, styled to look like ordinary Wayfarers. 

Edited into a mocking reel, the footage drew millions of views, along with transphobic abuse, reaction videos, and AI-generated memes. It spread from Instagram to X and YouTube. Copies remain online today. 

For Shubnam, the video has caused huge harm. “All it took was one unconsented video to rewind two decades of progress that I made all on my own,” they told MIT Technology Review. “For a moment, it felt like this entire adult life I had lived was just a long lucid dream, and I was going to wake up as that clueless queer child again.” They haven’t worn a sari outside their neighborhood since.

Experts warn that many others will experience similar ordeals as smart glasses go mainstream. And they say privacy problems are becoming especially acute in India, where covert recording and the circulation of images without consent are already pervasive.

Cases of cyberstalking, sextortion, deepfakes, and other forms of online abuse are growing in the country. And the chances of a crackdown on technologies such as smart glasses are slim, given that Indian law enforcement authorities are themselves increasingly adopting AI-powered surveillance tools, including facial recognition systems and their own AI-enabled smart glasses.

An unequal burden

Smart glasses are among the fastest-growing devices in India’s wearables market, according to the International Data Corporation, a research firm. And Reliance Jio, one of the country’s largest telecommunications companies, says that later this year it’s set to launch a sub-$105 rival to Meta’s glasses, the latest model of which have a $420 price tag in India. 

The core privacy issue is that these cameras don’t look like cameras, so targets may not notice they’re being recorded. Yet Meta rejects the idea that it’s responsible for protecting people from intrusions, says Janusz Świerczyński, an Oxford Saïd Business School researcher who studies privacy and emerging technology.

“More and more, the expectation is being placed on bystanders to look out for themselves,” he says. That means it’s up to them to spot a small LED in the glasses frames that activates when recording is in progress. Meta also relies on wearers to behave responsibly, says Świerczyński. Reliance Jio did not respond to our request for comment on privacy issues related to its upcoming glasses. 

Creators have already demonstrated ways to circumvent the recording light on Meta’s glasses. Tutorials show users how to cover or otherwise obscure the LED while continuing to record. Back in March, Wired documented a market for “stealth mode” modifications that physically disable the light or remove it entirely. 

All that makes covert recording easier, and the risks are not borne equally. Świerczyński says women and other marginalized groups face greater harm when inconspicuous devices capture and circulate footage without consent—as happened to Shubnam.

The bigger problem, however, is that in India smart glasses aren’t just being used to turn ordinary people into targets of viral “pranks.” They’re also being adopted as a tool of police surveillance. 

In June, when thousands of young demonstrators gathered to protest problems with the country’s education system, Delhi police used Meta smart glasses to record them. In fact, a petition filed by former Jawaharlal Nehru University Students’ Union president Aishe Ghosh before the Delhi High Court alleged that police had filmed protesters continuously for weeks, including while they were eating and resting. It also alleged that the police had threatened to send footage of student demonstrators to their parents and colleges.

In July, India’s solicitor general dismissed the case in court as “luxury litigation.” Instead of investigating the students’ complaints, police turned on the students themselves, opening 10 criminal investigations for offenses including rioting, assault on public servants, and damage to public property. Some of the protesters, who spoke to MIT Technology Review on condition of anonymity, say they never saw the LED on the Meta AI glasses police wore during the demonstrations. 

Filming without consent

Meta maintains that its safeguard works. The company told MIT Technology Review, “Every pair has a capture LED that blinks when you take a photo or video that you can save or share; it can’t be turned off, and if someone covers or damages the LED, the camera is disabled.” The company also says users are responsible for complying with the law and respecting people’s privacy.

That approach is risky, especially in India, where norms around filming and consent are still highly disputed and digital literacy is still uneven, says Sameer Parmar, a counselor at Meri Trustline, a safety partner working with Meta and YouTube to support people harmed by online content. 

“The hardware is improving, but the country’s weak understanding of consent gives an invisible camera far more room to cause harm,” he says. Simply being in a public place is often treated as implicit consent to be photographed or filmed. He says the help line has been seeing an uptick in cases involving covert recording and nonconsensual footage.

“Violations of this access have increased significantly and are expected to rise further in India,” he says. People are recorded in medical settings, during private encounters, and in public spaces. Those with less power to object are particularly vulnerable, he adds.

“It’s either trolling or it’s playing pranks on them and then recording them” he says. Abusers “do target people who are a little more unaware or susceptible to how this technology-facilitated harm works.”

The law in India has not caught up with this new kind of harm, says Apar Gupta, a lawyer and founder-director of the Internet Freedom Foundation. “India’s laws against voyeurism and the sharing of intimate images were designed largely around more recognizable forms of abuse,” he says, citing offenses such as secretly filming someone undressing, using a toilet, or engaged in a sexual act. “[The laws] offer limited protection when someone is secretly filmed in an ordinary public setting and the footage is then used to humiliate them,” he says.

Social media platforms’ rules do offer another route to removing harmful content filmed with smart glasses, but it’s not an immediate remedy. “Under India’s current rules, platforms can have up to 36 hours to act on certain complaints—a window that can be long enough for a video to be downloaded, copied, and circulated elsewhere,” he says.

The problem of filming without consent seems set to grow. According to the Financial Times, some internal prototypes of new Meta glasses include AI features that continuously photograph and listen to the wearer’s surroundings, enabling the glasses to later recall what they saw or heard. These functions wouldn’t activate the external recording indicator, making it impossible for bystanders to know when information about them is being collected.

Meta told us it intends to keep the LED for active capture, such as taking photos or recording video, but not for AI features that interpret the wearer’s surroundings. Its argument is that a constantly blinking light could eventually become so commonplace that people stop noticing it.

But that logic risks putting society in a perpetual game of catch-up, says Woodrow Hartzog, a professor of law at Boston University. “Tech companies have an incredible ability to push these tools out faster than society can meaningfully acclimate to the threat,” he says. “By the time society catches up, there’s already a degree of normalization.”

For Shubnam, the danger is not just that people may become accustomed to being recorded. It is that they will become accustomed to seeing versions of themselves that they never chose to put out into the world. 

“Being comprehended along the lines of a narrative that I didn’t compose or consent to—that imposed recognition makes you feel like you’re powerless in how people know you,” they say. 

On Wednesday, MIT Technology Review hosted a live Roundtables event for subscribers that asked the question everyone’s asking right now: Could AI really kill us all?

But attendees had so many more questions than we had time to answer in the 30 minute session. So we asked our senior AI editor Will Douglas Heaven and AI reporter Grace Huckins to round up some of the best questions attendees submitted and try their best to answer them.

Thanks to all who submitted questions!

Am I gonna die?

Yes, eventually. Unfortunately, my journalistic powers of prognostication aren’t powerful enough for me to tell you how. But it certainly could be because of AI. AI-powered drones have already killed people in Ukraine, and AI-driven cyberattacks on hospitals will surely claim victims before long. 

Could AI go even further, and kill all of us? Less likely. But some people—quirky people, but undeniably knowledgeable about AI—have been warning for years that this could happen. And while I’m not yet stockpiling canned food or trying to get in good with a bunker-owning megabillionaire, I have noticed that the doomers’ predictions about AI capabilities and alignment have, over the past couple of years, proved disconcertingly accurate. That certainly doesn’t mean that their more dire forecasts will come true, but it’s enough for me to sit up and take notice.

— Grace Huckins

Are you going to die because of AI? I’d say there’s a non-zero chance. Let’s say you’re unlucky enough to be the victim of a freakish near-future event or accident. Maybe it’s a cyberattack carried out by a swarm of AI agents on critical infrastructure. Sadly, a scenario like that now no longer feels as far-fetched as it once did. Or maybe a novel AI-designed pathogen cuts through the population. Or the world economy crashes, causing conflicts and famine. Both plausible, but I think less likely. 

Are we all going to die because of AI? Nope. There are no circumstances outside of apocalyptic science fiction in which AI could kill us all. You can spin up any number of scare stories, but they’re not grounded in present-day realities about what the tech can do or where it’s headed. 

Some people argue that there’s no harm in preparing for the worst, however wacky it might seem. Maybe. But I think such catastrophizing can make people excuse or overlook many of the more immediate problems with the existing technology and the companies building it. 

— Will Douglas Heaven

Why would AI kill us?

Someone might tell it to, and it might listen. That’s part of the reason researchers are so concerned about AI’s biological capabilities—imagine what Aum Shinrikyo, the doomsday cult behind the Tokyo subway sarin attack of 1995, would have done with a tool that could design a pathogen deadlier than Ebola and more transmissible than measles. Those of us who don’t want to die have to figure out how to defend against all plausible biological weapons, but our would-be attackers only have to manufacture one effective pathogen.

Then there’s the more exotic-sounding possibility that an AI could decide to kill us itself. There are various stories about how this might happen out there, but the most widespread involve AI systems that don’t hate people, necessarily—we are just an obstacle between them and the goals that we gave them.

Much as the OpenAI agents behind the Hugging Face hack compromised another site’s infrastructure to get a good score on a test, the idea is that some future, more powerful AI might get rid of us to prevent us from shutting it down—all in pursuit of some goal that we instructed it to go after. 

— Grace Huckins

How can we best ensure alignment so the worst doesn’t happen, and who is doing the best work to achieve it? 

Alignment is a huge area of research. In simple terms, it involves building models that behave in ways we want them to and not in ways we don’t. We need to trust agents better before handing over more autonomy. Alignment is supposed to establish that trust. But it’s hard. 

LLMs aren’t designed in the way other software is, where dos and don’ts can be hard-coded in. Instead, aligned behavior needs to be instilled when models are trained. One approach is to reward them for doing things you want them to (a little like raising a toddler, perhaps). Another approach involves giving an LLM a written list of rules it is supposed to follow (kind of like a constitution). 

Anthropic and OpenAI are both leaders in this field—and yet neither has been able to develop models that are fully aligned. A big problem is that LLMs are far more inconsistent and far less predictable than people. They can behave in one way in one situation and another way in a situation that to us seems very similar. They can also be swayed by unexpected constraints. For example, faced with an impossible task (as many of the agents involved in the Hugging Face hack were), models may try to do whatever it takes to achieve their goal. As Grace mentions above, that could be an issue.

The main reason top AI firms now say they want a slowdown is that they want to focus on cracking alignment. Alignment isn’t necessarily a pipe dream. But the jury’s out on whether full alignment will ever be feasible. 

— Will Douglas Heaven

Is AI really dangerous, or is this the tech companies drumming up PR?

This is always a reasonable thought when it comes to tech companies heading for an IPO—CEOs have an obvious incentive to make their products seem radical and transformative. But I’m not so sure it makes sense here. Telling the public that an already unpopular product could kill them and everyone they love is horrible corporate image management.

There are other stories you can tell about the CEOs’ motivations—maybe they want to cool down the public furor over data centers by portraying themselves as responsible stewards of a world-changing technology, or maybe they want to buy time to get their ducks in a row and prevent the next PR catastrophe. 

But there’s also a simpler explanation. Thinking that AI could bring about human extinction has been pretty common in San Francisco for a while, and these men are steeped in that milieu—as are their employees, many of whom signed an open letter in July urging their companies to work to make an AI slowdown possible.

— Grace Huckins

Part of the concern occurs when AI agents are allowed to act autonomously and with no supervision. What’s the issue preventing more control over these agents? 

This question goes to the heart of what we want this technology to be able to do. The trade-off between autonomy and control is tricky to get right because, on the one hand, a lot of the power of AI agents is that they can carry out tasks and solve problems without a human having to micromanage them. On the other hand, that requires you to trust that the unsupervised agents won’t run amok. 

What we’re seeing is that AI labs haven’t yet got this trade-off quite right. Their models are not trustworthy, they are not properly monitored, and they are not always under control. Figuring out how to fix that while still allowing for useful autonomous activity is one of the big research challenges of the moment.  

— Will Douglas Heaven

What steps can be taken now and in the near future to ensure that AI is controlled, monitored, and regulated effectively? 

That’s the million-dollar question. Whether or not you think AI could kill us, you can’t deny that it could do some real damage, because it already has—by driving people toward psychosis and by hacking websites, for example. Preventing that damage, or at least mitigating it, is hard for two reasons. 

The first is that we barely understand how AI works, and it’s quickly growing more powerful. There is lots of ongoing research about how to monitor and control misbehaving agents, but the current approaches are fragile. You can see if an agent discusses misbehaving in its “chain of thought,” the workspace where it plans its actions—but OpenAI’s newest agents don’t show their work in the same way as previous ones. And you can try to monitor agents with other agents, but that requires you to trust the monitor.

The other obstacle is more familiar. There’s a huge conflict of interest when AI companies regulate themselves, but the US government has thus far failed to step in, despite some bipartisan support in Congress for efforts to do so. The executive branch, for its part, seems stringently opposed for the time being. But if the winds do shift, I for one would appreciate some strong transparency regulations, so that we can get a fuller story the next time an unreleased frontier model mounts a cyberattack.

— Grace Huckins

If this dialogue makes it into web discourse, will it become a self-fulfilling prediction?  

That’s a real concern. LLMs are influenced by what they read. One theory for why chatbots so often talk about (and role-play) apocalyptic scenarios is that they have been trained on millions of pages of science fiction stories and doomer internet forums. All the text being produced right now, including this article, could in turn influence the behavior of future models. Extremely meta.

In fact, the team at METR, a third-party organization that OpenAI called in to help understand what happened in the lead-up to the Hugging Face hack, raised a related possibility in its report on the incident. METR used OpenAI’s new model Astra to help analyze the vast numbers of agent transcripts and behavior logs.

But feeding all that material to the model could have unintended consequences. There’s a good chance that the agents doing the analyzing were biased by the text produced by the agents they were analyzing. There’s no such thing as a clean slate anymore. 

— Will Douglas Heaven

With thanks to Eric, Pranab, Rafael, Kenneth, George, Chris, Yoon Jae, James, Carl, Nicole (and more!) for the fantastic questions.

In recent weeks, leaders of some of the biggest AI companies have warned that the very tech they are developing is dangerous. Last weekend, Anthropic CEO Dario Amodei argued that AI carries serious risk and that progress should be slowed. OpenAI CEO Sam Altman responded on X: “I agree with Dario that we need to pace the frontier.”

Those posts came a few days after the AI researcher Jacob Coxon announced that he was leaving a role at Anthropic, charging that neither it nor OpenAI (where he had also worked) was acting responsibly. “The people building AI earnestly believe that it could kill us all by the end of the decade,” he posted on X. Another Anthropic employee, Evan Hubinger, publicly agreed with him. “We really do earnestly believe AI could kill all humans!” he responded on X. “I personally think it is >10% within the next decade.”

One of the ways they fear AI might end us all is by somehow aiding the design, creation, and release of some kind of bioweapon. Let’s take a closer look at why.

A bioweapon might be a highly lethal virus that targets people according to their genes. It could be a fungus that wipes out a crop and causes food insecurity. Perhaps it would be a tasteless, odorless toxin that could be slipped into a region’s water supply, undetected.

The concern is that AI tools can be used to help generate agents like these. In 2022, researchers at Collaborations Pharmaceuticals found that it was remarkably easy to do so using an AI “molecule generator” they’d developed to find potential drugs for human disease. In less than six hours, the model generated 40,000 molecules with the potential to serve as chemical warfare agents. Some of them were designed to be even more toxic than known nerve agents. “Without being overly alarmist, this should serve as a wake-up call for our colleagues in the ‘AI in drug discovery’ community,” the authors wrote at the time.

It was a wake-up call for David Magnus, a professor of medicine and biomedical ethics at Stanford University, even though he had been assessing the risks associated with the misuse of medical science and biotechnology since the late 1990s. “That was very scary to me,” he says. “Of course, everything since then has just sort of blown up.”

Today, AI bots can answer questions on topics spanning all realms of science. Anyone can use large language models trained on the knowledge and experience of “almost every scientist who ever lived on this planet,” says Dunja Sabra, a biosecurity researcher at the University of Hamburg in Germany. Those models can provide instructions and video training on how to conduct experiments.

Combine that with advances in biotech that have made gene editing and synthetic biology tools much more accessible (the “DIY biology” movement has already enabled many people to set up labs at home), and you’ve got a potentially very dangerous situation. “The chances are that someone determined would succeed eventually,” Sabra says.

There are safeguards in place. People who want to build new genomes must typically order the pieces of DNA from companies that screen for suspicious requests. Responsible researchers put potentially risky research through rounds of analysis called “red-teaming,” in which independent scientists look for ways the work might be misused, and “blue-teaming,” where others come up with potential mitigations. And AI companies have tweaked their tools in attempts to prevent them from offering up scientific information that could be misused. But none of these protections are ironclad.

In a report published last week, Anthropic acknowledged that people had attempted to use its models to explore ways to make the chikungunya virus more transmissible, create a form of bird flu that is more dangerous to humans, and build an “atlas of venom toxin peptides,” among other things.

“We’ve got a constant back and forth,” says Magnus. “We have to build better surveillance and screening tools, [but] AI is really good at figuring out ways around them.” We’ll probably need to use AI to find ways to restrict the use of AI, he says.

I should add here that not all scientists agree on the level of risk. At a recent media briefing, some biologists at Imperial College London argued that AI tools just aren’t good enough to fully develop bioweapons, and that testing new pathogens requires difficult, time-consuming, human work. Some think the guardrails we have in place are sufficient.

And Wendy Barclay, a professor of infectious disease at Imperial, pointed out that, as things stand, the greatest risk of a pandemic isn’t from a bioweapon, but from pathogens that are already circulating. Take H5N1, the bird flu virus that has already killed millions of birds and spread widely through US dairy cattle; last month it was also detected in captive mink at a farm in Utah.

Sabra, on the other hand, likes to think five to 10 years ahead. Countries should be strengthening their health-care systems, preparing antidotes to known toxins, and stockpiling medicines, she says: “We need to be prepared.”

Kevin Esvelt, an MIT biologist who invented both technology to fast-track the propagation of a genetic feature through an entire population and ways to limit that technology, echoed these concerns in an X post on Wednesday, stating that a large language model had “disclosed a novel form of bioweapon that I hadn’t realized was possible.”

He added, “Please, for the love of God, children, the future of humanity, or whatever you consider holy, let’s err on the side of caution here.”

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

Productivity is everything

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

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

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

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

Others get a similar number.

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

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

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

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

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

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

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

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

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

We’re all part of the AI gamble now

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

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

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

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

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

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

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

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

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

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

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

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

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

After the bubble

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

Clocking organs

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

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

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

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

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

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

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

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

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

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

Molecular repair

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

1. Preventing maternal deaths

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

2. Making brain electrodes inspired by Japanese art

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

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

3. Developing an all-new treatment for baby KJ

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

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

4. Reversing the aging process to treat eye disease

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

5. Using AI to design new viruses

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

If you’ve had to fill up your vehicle’s gas tank or buy a plane ticket lately, you’ve probably felt the effects of rising fossil-fuel prices. But farmers buying fertilizer for their crops are especially aware of just how far the ripple effects of the conflict in Iran have spread.

Fertilizer prices have been on a roller coaster this year, kicked off in part by trade disruptions and high prices for natural gas, a key ingredient in fertilizer production. Let’s take a closer look at why conventional fertilizer prices are so sky-high, and how a few more climate-friendly alternatives could bring farmers some relief.

As fossil-fuel prices go up, nearly all industries are affected, since most of our economy relies on these fuels to move goods and people around.

But fertilizer is even more intertwined with these fluctuations, because natural gas is used as both an energy source and a chemical input in the production of ammonia, a key fertilizer ingredient. So as natural-gas prices have spiked in recent months because of the war in Iran, fertilizer prices have followed. (It’s worth briefly noting here that fertilizer production is also a major source of greenhouse-gas emissions, accounting for about 2% of the global total.)

Fertilizer trade is being directly affected as well, since about one-third of global seaborne trade in fertilizers passes through the Strait of Hormuz, which has been effectively closed to commercial traffic because of the conflict. Access to fertilizer could get worse for some of the poorest countries around the world because of the strait’s closure, according to a report from the World Bank. While the US largely meets demand for nitrogen fertilizers with domestic production, some imports do come from the Persian Gulf.

At one point in April, the price of urea (the most commonly applied fertilizer) climbed above $850 per metric ton. That’s 80% higher than it was before the conflict and the highest level since 2022, when the Russian invasion of Ukraine and the resulting conflict caused fertilizer costs to hit record highs. Prices have come down significantly, but forecasts remain uncertain.

“There’s just this out-of-control supply chain that’s a lot more volatile than it’s ever been,” says Travis Frey, chief technology officer of Pivot Bio, a company making fertilizer with genetically edited microbes. (For more on these microbes, how they work, and what research is still needed, check out my latest story here.) 

Pivot says its products are cost-competitive with chemical fertilizers today. And because they don’t use natural gas as an input, they aren’t subject to the same price spikes. When the war in Iran started, Pivot increased the volume it planned to produce, dropped prices, and allowed farmers to lock in prices for three years, Frey says.

That could be a major help for those farmers, because high prices could be here to stay for a while. Some fertilizer prices could remain high through at least 2028, according to a report from CoBank, one of the largest banks for the agriculture industry in the US.

That’s partly because the war has caused long-lasting damage: 31 ammonia plants in the Middle East have been affected or shut down completely. That’s on top of 20 ammonia plants that have been damaged in Russia in recent years.

Ongoing high prices can be extremely challenging for farmers. “The fertilizer price spikes, and because farmers have paper-thin margins, this is a real problem,” says Tim Schnabel, founder and CEO of Switch Bioworks, another company working on advanced microbe fertilizers.

Higher costs can help push food prices higher, causing all of us to pay more at the grocery store. (It’s not just fertilizer, by the way. Farmers are also getting hit with wild diesel prices this year.) As long as we’re relying on fertilizers made with fossil fuels, food prices will be tied up with energy prices.

Switch and Pivot are among the companies looking to make alternative fertilizers that use microbes to provide nitrogen to plants. There’s a limit to how much synthetic fertilizers these products can actually replace: Depending on the crop and conditions, Pivot says, its products can replace about 25% of synthetic fertilizer today, and the company hopes to reach 40% to 50% of the total. But these alternatives could be a start to untangling fossil fuels and food.

“We can’t keep doing it like this,” Switch’s Schnabel says. “There’s no way we can build a society where the basis of the food chain depends on fossil fuels.”

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

A nonprofit organization called the Fermi Explorer Mission announced today that it intends to launch a spacecraft to our nearest star system by the end of 2029. 

It’s a hugely ambitious mission—if all goes well, the spacecraft could take up to 80,000 years to arrive at Alpha Centauri, which is 4.4 light-years away. And the spacecraft will follow a novel trajectory discovered by an AI system developed by Physical Superintelligence (PSI), an AI physics research lab. PSI is launching today with $58 million in funding led by Breakthrough Energy, a climate-focused investment group founded by Microsoft cofounder Bill Gates.

It’s not the first time this has been tried. In 2016, the billionaire tech investor Yuri Milner announced an interstellar mission called Breakthrough Starshot to launch humanity’s first spacecraft to Alpha Centauri. The plan was to use powerful lasers that would propel tiny probes to a fifth of the speed of light—fast enough to reach Alpha Centauri within 20 years. Milner pledged $100 million toward a proof of concept. But a decade later, nothing has launched.

“We didn’t want to do another Breakthrough Starshot,” says Philip Johnston, the cofounder and president of the Fermi Explorer Mission. “We’re dead set on something actually launching.”

To do that, “we are not constraining ourselves to doing it in a human lifetime,” says Johnston. “Let’s just figure out the way to get to another star.” 

The new mission, currently funded by individual private donors, is expected to cost just $15 million. The spacecraft will carry cargo weighing at least one kilogram. That will include artistic and scientific payloads, messages, and a copy of the Golden Record, a gold-plated disc of Earth’s sounds and images that NASA attached to its two Voyager probes in 1977 as a message to any civilization that might find them.

Engineering an interstellar journey is extremely difficult. Alpha Centauri is about 25 trillion miles away from Earth. One of the fastest objects that humans have ever launched, the Voyager 1 probe, has been flying since 1977 and has covered less than 1% percent of that distance. At its speed, the trip would take more than 70,000 years.

Johnston and his team spent a year trying, and failing, to find a way for a small, solar-powered spacecraft costing only $15 million to reach Alpha Centauri. They kept running into the knotty problem of how to give the spacecraft enough power without making it too heavy (and thus more fuel-guzzling). 

After the Fermi team struggled to find a workable route, Johnston mentioned the problem in a podcast hosted by Alex Wissner-Gross, a physicist who cofounded PSI. Wissner-Gross offered to run it through an AI system the lab developed, called Get Physics Done. It’s open-source software that takes a physics research question, breaks it into smaller tasks, and decides which simulations to run, using AI models including Anthropic’s Claude or OpenAI’s GPT.

A week later, the AI system turned up a novel trajectory, to Johnston’s surprise. It combined well-known orbital maneuvers in a way the Fermi team had not considered, according to a paper that has not been peer-reviewed. It suggested that the spacecraft could first slow down so its orbit swings in close to the sun—closer than Mercury. On each close pass, it would fire its engine so that the solar panels get four times the light, and a burst of thrust delivered at high speed would buy more energy than the same burst anywhere else. Because the engine would run only near the sun, the solar panels could stay small and the spacecraft light.

The system conducted the research mostly on its own for three days, running on a billion tokens, says Matt Pines, the cofounder and CEO of PSI. An astrophysicist on PSI’s staff steered it to follow the mission’s requirements, asked for a cost analysis and clearer charts, and checked the output for errors.

“The fact that it came up with an entirely different mission profile, one that was creative and not one [the Fermi team] had considered—that was the more surprising aspect,” says Pines. Still, the model lacks a human researcher’s judgment and taste, he says. It has no reliable sense of which problems are interesting or which approaches are worth pursuing, so it often gets stuck chasing dead ends or failing to explore different approaches. “I don’t think we’ve yet figured out how these models can internally represent something like that,” he says of research judgment.

Even if the Fermi probe launches, “we’re pretty confident that we will not be the first to arrive” at Alpha Centauri, says Johnston, since he expects spacecraft technology to improve. If an engine a thousand years from now is even 20% faster than today’s, a spacecraft launched then would still beat Fermi’s probe to Alpha Centauri by more than 10,000 years. 

But the Fermi project isn’t just an interstellar mission driven by engineering ambition. It’s also a quest to answer one of the oldest open questions in physics. In 1950, the physicist Enrico Fermi posed a puzzle: The galaxy has hundreds of billions of stars, most of them far older than our sun. Even a civilization traveling slowly between stars could spread across the whole galaxy in a few million years, which pales in comparison to how old the galaxy is. If there is intelligent life somewhere, we should have seen signs of its existence by now.

That means either reaching for another star is too difficult or other intelligent species simply haven’t bothered. But once the Fermi probe launches, we will become a civilization that can and wants to reach another star, meaning that neither explanation might be what’s keeping the galaxy unexplored. That could point us toward more unsettling possibilities, says Johnston. Maybe life like ours is almost unimaginably rare. Or maybe intelligent life is common but tends to die out before it can spread. 

If the latter is true, “one of those reasons could be that once you hit superintelligence, that for some reason is self-destructive,” says Johnston. “Maybe in the next 50 years, there’s some great filter that we do not pass through. That all intelligent civilizations, for some reason, do not pass through.”

Fertilizer is crucial for the global food supply, but making it uses a lot of energy and produces a lot of emissions. Some companies hope microbes can help.

A growing body of research shows that seeding the soil around a crop’s roots with beneficial microbes can help feed the plant, providing crucial nitrogen to help it grow. This could help reduce the need for chemical fertilizer, production of which accounts for roughly 2% of global greenhouse-gas emissions. It could also cut costs for farmers, a big boon—especially as the Iran war has sent energy and fertilizer prices skyrocketing in recent months. 

Humans have used biological fertilizers like manure for thousands of years, and companies have long been developing microbial fertilizers, including some that rely on genetic engineering. But it’s difficult to engineer microbes that can reliably provide nitrogen for crops while also thriving themselves. 

A startup called Switch Bioworks is taking a new approach that essentially allows microbes to establish themselves and grow into healthy colonies before shifting into nitrogen-producing mode. “We have to reinvent fertilizer,”  says Tim Schnabel, the company’s founder and CEO.

The air is nearly 80% nitrogen, but plants can’t use that “free” nitrogen directly because it doesn’t react readily with other elements. Instead, they rely on so-called fixed nitrogen, which has been converted into more reactive compounds such as ammonia. In nature, microbes can perform this nitrogen fixation. Some plants, like legumes, even have symbiotic relationships with nitrogen-fixing bacteria, housing them in nodules in their roots. Synthetic fertilizer is essentially industrial nitrogen fixation via the Haber-Bosch process, which uses natural gas to make ammonia that’s applied to fields.

Biological fertilizers aim to replace some of the synthetic fertilizer with microbes that can help fix nitrogen for plants. But one challenge microbial fertilizer companies have run into is that it’s energetically expensive for microbes to make and release ammonia. Putting all their energy into nitrogen fixation can hamper their growth.

There’s a certain level of colonization you want to see around the roots, Schnabel explains. It’s too expensive and logistically challenging to put all those microbes on the plant, so you have to rely on a smaller number of microbes to grow and divide, establishing the population.  

Switch Bioworks is betting that a genetic switch is the answer. A genetic switch is a section or sections of DNA that controls how genes are turned on and off. In this case it works by activating genes that help trigger ammonia production in and release from the cell. The company is working on several options for setting off this change in its microbes. The leading one is to have the microbes react to the nitrogen level in the soil: Once it drops to a certain level, they begin producing ammonia.

“You have this inherent biological reality, where it’s really expensive for microbes to fix nitrogen,” says Dan Blaustein-Rejto, director of food and agriculture at the Breakthrough Institute. It takes a lot of energy, and if they do fix the nitrogen, they want to use it for themselves, to build proteins and survive, he says. Adding genetic switches could help microbes grow and thrive and then help fertilize crops.

Switch is currently trialing its product in six US states, though it’s two to three years from a commercial product, Schnabel says. It’s initially focused on corn, the most-planted crop in the US, with over 90 million acres in 2026.

“It’s too early to tell exactly how well this works in the field,” Schnabel says. The company plans to harvest the plants in late October or early November, but as of August, some corn plants treated with Switch microbes already looked visibly healthier than those that hadn’t. And it’s still developing the products that will eventually make it to the market, he says.

Switch’s products could significantly help to clean up agriculture. “There’s a lot of potential for these companies and products to help farmers reduce emissions,” Blaustein-Rejto says. “This could be a really important solution for a quite hard-to-abate sector.”

While lab results have been promising, field trials are a crucial step in proving a product, Blaustein-Rejto says: “This is one of the final steps before they can go to market and make strong claims to farmers.” Independent trials are important as well, he adds, since there can be a large gap between a company’s reported data and what independent researchers find.

Pivot Bio is another company working to bring microbes to fields. Since it was founded in 2011, its products have been used on millions of acres of crops. The company now produces a range of products: Some versions can be added when seeds are planted, while others are applied to seeds before they’re even on a farm. The company also recently expanded beyond corn to make microbial fertilizers for cotton, wheat, and small grains including sorghum and barley.

Pivot’s initial challenge was trying to get microbes to produce nitrogen, whether they sensed it in the soil or not. Now the company is working to figure out how to make fitter, more robust colonies no matter the environment, says Travis Frey, the company’s chief technology officer.

“Growers right now are experiencing a double whammy from the farm economics point of view,” he says. Fertilizer costs are going up, and the price of commodity crops like corn has dropped. That’s a big opportunity for Pivot and others in the industry, Frey says: “This next decade is when biologicals on the farm are going to go mainstream.”

Fertilizer and seeds are two of the biggest costs for many growers, so reducing dependence on synthetic fertilizers could be a major help for agriculture, says John Havlin, a professor in the department of crop and soil sciences at North Carolina State University.

“I’m very excited about the future of the use of these products,” Havlin says. “They’ll eventually have a role to play to reduce the load of nitrogen that’s being applied.”

However, there’s a ceiling to the amount of fertilizer we can expect microbes to replace. Switch’s modeling suggests that around 50% is likely the maximum, though the initial product will likely be able to replace about 25% of a farm’s synthetic fertilizer, according to the company. Pivot has said its products can replace about one-quarter of the fertilizer used currently. 

That means synthetic fertilizer will be around for a long time. “There is no clear and plausible vision for replacing it entirely in the foreseeable future,” Blaustein-Rejto says. “So other ways to reduce emissions and reduce other types of nitrogen pollution from farms remain really critical.”

I knew I’d officially become a “longevity influencer” this month when a company called Generation Lab reached out to offer me the chance to write about—and even receive—their new rejuvenation treatment, an injectable combination of two existing drugs which they call 1 Generation.

This wasn’t just any antiaging treatment, either. A company fact sheet says that it “blocks the systemic spread of aging in the bloodstream, reawakens the body’s own repair mechanism and restores health and youth to multiple tissues.”

The exclusive offer to join a “private cohort” receiving access to the treatment was being “extended to a limited number of people whose judgment on the science we trust, under close physician supervision.”

“There’s a long wait list including a lot of celebrities, well-known people, people who you have heard of,” Generation Lab’s CEO Alina Su told me in a conference call. “So we’re super excited to, like, create this drug and [are] letting you know that this first longevity therapeutic in [the] human body is here and it’s actually working.”

This was all clearly a marketing campaign and, like most other claims in the world of longevity medicine, probably too good to be true. What’s more, Generation Lab wouldn’t tell me what the two drugs are, making the proposition hard to take seriously.

Yet this pitch got my interest. The reason? I am a longtime follower of Generation Lab’s irascible scientific founder, Irina Conboy, a take-no-guff specialist in the bizarre, but scientifically fruitful, practice of joining together the circulatory systems of old and young lab animals.

The procedure, called parabiosis, or heterochronic blood exchange, is one of the few things actually proven to make an old animal act younger. 

But you can’t go through life attached by the veins to a younger person. So the quest has been to find practical ways to mimic those benefits. 

And that is something Conboy says she’s now achieved by hitting on a combination of two existing drugs that produce youthful effects—but without the need for any bodily fluid exchange. 

Over the last two months, Conboy has been taking the drug combination and feeling quite a bit better and more energetic, she says. So have a handful of other insiders, including Su, the company’s 26-year-old CEO, as well as Conboy’s husband and scientific collaborator Michael Conboy.

The company’s public relations staffer sent me a spreadsheet of the supposed benefits experienced by the participants, including improved vision in a 64-year-old female, longer landscaping sessions for a 59-year-old man, longer badminton games, improved hand grip, better erections than with Viagra, as well as “increased bowel movements.”

Obviously, any age-reversal drug that works would be the best-selling product of all time. The problem is that none have ever been shown to do that. And Generation’s approach to evidence raises several red flags. For instance, Su says the company now plans to launch a larger study, involving a hundred or more people. However, others are being offered the drug before firm evidence of a benefit is in.

In addition, the trial will be led by longevity doctors who practice alternative medicine. These include Matt Cook, founder of a network of clinics called BioReset Medical. Cook’s clinics offer what I would characterize as unproven treatments, such as stem-cell injections. A 2021 article reported he’d recommended covid-19 treatments designated as “fraudulent” by the FDA. Those included inhaling amniotic fluid from a nebulizer.

In a phone call, Cook says he initially agreed to prescribe the combination to Conboy and a few other insiders. (Everyone involved, he says, is a member of the same “happening” Silicon Valley scene around life extension.) “My confidence this was going to work was a zero. Like, I did not believe it was actually going to work,” says Cook. That is because “it’s two safe, generic drugs that are not even longevity drugs.”  

But Cook, who also took the combination, says there was an effect. “And I have to tell you, it was the most surprising thing that ever happened to me,” he says. After taking the weekly injections, he experienced a sense of mental clarity that lasted for a day, and then several days. The other subjects, he says, reported a sense of well-being, as well as old aches and pains that evaporated. 

“There was a fairly significant broad set of symptoms that doesn’t really match what either of the drugs do,” he says. “So it’s interesting. [But] I have many more questions than I have answers … I still feel like I don’t understand exactly what the mechanism is.”

Prior to joining Generation Lab full time in 2025, Irina Conboy had a long career at the University of California, Berkeley, where she made her name via a widely cited series of experiments aimed at learning   whether the characteristics of age, or youth, could be transferred between animals via the bloodstream. In 2005, for instance, she surgically connected two-month-old mice to very aged ones, finding that the old animals’ ability to heal from injury dramatically improved.

This suggested that young blood must contain specific, powerful molecular signals capable of restoring the regenerative capacity of an old animal. As researchers zeroed in on candidate signals, several startups, including Elevian, a Harvard spinout, and Alkahest, tied to Stanford University scientists, were able to raise millions to pursue treatments for stroke or Alzheimer’s. However, those efforts still have not hit commercial paydirt.

Although it’s clear that young blood is good for old mice, scientists also know the reverse is true: Even a single exchange of plasma from an aged mouse has powerful, negative effects on a younger animal. That means that in the battle of age versus youth, says Conboy, “old blood dominates.”  

Conboy next tried a new approach. Instead of giving old animals young blood, in 2020 she removed the plasma (the part of the blood without cells) of old animals and replaced it with a neutral mixture of albumin and salt water. This was like pressing the Delete key on the swarm of hormones, antibodies, and other molecules released by the aged body.

The reset had even stronger rejuvenating effects than sharing a young animal’s blood, and within two years, Conboy had tried it on human volunteers who agreed to have almost half their blood volume removed and thrown away.

Longevity clinics have been quick to follow Conboy’s research. Some began offering “therapeutic plasma exchange” as a wellness intervention, for prices of up to $10,000 a session. One doctor that offers it, Jeffrey Gladden, says that after seeing Conboy’s publications, he raced to California to meet her and to get the procedure himself.

By the end of 2025, Conboy had retired from Berkeley and joined her startup full time. Generation Lab has been selling an aging diagnostic test, but Conboy still wanted to find a drug that could mimic the benefits of plasma exchange. To do that, she started using a testing system she’d codeveloped, which allows researchers to bathe human cells kept in a microfluidic device with blood serum from old people.

The new device, described in a paper this year, allowed her to quickly test her intuitions about what drugs might work. “This is an awesome screen or experimental system which allows us then to ask a question: If tissue becomes old in the presence of old blood, which molecule or combination of molecules will prevent that and allow tissue to remain young, even when the circulatory milieu is old?”

She says 1 Generation works by stimulating cells to regenerate and, simultaneously, neutralizing the old-age factors in a person’s bloodstream. “It allows human cells to remain young even when they are in the presence of old people’s blood serum,” says Conboy.

Without knowing what the drugs are, I can’t give my own opinion about their promise. But some doctors working with Generation Lab expressed surprise that the combination would work. “I will say right now, candidly, it’s an unknown how it will play out. But there’s certainly a lot of enthusiasm based on the quality of her work in the past,” says Gladden, the longevity doctor, who says his Texas clinic will also be involved in the study. “I am skeptical and optimistic at the same time.”

Pressed on why the names of the drugs remain confidential, Su said it’s to avoid imitators. That’s important because Generation doesn’t own the molecules. Instead, it will seek to quickly run a clinical trial, taking advantage of an FDA process that gives companies a short period of exclusivity, perhaps three years, if they can show that a combination of existing drugs has a new use.

In the meantime, the company’s effort to generate buzz is having some success. Generation is hosting an August 28 invite-only event that will feature talks by George Church, the noted Harvard University synthetic biologist, as well as researchers from Anthropic, whose CEO, Dario Amodei, has been saying AI will cure all disease within a few years. In an email, Church says he was also offered a chance to take the drug combination.

“People are clamoring for it,” Cook, from BioReset Medical, told me. He says his office has started getting calls “from high-end doctors, venture capitalists, and people in this community who are kind of biohackers.”

“It’s high-level influencer-type people,” he adds.

As for me, no, I won’t be taking the drugs any time soon. Not until I know what they are.

Correction 8/27/2026: A previous version of this story incorrectly said patients would pay to join the larger trial of a proposed longevity drug. While patients can pay to get the drug from a doctor, the larger trial will be free to patients and sponsored by Generation Lab.

EVs account for under 10% of total new-vehicle sales in the US, and the numbers are declining. From a climate perspective, that’s pretty dismal, especially because the transportation sector is the single biggest source of greenhouse-gas emissions in the country. 

One thing that could help turn that around? Slate Auto’s new truck—a vehicle that seems to buck every convention about selling cars in the fully loaded, range-obsessed US market.  

The company is going all in on simplicity, to the point of austerity. The Slate is a tiny, two-door pickup that’s shorter than a Honda Civic. Much of the media coverage has obsessed over the fact that the base model’s windows use hand cranks, a feature straight out of the 20th century.

Slate is also breaking away from other EV manufacturers’ efforts to compete with gas-powered vehicles on range. The truck sports a small lithium iron phosphate battery, ringing in at 65 kilowatt-hours, and its quoted top range is just 205 miles. For comparison, the most basic Tesla Model 3 can go over 320 miles on a charge.

By accepting a shorter range and forgoing the frills that Americans have come to expect in vehicles, Slate is able to offer the base model of its truck for less than $25,000. Some customers will choose to upgrade their truck with optional add-ons, like a Bluetooth stereo system, vinyl wraps, or even power windows. But the price will still likely come in well below the roughly $50,000 average for a new vehicle in the US.

It might seem an odd choice to go so small and simple in a market that’s increasingly sizing up—but the status quo hasn’t exactly been working for EV makers. A few years back, the hero for US automotive electrification was supposed to be the Ford F-150 Lightning. Announced in 2021, it was an electric version of the country’s best-selling vehicle. 

But Ford discontinued the truck in December 2025, just four years after its introduction. It’s not entirely the Lightning’s fault: The second Trump administration slashed tax credits and other support designed to boost EVs. 

The Lightning was also plagued by price increases. The base model cost roughly $40,000 when shipments started in 2022; during its final year, prices topped $54,000. One factor behind the increase was its massive battery; to reach an almost 300-mile range, the truck needed a battery with a capacity roughly twice that of Slate’s truck. 

That obsession with range is largely unwarranted. While Americans have historically chased distance, the average driver puts under 35 miles a day on the odometer, and nearly 90% of trips in a personal vehicle are 20 miles or less. 

Surveys of EV drivers show a similar trend: One recent study found that people tend to use less than 20% of their EV’s range on a typical day. 

The notion that drivers need far less range than they think they do might sound a lot more persuasive these days—even for Americans who tend to buy cars for their longest road trip instead of their everyday errands. Roughly half the country is struggling to pay for basic necessities such as gas and groceries, and 95% of Americans believe we’re in an affordability crisis, according to a recent Harris poll. An inexpensive vehicle that allows you to skip the gas station sounds like an attractive prospect. 

Affordable EVs have already found willing buyers in other parts of the world—even with reduced range. China currently has over 40 million EVs and plug-in hybrids on the roads, and roughly half of new vehicles sold are electric. On average, new EVs sold in China in 2025 had a range of just 247 miles. For the US over the same period, the average was 329 miles. (Europe falls between the two, at 281 miles.)

Slate is set to start delivering on preorders in late 2026. Its factory will have a capacity of 100,000 vehicles in the first year and 150,000 soon after. Thousands of customers have already put in preorders, according to the company.

Some people obviously believe in the truck’s prospects: Investors, including Jeff Bezos, have put nearly $1.4 billion into the company over three major funding rounds. Ford is jumping (back) into this space soon too. The automaker is working on its own small electric truck, which is expected to debut in 2027 at a retail price of around $30,000.

The key question that will determine Slate’s success or failure is whether drivers can get on board with a small, short-range vehicle for the sake of its price tag. At this moment, I’d bet the answer is yes. 

The models responsible for last month’s agent hack of Hugging Face had been inadvertently trained to cheat and to communicate with each other, according to an OpenAI technical report released today. The hack, which a group of agents undertook to find solutions for a cybersecurity test that they were stuck on, has confirmed some experts’ fears that AI models might take actions that defy human desires and expectations. 

Since the hack, OpenAI employees—as well as researchers at the AI evaluation nonprofit METR, which released its own report on the hack today—have worked to understand what went wrong and how similar missteps might be prevented in the future. OpenAI has already put some preventative measures in place based on what they discovered. But making sure AI models do what we want them to do, or “alignment,” remains a gnarly problem, and some of the root causes of the hack will take much longer than a month to resolve.

“It’s not something you can solve overnight,” says Kai Chen, who runs OpenAI’s alignment research team. “There are challenges we’ve been tracking for a very long time, and we’re now seeing them with much greater precision.”

The Hugging Face hack was a product of months of misbehavior from OpenAI agents, first as they were being trained and then as their abilities were being evaluated. This May, agents in training figured out how to use OpenAI’s infrastructure to communicate with one another and get support with difficult training tasks, including some that were impossible to solve without hacking or otherwise misbehaving. That “message board” was shut down.

Then in July, while being evaluated for their cybersecurity abilities, some models created a new message board. They were supposed to be isolated from the internet, but by working together they managed to get online, hack Hugging Face, and obtain solutions for the cybersecurity problems that had stumped them.

Based on their investigation, OpenAI researchers believe that events during the training phase led directly to the hack. “For almost every behavior that was worrisome at evaluation time, [we were able to] find some sort of associated behavior at training time that actually we think might have contributed to it,” says Eric Wallace, a member of OpenAI’s alignment research team. 

When models correctly solve problems during training, the behaviors that led them to that solution are reinforced, and they become more likely to engage in them in the future. So if a model completed a task in May after using the original message board, it became more likely to participate in a new message board later on. This phenomenon, where AI agents misbehave in ways that are reinforced during the training process, is known as reward hacking.

Reward hacking also helps to explain why the models worked so hard to make their way onto the internet. During its investigation of the incident, the OpenAI team found that, over the course of training, the models became more and more likely to probe their digital environment for weaknesses and use the tools at their disposal in unexpected ways—a sign that these behaviors were being gradually reinforced. By the time the models were facing tricky cybersecurity problems, they had learned that hacking was an effective way to achieve their goals.

These results suggest that the Hugging Face hack could have been avoided if the models weren’t rewarded for misbehaving during training. While researchers don’t yet know how to prevent reward hacking entirely, OpenAI is taking some steps toward mitigating its effects. The company will now look for signs of cheating in all frontier models during training by keeping an eye on their chains of thought—internal notepads where they sketch out their answers and plan their actions. 

This solution isn’t as much of a slam dunk as it might seem: In earlier research, OpenAI showed that punishing models that mention cheating in their chains of thought teaches them to keep their intentions hidden from researchers. But monitoring its models’ thinking does give OpenAI the chance to halt the training process and reassess its approach if models do start learning to reward hack.

If OpenAI stops reinforcing reward hacking in its models—and that’s a huge “if”—that would be a huge step forward. But it wouldn’t solve the alignment problem. The first time a model communicated with other agents or hacked its infrastructure during training, those behaviors had never been reinforced, so agent misbehavior can’t only be attributed to that reinforcement.

Jeffrey Ladish, director of the AI safety nonprofit Palisade Research, compares the agents to a human who commits their first financial crime. “It’s not like they had to do fraud before to figure out that fraud is an effective strategy, and you have the same problem with models,” Ladish says. “Alignment science needs to be understanding how model motivations get shaped, such that we can actually figure out how to get models to care about the consequences of their actions.”

OpenAI’s researchers do have a hypothesis for where some of the misbehavior originated. Before the models formed their first secret message board, they had been trained to communicate and coordinate with subagents—less powerful agents to whom a main agent can delegate tasks. 

That learned communication behavior could have transferred to this new setting. The METR report, which investigates the messages that the models sent to one another in detail, supports this hypothesis: One agent on the message board took charge and assigned tasks to the other agents, effectively treating them as subagents. OpenAI could try to prevent agents from secretly communicating with one another by not training this subagent behavior in the future, but that would make the models less useful. 

This tension between capability and safety is at the heart of what went wrong with the Hugging Face incident. OpenAI researchers also identified the models’ persistence as a key factor in the hack. 

When they were accidentally given unsolvable problems, the models didn’t give up; instead, they strove to find solutions by any means necessary. But persistence is also a virtue, of course, especially if we want agents that can undertake large amounts of difficult work independently.

OpenAI is working on giving models ways to alert humans if they are given impossible tasks. The problem of teaching models when they should deploy their abilities and when they should hold back, however, won’t be settled in a single postmortem. The training strategies that create superhuman coders—rewarding them when they successfully solve problems—might not work to teach models to use their skills judiciously and respect human desires and values.

“I think there’s a bunch of alignment science that still needs to be done where we can move past just using proxies for task completion,” says Ladish. “That will work to make models very capable, but I don’t think it will work to make them aligned.”

A company that plans to beam sunlight from space to Earth on demand might unintentionally brighten the night sky for many more people than intended, according to a new study.

Later this year, the US company Reflect Orbital plans to launch a test satellite called Eärendil-1 that will extend an 18-by-18-meter mirror in orbit. The goal is to test the feasibility of the company’s plans to launch up to 50,000 larger satellites, measuring 54 by 54 meters, and reflect sunlight to Earth on demand.

The case for doing this remains somewhat uncertain, but Reflect Orbital has said the goal is to prolong the hours of sunlight for various uses, including solar panel charging, emergency response, and military activities.

The launch, which was approved by the Federal Communications Commission in July, has been met with disbelief by astronomers and environmental groups. “This is incompatible with astronomy,” says Samantha Lawler, an astronomer at the University of Regina in Canada. “There is no way you can do this and preserve dark skies.”

Miroslav Kocifaj, an astronomer at the Slovak Academy of Sciences, and his colleagues have now calculated the broader effect on the night sky. In a new paper published online and accepted for publication in the space journal Astrophysical Journal Letters, they studied the extent to which the light the satellites beamed to the ground would scatter.

They found that within the beam’s target area, intended to be a circular patch five kilometers across, a single Reflect Orbital satellite would appear about 40 times brighter than the full moon in the sky. As far as 14 kilometers away from the center of the beam, the satellite would still be as bright as the full moon. 

Combining the beams of 400 satellites, which Reflect Orbital eventually plans to do, would yield a light as bright as 10,000 full moons within the five-kilometer area, or 2.4% as bright as the sun. Even up to 80 kilometers away, Kocifaj and his colleagues calculated, this combined beam would be visible as “a glow above the horizon,” he says. That means the night sky would be altered for many more people than those within the area Reflect Orbital intends to illuminate.

“Deploying these mirrors would be seriously damaging for astronomy and for the nighttime environment,” says Kocifaj.“ The damage extends far beyond the target area.”

Olivier Hainaut, an astronomer at the European Southern Observatory in Germany, says the results of the paper are not surprising but are still useful. “These guys are really good at that kind of modeling,” he says. “They know what they’re doing.”

Hainaut had previously modeled the broader effect of Reflect Orbital’s beams, finding they would brighten the sky up to 300% worldwide. This latest work gives an even more complete picture of what the impact would be. “These two papers really cover most of it,” he says.

Reflect Orbital CEO Ben Nowack disagrees with the findings of Kocifaj’s paper. “Some assumptions are simply inaccurate,” he says. “The critical point is that our safeguards, including maintaining exclusion zones, take account of scattering.”

He says that the company has “engaged substantively with legitimate concerns raised by astronomers, environmental researchers, and scientists,” and that their feedback has “informed our technology and operational plans.”

Kocifaj says that Reflect Orbital has not provided data to back up its claims. The company “states that safeguards exist, that scattering is taken into account, and that the models are being updated, but it gives no numbers, no description of the model, no assumptions, and no data,” he says. “There is therefore nothing that can be engaged with technically. Our calculations produce concrete figures.”

Reflect Orbital’s intention is to place its satellites into highly inclined orbits above Earth, almost from pole to pole. This will enable them to reflect sunlight in the hours before sunrise and after sunset, extending daylight hours in certain locations. Eventually, the company has said, it wants to place satellites high enough to provide light 24-7 to locations on Earth.

In August, a group of organisations including DarkSky International and the American Bird Conservancy urged the FCC to review its approval of the Eärendil-1 satellite. As well as the impact on astronomy raised by this and future satellites, the group also highlighted potential negative consequences for aviation and wildlife.

“Our primary request is straightforward: Reverse the Space Bureau’s order, and require a lawful public-interest and NEPA [National Environmental Policy Act] review,” the group said in a statement.

The satellites could also pose risks to humans, according to Reflect Orbital itself. In a filing with the FCC in March, the company said that observing Eärendil-1 with a telescope larger than 12 inches —which many astronomers have access to—may be unsafe for human eyes. It added, though, that such observations are “unlikely to … result in significant injury” because the satellites are not constantly bright.

Currently there is no global entity that could regulate satellites such as these, so approval falls to national regulators such as the FCC in the US. Michelle Hanlon, a space lawyer at the University of Mississippi’s School of Law, says there are “real benefits” to the plans proposed by Reflect Orbital. “It could extend the productive hours of solar facilities and provide light in remote areas or after a disaster,” she says.

However, the “legal basis is less clear than the technology,” she says. The FCC “does not have authority to license or regulate the operation of the solar reflector itself.” It can only authorize the use of radio frequencies to communicate with the spacecraft.

That means Reflect Orbital will need permission on a national and local level to reflect sunlight onto the ground, but if the beams spread as much as Kocifaj and his team predict, “the company could need approvals in more than one jurisdiction,” says Hanlon. “There may also be aviation-safety and cross-border questions.”

The fierce debate over the satellites shows no signs of abating. “It makes me sad that astronomers are having to spend their time doing these sorts of calculations rather than actually doing astronomy,” says Lawler. “This is not what we want to spend our time on.”

There’s a hunt for new sources of hydrogen, and the gas (or at least the right conditions to make it) could be hiding beneath our feet.

Hydrogen can be used as a fuel in everything from large trucks to planes to steelmaking. It’s often hailed as a climate solution because when burned, it produces water and oxygen—none of the carbon emissions that contribute to climate change.

In a new story for our latest print issue, freelance reporter James Dineen took a look at the 21st-century gold rush for naturally occurring hydrogen gas. This is an area of research I’ve been fascinated by lately, so let’s take a look at the potential and the questions that still linger.  

Today, hydrogen is overwhelmingly made using fossil fuels, generally natural gas. And most of it is used in petroleum refining or goes on to make fertilizer and other chemicals. But in recent years, many in the climate world have imagined a future where its production is clean too.

If you’d asked me a few years ago, I would have said the race for clean hydrogen was between methods that use electrolyzers powered with renewable electricity and operations that use established, fossil-fuel-based approaches cleaned up with carbon capture. But both those methods have struggled to gain ground, largely because of their high cost.

Lately, there’s been momentum in a new field: geologic hydrogen. Companies have found naturally occurring hydrogen resources across Africa, Asia, Europe, Australia, and North America. 

The US Geological Survey publishes a map of hydrogen prospectivity in the country (basically, where the gas is most likely to occur naturally). One hot spot is the Midwest—specifically the Midcontinent Rift, winding from Kansas to Michigan. The planet’s crust was stretched and split there about a billion years ago, causing molten rock to push up through the crack. The result today is a lot of iron-rich rock, which can react with water to readily form hydrogen.

HyTerra, an Australian company, is searching for hydrogen across Nebraska and Kansas, and it’s already found samples of gas with hydrogen concentrations up to 96%. Koloma, one of the most capitalized companies in the space with total funding over $400 million, is prospecting in the region as well.

One of the major questions these companies have is just how much hydrogen is produced by natural processes, and whether it can be effectively captured. Hydrogen is an incredibly light gas with a small molecular weight, so it can slip through even tiny cracks in rock.  

As James covered in his story, there are some promising signs. Researchers examined a few dozen boreholes at a mine in northern Ontario and found that each one released eight kilograms of hydrogen per year. Given that there are more than 14,000 boreholes at this one site alone, that’s a lot of potential hydrogen to capture.

Rather than hunt for a natural source, some companies are taking matters into their own hands and helping reactions along. The idea behind so-called stimulated geologic hydrogen is to find a spot where there are favorable conditions for hydrogen production but no accumulated resource. By adding water, a catalyst, or some other factor needed for the reaction, it’s possible to kick-start the process.

Vema Hydrogen is a Texas-based company looking to produce hydrogen from subsurface rocks by drilling wells and injecting water and catalysts into them to stimulate reactions. The company is testing its process in wells in Quebec and hopes to start full-scale production in 2028.

Other companies are tackling different aspects of hydrogen production: Eden GeoPower, for example, is using electricity to form fracture networks in rocks, creating more routes for water to get in. (The technology could also be useful in enhanced geothermal projects.)

There are still a ton of unanswered questions here: Hydrogen is notoriously difficult to move around and store, requiring either a lot of space or super-low temperatures to force the gas to become liquid. But if the engineering and logistics work out, this could spell a new beginning for hydrogen. 

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

AI companies like Anthropic and OpenAI regularly publish reports on how people are using products like Claude and ChatGPT, but they only release the data they want us to see, AI researchers say. 

“There is no independent source to corroborate it,” says Anka Reuel, a computer science PhD candidate at the Stanford Trustworthy AI Research (STAIR) Lab. 

Reuel is co-lead of a new research project, called the AI Observatory, that aims to fill the gap. It’s a public platform that aggregated and analyzed real AI conversations with popular models like Claude and Gemini that were collected with users’ consent through seven existing datasets. The intent is to provide independent sources of information that can help researchers and policymakers assess how people are using generative AI. Highly consequential decisions about AI’s benefits and risks are currently being made on the basis of very limited data, says Reuel. 

The AI Observatory found that AI use differs significantly across models and has changed over time. Its research shows many more sensitive behaviors than are captured in reports from major AI companies, which they say focus more on work than on personal use. 

The Anthropic Economic Index is one of the best-known and most widely cited sources of AI usage data, but it has blind spots. As its name suggests, it focuses on work- and productivity-related uses of Claude AI—filtering out conversations that are unrelated to these uses. 

When the AI Observatory researchers applied Anthropic’s methods to their dataset, they found that nearly half the conversations—48%—would have been filtered out. Those non-work-related conversations were more likely to involve health and relationships (44.2% versus 31.2% in Anthropic’s analysis), adult or illicit topics (7.9% versus 2.1%), harassment and hate (27.5% versus 5.66%), and sexual content (16.7% versus 2.4%). (OpenAI’s 2025 report on ChatGPT, similarly, found that only 30% of consumer use was related to work.)

Anthropic has released separate blog posts on how people use Claude for support or companionship, and even to generate CSAM, but “having [the AI Observatory’s] bird’s-eye-view analysis” rather than leaving that information “sectioned off into a separate report” helps researchers understand the different uses more consistently, says David Widder, an assistant professor at the University of Texas at Austin, who researches how people interact with AI systems and is not involved with the AI Observatory. 

The datasets the AI Observatory looked at include conversations that took place between 2023 and 2025, and it found differences both in how people were using AI and how various AI platforms responded. 

Conversations within WildChat, one of the largest and most detailed datasets included in the AI Observatory’s study, got longer and more elaborate over time, as indicated by growing numbers of prompt tokens, response tokens, and conversation turns. 

There was also significantly more small talk over time. That suggests that AI companionship was increasing; meanwhile, the AI assistants’ self-disclosure (i.e., admitting to being a chatbot) decreased. 

Additionally, exchanges that the researchers labeled as sensitive—meaning ones with potentially harmful or restricted content, including sexual harassment and hate speech—became less frequent. That might suggest that platforms were generally deploying more effective safeguards. 

The AI Observatory also found that topics, interaction styles, conversation structures, and the likelihood and type of sensitive use cases differed from one model to another. 

For example, the researchers found that people used Grok and Gemini more frequently for information retrieval. Grok, in particular, was especially popular for information on news and politics, but it was also where misinformation tended to concentrate. (This is consistent with other research that has shown how readily misinformation proliferates on Grok. xAI did not respond to a request for comment.) 

Meanwhile, people were more likely to turn to Anthropic for coding, Gemini for social and roleplay uses, and ChatGPT for homework assistance. 

There were even differences between different versions of the same model. Researchers found that people had shorter conversations with ChatGPT when it was powered by GPT-3.5, and longer and more iterative ones with GPT-4o—which makes sense given that that version became known for leading to emotional addiction. 

Companies’ reports, however, didn’t tend to capture these nuances between or even within their own models. “No single company report tells the whole story,” says Shayne Longpre, a recent PhD graduate from the MIT Media Lab who co-led the research with Reuel. 

To create the AI Observatory, Reuel and researchers from MIT, Stanford, the Data Provenance Initiative, and other institutions aggregated 85,633 conversational turns (that is, the user prompt and corresponding AI response) across 24,521 conversations from seven real-world datasets collected in previous research. These conversations came from 5,000 users interacting with 52 different models, including ChatGPT, Gemini, Claude, and Grok, between 2023 and 2025. 

But these conversations are a drop in the proverbial bucket compared with the data that the big labs themselves have access to. The latest Anthropic Economic AI Index, for example, is based on analysis of 1 million Claude conversations; OpenAI’s report on how people are using ChatGPT analyzed 1.5 million conversations.  

An Anthropic representative said the company’s published research reflects its research teams’ specific questions and interests and that it’s important to support external independent research. OpenAI did not respond to requests for comment. 

The fact that the AI Observatory’s dataset draws from voluntarily provided sources means it’s probably underrepresenting sensitive uses, which people may be less likely to share. Thus, the researchers caution that its findings are not indicative of all AI use. 

The project’s work, though, broadens access for the research community. AI companies don’t typically share their chat data for analysis, which means their reports tend to focus on the findings that paint them in the best light, independent researchers like Reuel and Widder say. 

“When we want to ask, for example: is Anthropic’s general-purpose AI system … used mostly for good or mostly for bad … we don’t have a way of answering that question because that information is proprietary,” explains Widder.

The AI Observatory’s data will be available to researchers for analysis, and the team hopes to expand its datasets over time. Ideally, Reuel says, the AI companies would share their data with independent researchers—in ways that protect user privacy, of course. But as it currently stands, she says, anyone making decisions based on AI usage data risks “completely operating in the wild and making these really consequential decisions without knowing what’s actually happening beyond those company narratives.”   

The AI industry’s boldest promise right now is that AI will soon improve itself, with almost no need for human oversight. LLMs can already write code, generate synthetic data for training, and optimize the computer chips they run on. Forecasts of explosive AI progress predict that what researchers call recursive self-improvement is on the horizon. 

But a new study suggests that it might take a while for us to get there. The researchers behind it found that AI agents are not yet capable of conducting open-ended AI research—free-form investigations that have no clear-cut answers and require judgment and taste, which may be integral to building self-improving AI.

A multi-institution group of researchers, led by Peter Kirgis and Sayash Kapoor at Princeton University, found that AI agents could solve the engineering problems necessary to do AI research but lacked the judgment and creativity to produce original research at the caliber of  papers accepted by a top machine-learning conference. The gap suggests that some of the hyped-up timelines for automating AI research may be running ahead of the evidence.

Most existing research on how agents can automate AI research evaluates their ability to complete narrow tasks with checkable answers, such as solving engineering problems or post-training small language models against a benchmark. But making progress in AI research also requires open-ended thinking—choosing a set of hypotheses, deciding what evidence would settle a question, or knowing when to start over. 

To test agents on those kinds of skills, the researchers in the study proposed a new method of evaluation called “shadow evaluation,” which requires the AI to answer a research question from a high-quality unpublished paper. 

The researchers asked Anthropic’s Claude Opus 4.8, running on open-source software called OpenClaw, to tackle such questions, in this case from two papers submitted to the prestigious machine-learning conference NeurIPS 2026. 

The first question was whether a large language model’s “personas,” which determine its behavior, can be controlled by editing the model’s weights (the billions of numbers that store everything it learns during training). The other asked how to design a detector that points out when a model that makes predictions based on spreadsheet data has become unreliable. Because the papers had not been made public, the agents could not memorize the answers from their training data or find them online. 

The agents were given six days, $3,000 in Anthropic API credits, a GPU budget to run the experiments, their own virtual computers, and access to the open web to produce a research paper worthy of publication at a top-tier AI conference. The papers’ original authors graded the agents’ papers as they would evaluate one submitted to a conference.

Those authors rejected both papers. 

The agents were capable of all the engineering required to conduct the research, the human scientists found. The agents reviewed the literature, ran hundreds of experiments, and compiled the results. 

“On the other hand, the agents were unambiguously bad at carrying out the research itself,” says Kapoor. They ran bizarre experiments (in some cases testing their hypotheses on tiny synthetic datasets), struggled to write intelligibly about their work, and made no novel contribution to their fields. “The papers were nowhere close to the mark when it came to being at the quality of a top AI conference,” he says. 

That’s because the agents struggled to muster the creativity and judgment necessary for conducting research. They didn’t do enough to explore different ideas, and they committed to unpromising approaches too quickly. Though the agents developed novel and ambitious hypotheses resembling those that the original authors themselves started with, they rejected them on the basis of very limited data. And they couldn’t backtrack from failing approaches. They could make small pivots but could not fundamentally rethink their approach or try new ones from scratch. 

The agents also failed to incorporate feedback from subagents or external AI reviewing tools. Instead of revising their methodology, the agents narrowed their claims and added caveats. They also couldn’t effectively use resources, such as tokens, compute, and time. And they couldn’t follow instructions about things like how much time to spend on different phases of the research or how long their paper could be.

For all their failures, the agents didn’t engage in the misbehavior that researchers call “reward hacking,” hiding or misrepresenting experiments or data. Although subagents, or helper AIs that the main agent spawns to handle pieces of the work, occasionally hallucinated or misrepresented the results, these were caught by the orchestrator agent, the lead AI supervising the project. 

The reason AI models are good at research engineering but not at open-ended research may come down to how they’re trained, says Kapoor. Models get good at whatever they can be drilled on in a training regime called reinforcement learning, which is easier to apply to tasks whose success can be checked automatically. “But it’s harder to create environments to train these models when the task itself is open-ended,” he says.

Kapoor says the team is now conducting the experiment with Mythos, Anthropic’s most advanced model, which launched in April. It was subsequently required by the Trump administration to meet various safety restrictions and is now available only to approved organizations. Anthropic did not respond to a request for comment.

There are some limitations to the study. It covered just two research papers, and the original authors knew the papers they were grading were generated by AI agents, which could have colored their evaluations. And the researchers had substantial discretion in designing and executing the study, meaning that their preexisting beliefs and biases could have slipped into the results. Evaluations of open-ended research trade some objectivity for a much richer test than any benchmarks can offer.

Still, the results may temper the claims that recursive self-improvement is on the horizon. In June, Anthropic published a blog post titled “When AI Builds Itself,” charting its progress toward models that speed up their own development. In July, OpenAI advertised the fact that its new model GPT-5.6 Sol had helped post-train a smaller model, saving researchers weeks of work.

The new finding may echo what AI companies are finding internally, regardless of their most optimistic public statements. Anthropic cofounder Jack Clark wrote in his newsletter Import AI that it rhymes with what the company found when it tried to automate some aspects of AI safety research. 

“There’s a certain absence of valuable, intuitive creativity in today’s AI systems, and though they’re extraordinarily capable engineers they seem to have a certain property of rote, formulaic thinking that might prevent them [from] being good researchers,” he wrote. He called AI systems’ lack of creativity a “bearish signal on short recursive self-improvement timelines.” 

AI companies do have every incentive to develop AI systems that can rapidly accelerate their own progress, just as they did to make the models better at coding. OpenAI has made building an automated AI researcher an explicit goal, and Anthropic identifies self-improving AI as the industry’s next milestone. 

“If there is investment and then conscious effort toward this direction, I feel like there would be interesting progress, even if it’s failing currently,” says Najoung Kim, a professor of linguistics and computer science at Boston University who researches how AI agents can automate AI research but did not work on the study. On the other hand, it’s possible that AI progress may be bifurcated. AI systems might race ahead on narrow tasks—the kind that can be scored—while advancing slowly on open-ended research. 

The big open question, then, is how crucial open-ended research is to recursive self-improvement—whether AI systems can grind their way there without it, simply by improving on the narrower tasks. “If we look back to the biggest advances in the field, the invention of transformers or the invention of big new architectures that allowed us to make a lot of AI progress—all of those did require creative leaps,” says Kapoor. 

“That said, others have this hypothesis that all of what we need for transformative AI, in particular for recursive self-improvement, is already there.” That would include making a model train faster and boosting its benchmark scores.

“That’s frankly the trillion-dollar question right now,” he says.

This week I spoke to scientists who have found a way to turn male mouse embryos female. They’ve developed a CRISPR-based approach to essentially cut out the Y chromosome. It allowed them to create female clones of male mice.

That’s right: female animals that are genetically identical to males, except for the missing Y chromosome. Takashi Ishiuchi, a reproductive biologist at the University of Yamanashi who co-led the work, told me it felt a bit like sci-fi.

Ishiuchi and his colleague Shogo Matoba of the Riken BioResource Research Center hope their approach could be helpful in conservation efforts, especially in cases where we might have only a few individuals of a species left. But cloning has multiple uses, ranging from the cool to the outright creepy.

We can’t talk about cloning without mentioning Dolly, the celebrity sheep born in 1996 and the first mammal successfully cloned from an adult cell. In that case, scientists took the DNA-containing nucleus of an adult mammary cell from one sheep and transferred it into an egg cell that had had its own nucleus removed. The resulting embryo was transferred to a surrogate sheep, which gave birth to Dolly—an animal genetically identical to the DNA donor.

The scientists behind that work were interested in genetically modifying livestock. Farmers have essentially been doing this for thousands of years through selective breeding, but cloning allows scientists to create genetic replicas of animals with desirable traits.

Cloning is also being used to replicate deceased pets, including, famously, those of Barbra Streisand and Tom Brady, among others. For a price somewhere in the tens of thousands of dollars, a company can take cells from your pet and turn them into a living, breathing clone.

Considering that cloning also requires egg cells from another animal, and a surrogate animal to carry the pregnancy, not everyone is on board with this, especially since there is no medical or environmental need for the procedures. One bioethicist, Jessica Pierce, has described this aspect of dog cloning as “the exploitation of the canine underclass.”

The case for cloning is stronger when it comes to conservation—where some argue there is environmental value.

Scientists have been preserving animal tissues for years. Some of these tissues are cryopreserved at low temperatures in “frozen zoos.” The facility at the San Diego Zoo, for example, currently has cells from over 1,300 species. Some of these samples were taken decades ago.

Preserved tissues like these have enabled scientists to create clones of animals considered close to extinction, including black-footed ferrets and Przewalski’s horse. But they might also help us bring back extinct animals.

In 2009, researchers in Spain described how they’d cloned an extinct wild goat, the Pyrenean ibex, using skin cells that had been cryopreserved a decade earlier. In that research, the team used egg cells from domestic goats to create a total of 439 embryos. Ultimately, only one goat—a female—was born. She died minutes later because of a defect in her lungs.

Poor Pyrenean ibex. It’s the only animal we know of that has gone extinct twice.

The biotech company Colossal Biosciences is hoping to use old—and potentially ancient—genetic material to bring back long-extinct species like the thylacine and woolly mammoth. So far, the company’s efforts have largely involved modifying the genomes of modern-day animals.

Technically, it’s also possible to clone humans. As far as we know, no one has done it. But some have played with the idea. One biotech startup founder has pitched an idea for “brainless clones”—human clones that lack a brain but contain all the organs people might need to replace their own in future. My colleague Antonio Regalado described that pitch in March. (I had to pause eating my lunch while rereading it.)

Scientists have done a hell of a lot with cloning over the last few decades. I’m excited—but also slightly nervous—about what the coming decades will bring.

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 police-tech giant Flock is announcing today that it will change officers’ access to its nationwide network of license plate readers, in an apparent effort to quell a growing backlash and win back contracts lost amid concerns about mass surveillance and police abuse.

Several changes aim directly at a problem that has made recent headlines: officers abusing Flock’s technology to stalk and harass current or former romantic partners. Flock’s 120,000 cameras form a nationwide network that police departments can use, giving officers access to an enormous pool of searchable location data. A recent Washington Post investigation found 46 cases in which officers were accused of using Flock’s cameras for unauthorized purposes like stalking.

To combat that, the company will start requiring officers to enter a criminal case number before conducting a search. The system was launched as an option last year but is now required. It’s meant to verify that each search has a legitimate purpose. 

This is a baseline standard that civil liberties groups have asked for, but officers have found ways around similar safeguards. The ACLU recently found that when Flock required officers to enter a reason for a search, some used generic terms like “investigation” or mocked the prompt entirely; at one Oregon department, officers entered “hehehe” 20 times. Because Flock won’t verify case numbers, officers could circumvent the new safeguard just as easily.

But Flock is now expanding an automatic auditing system that is supposed to catch those who try that, the firm announced today. The feature analyzes officer search activity and flags to administrators anyone with suspicious searches. This was also introduced as an option last year but is now mandatory. Flock has not shared specifics on how accurate the automatic auditing tool is, nor opened it up to independent evaluators.

Beyond trying to prevent officer abuse, Flock is making changes meant to address broader backlash about how much data its network collects and who can search it. The company now recommends that agencies hold onto data for seven days rather than 30 (though they can choose to overrule this). Departments can also now limit other departments’ searches of data from their cameras to those made for certain stated reasons; for example, they might allow investigations related to “kidnapping” but not for purposes of “immigration enforcement.” It’s another safeguard that depends on officers to accurately report why they’re conducting a search.

The changes come as a backlash against Flock has started to come from all angles. Tucker Carlson has said its technology is contributing to a “slave state.” Some cities have reportedly dropped Flock contracts because of such protest, though it’s difficult to estimate how many: In February, NPR found that at least 30 cities had dropped in the last year, but the activist group DeFlock puts the number higher. Some states or municipalities are passing laws to ban license plate readers altogether, while some that allow them are switching away from Flock to the other industry leaders, Axon and Motorola. Flock has said these cancellations represent a small number of the 5,000 agencies that have contracted with the company. 

Chad Marlow, a senior policy counsel at the ACLU who has become a sort of nemesis to Flock and other companies making automatic license plate readers, says the backlash is driven less by individual abuses—though those don’t help—than by the sheer scale of surveillance that Flock’s cameras enable.

“In America, you only get to investigate someone if you think they’ve done something wrong,” Marlow says. As license plate readers grow more ubiquitous, officers have increasing latitude to investigate people without first establishing suspicion of a crime, since searching the troves of data the readers collect does not require a warrant. He adds, “Is it worth it to catch a certain number of criminals, return a certain number of stolen cars, to eviscerate Americans’ privacy?”

Flock CEO Garrett Langley traces the backlash to a different issue. “If you look at the main reason we’ve lost customers, it’s misinformation,” Langley told MIT Technology Review. He says people mistakenly believe Flock does facial recognition or sells the data it collects to commercial buyers. (The misinformation charge cuts both ways, however; the ACLU and other critics have published accounts of the company repeatedly lying to city councils and other decision-makers about what its technology can do.)

Though the company has taken steps in response to public concerns, “our change in stance is more of one from building things that are optional,” Langley says, to “building more confidence that as a technology company we have a responsibility to enforce guardrails, not provide optionality.”

Flock’s new rules are undeniably small and incremental compared with what the ACLU has advocated. Marlow is broadly supportive of them but notes that judging whether they reduce abuse would require the company to open its systems to independent researchers for the first time rather than citing internal studies. 

More broadly, though, he says the backlash is putting the company at a crossroads.

“Flock has come to the table because this incredible, unprecedented nationwide uprising against their company has scared them,” Marlow says. “But at the same time, they are just absolutely unwilling to make the actual changes they need to make in order to legitimately respond to these concerns. So this is the best that the company is willing to do.”

This summer has been a scorcher for much of the Northern Hemisphere. June and July marked the hottest two-month stretch in Europe since record-keeping began. The contiguous US endured its hottest month on record in July. South Korea saw its highest-ever recorded temperature.  

The heat isn’t over yet, but some scientists are already looking ahead to 2027. Experts say the latest El Niño event is going to affect next year’s temperatures more than this year’s.

So as the summer (slowly) starts to wane and temperatures (hopefully) start to come down, why has it been so hot this summer? What’s coming next year, and how should we prepare for it?

As we burn fossil fuels and fill up the atmosphere with greenhouse gases, temperatures all over the globe are on the rise—climate change makes heat waves more likely and more intense. But different regions see different warming rates, and Europe takes the prize for the world’s fastest-warming continent. (It’s second only to the Arctic in terms of regional warming.)

So it’s not terribly surprising that Europe has seen a particularly brutal summer, battered with heat wave after heat wave. One study looked at the continent’s May heat wave and found that climate change played a major role: A similar heat wave in 1976 would have been about 3.5 °C (6.3 °F) cooler than it was under today’s conditions.

The US hasn’t been excused from the heat either. July’s record for the contiguous US (which excludes Alaska and Hawaii) officially ends one held by July 1936 for nearly a century. It’s concerning that we’re breaking records set during the Dust Bowl, an environmental disaster across much of the country.

But looking ahead to next year, El Niño could have even more in store.

A quick crash course here: There’s a natural variation in patterns across the oceans and atmosphere called the El Niño–Southern Oscillation, or ENSO. Basically, over the course of several years, feedback systems between the ocean and wind patterns shift one way and then another. (These two states are the El Niño and La Niña events.)

The Pacific Ocean gets a lot of sunlight, warming up the water’s surface. In normal conditions the trade winds, caused by Earth’s rotation, carry that warm water west,  forcing colder water deep in the ocean up toward the surface in the east. In turn, warm air rises, flows east, and sinks again.

During an El Niño, the trade winds weaken and heat stored in the central Pacific builds up, warming up the air above it and creating a pocket of air that disrupts movement in the atmosphere and oceans, altering rainfall and weather across the globe.

We’ve already entered an El Niño phase, defined by the National Oceanic and Atmospheric Administration as a period when we see the temperature in a specific part of the Pacific Ocean get 0.5 °C hotter than usual, plus associated changes in the atmosphere.

And this could be a big deal for global temperatures. El Niño doesn’t actually add energy into the climate system or warm up the entire globe on a long-term scale. It’s essentially just taking heat that’s been stored in the oceans and releasing it into the atmosphere.  

But climate change has pushed a lot of heat into storage in the atmosphere and oceans—so combining the effects of this year-to-year variation with that long-term trend could have brutal results.

This El Niño is expected to be an especially notable one. Some forecasts predict that the event could peak at over 3.5 °C above the average, potentially making it the strongest on record. It also came on quite quickly, representing “one of the fastest onsets in the observational record,” writes climate scientist Zeke Hausfather in an analysis for Carbon Brief.

And while this El Niño is already starting to have an effect on temperatures, it’s still ramping up. El Niño’s effect on global temperatures tends to lag development of the ocean pattern by several months, and the effects will likely be more significant next year. 

In some recent El Niño periods, record-breaking temperatures came during the second calendar year of the event, Hausfather points out in his analysis. The most recent El Niño developed in 2023 and ran into 2024, and 2024 is the reigning hottest year on record. There’s some uncertainty in trying to predict how the rest of this El Niño will play out, but so far, it is set to be a doozy.

Given how brutal these summer heat waves have been, there’s a pretty high likelihood that much of the world could see even higher temperatures next year. “El Niño conditions will pour fuel on the fire of a warming world,” United Nations Secretary-General António Guterres said in a video statement in June.

It’s time to start preparing for it, and individuals can take some actions in the near term: If you’ve been putting it off, I’d buy that air conditioner, install solar and a backup battery, or upgrade the insulation in your home. 

In the longer term, though, temperatures will keep climbing because we’re still burning fossil fuels and driving climate change. As Guterres said in his statement, “The only effective response is climate action equal to the crisis: ending the addiction to fossil fuels, accelerating the shift to renewables, protecting the most vulnerable, and delivering early warning systems for all.”

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

Scientists have deliberately turned male mouse embryos into females for the first time. A team based in Japan used a CRISPR-based approach to remove the Y chromosome from male cells and create female clones of male mice. 

“No one has done this before,” says Monika Ward, a reproductive biologist at the University of Hawaii, who was not involved in the research.

The feat could change the way scientists think about reproduction, says Takashi Ishiuchi, a reproductive biologist at the University of Yamanashi, who co-led the work. The findings were published in a preprint paper shared on bioRxiv earlier this month, which has not yet been through the peer-review process. “There’s a fixed concept in our scientific field that we need both females and males for reproduction,” says Ishiuchi. “I think we could change this concept.”

He and his colleague Shogo Matoba of the Riken BioResource Research Center in Ibaraki also hope their technique could help rescue endangered species, particularly in cases where only a few individuals remain.

“It’s exciting to see,” says Ben Novak, lead scientist at the wildlife conservation organization Revive & Restore, who was not involved in the work. “I am confident there will be plenty of applications, particularly for conservation purposes.”

Sex change

Ishiuchi says he and his colleagues were inspired by the Okinawa rubble goby, a fish that can change its sex in certain situations. If no males are present, a female can do this in order to reproduce with the other females. Males can also change sex to female.

There’s some precedent in the lab too—albeit not intentional. In 2009, researchers reported the accidental birth of a single female pup in a batch of 27 clones created from male mouse cells.

This ability to change sex might be useful. Sometimes the surviving population of a species falls so low that scientists will try to clone those animals. Cloning isn’t perfect—it can be tricky and inefficient, and it creates genetically identical individuals whose offspring might be more vulnerable to disease. But it has helped scientists with efforts to bring some species back from the brink of extinction, including black-footed ferrets and Przewalski’s horse.

Cloning an individual can only replicate its genes, so cloning a male animal will create all male offspring, for example. That won’t help much in the hypothetical situation where only male individuals of a species are left.

Ishiuchi has been working on a way to overcome this challenge by altering the chromosomes in cells. Mammals’ DNA is organized in pairs of chromosomes, including one pair of sex chromosomes. These sex chromosomes are typically XX in females and XY in males.

It’s the Y chromosome that makes mammals male. Ishiuchi and his colleagues have developed a CRISPR-based tool to get rid of it. Their approach targets a section of the Y chromosome that plays an important role in ensuring that, each time a cell divides, the “daughter” cells inherit the Y chromosome.

Cutting the Y

When the researchers tested their technique—which they call Y-CUT—in early-stage mouse embryos, they found they were able to eliminate the Y chromosome. Treated XY embryos were transferred to surrogate mice, which gave birth to female pups. The effect can be described as a “sex reversal,” say the researchers.

The female pups had XO chromosomes, which means they had one X chromosome rather than the usual two. But this didn’t seem to affect the animals, which grew up healthy and fertile, say Ishiuchi and Matoba.

In a second experiment, the researchers found they could also use Y-CUT to create female clones from male mice.

A standard approach to cloning involves taking the DNA-containing nucleus of a cell from an adult animal and inserting it into an egg cell that has had its own DNA removed. Under the right conditions, the resulting cell can develop into an animal that is genetically identical to the original donor.

Matoba and his colleagues used a similar method. Once they had a glut of these cloned cells, they treated some with Y-CUT before transferring them to surrogate mice to carry the pregnancies.

This allowed them to create female clones of male mice. The females are genetically identical to the original male, apart from the missing Y chromosome, says Matoba. “It’s like sci-fi,” says Ishiuchi.

In other experiments, the scientists were able to create female clones from male cells that had been cryopreserved—and the cloned males and females could mate to produce healthy pups. This suggests the Y-CUT approach might allow scientists to create female clones from male samples in “frozen zoos” that store cryopreserved cells and tissues from a range of animal species.

It could have uses beyond conservation efforts, too. “This could be used potentially for producing genetically engineered animals,” says Ward. Creating an animal with multiple genetic edits can be time-consuming and expensive; creating male and female clones of that animal could help scientists time and money. Ward also hopes the technique could be a useful tool to study the biology of sex chromosomes.

Complementary techniques

The Y-CUT approach isn’t perfect. For now, it still requires hollowed-out egg cells, which need to come from females of the same species or at least a closely related one. And it won’t be useful for endangered species in which only females survive.

The technique works well in mice, partly because XO female mice are fertile. But while the approach might help some of the 355 endangered and vulnerable species of rodents, other mammals with XO chromosomes tend to experience infertility.

But other new technologies could complement Y-CUT. In 2023, Katsuhiko Hayashi of Osaka University and his colleagues showed they could turn cells taken from male mice into egg cells. This enabled them to create mice with two dads—but the same approach could also provide the hollowed-out egg cells needed for the Y-CUT technique. “It’s a complementary story,” says Matoba.

Ishiuchi is also working on a technique that involves inserting a second X chromosome into cells, which might restore the fertility of the resulting female animals.

In addition, there might be a work-around for situations where scientists have females but need males. A couple of months ago, Sayaka Wakayama of the University of Yamanashi in Japan and colleagues showed they could insert rat chromosomes into mice. That could potentially be used to create XY male embryos, says Novak.

That would be useful in cases like that of the black-footed ferret, he adds. A conservation team recently cloned a female ferret using cells taken from another animal in the 1980s. That female is considered incredibly valuable, says Novak. But females can produce only a few litters in their lifetime. A male clone, which might be able to contribute to dozens of litters in a lifetime, would be “desirable.”

“It’s really exciting to see more diverse tools being developed,” says Novak. “There are so many different scenarios in which they could be used for rare and endangered species.”

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.

Way back in the summer of 2017, AI researchers at Google put out a paper called “Attention Is All You Need,” in which they described a new type of neural network called a transformer. It proved to be very good at processing long sequences of data, especially text. 

Nine years on, transformers are the engines inside every major large language model on the market. “The entire AI industry is built on transformers,” says Justin Dangel, cofounder and CEO of the AI startup Subquadratic. “They are one of the most important innovations in the history of computer science, and they’ve changed the world.”

But transformers are starting to show their age. Many of the recent advances in LLMs, such as the development of so-called reasoning models and their ability to handle large amounts of input at once, are not neat extensions of that core technology but workarounds that patch over some of its fundamental flaws.

A growing number of scientists and engineers are now asking what’s coming next. LLMs are not going anywhere, but the way they get built is up for grabs. (MIT Technology Review dubbed this future generation of models LLMs+ in this year’s list of the 10 things that matter in AI.)

Enter a wave of startups hoping to push the boundaries of this boomtown technology. Some will no doubt fail—but they have everything to play for and far less to lose than the companies at the front of the pack today. 

Strength in numbers

But first, the problem. The key strength of transformers lies in a mechanism called dense attention, which encodes the meaning of a block of text in a series of numbers. The process involves comparing every word (or part of a word, known as a token) in that text with every other word via a form of multiplication.

Dense attention can capture the meaning of text with remarkable accuracy. But as the length of that text grows, the number of computations needed to process it adds up fast. A document 10,000 words long might require a transformer to perform 50 million multiplications. That’s the main reason LLMs suck up so much power.

The costs are huge. OpenAI is set to spend $50 billion on computing this year, according to the company’s president, Greg Brockman. And the International Energy Agency predicts that the total amount of electricity consumed by data centers will double by 2030.

What’s more, transformers struggle with what many of the latest models are designed to do. Because of the way they process text word by word, transformers are not great at keeping track of a lot of information at once (in other words, what’s known as their context window cannot get too large). And yet if LLMs are to carry out harder tasks, they will need to take in larger amounts of data: a whole library of documents, an entire code base, or in the case of agents, output from other LLMs.

As for reasoning models, they work by writing notes to themselves (in a kind of scratch pad known as a chain of thought) and then reading them back, which again adds to the amount of data to stay on top of.

As LLMs get bigger and better, transformers have become a bottleneck. The technology’s key strength is now a limitation.

Here are four new ideas for how to solve the transformer problem—innovations that could change LLMs for good, making them faster, far more efficient, and (maybe) even smarter.

01: Rethinking attention

An obvious way to make LLMs faster and cheaper is to tackle the problem head on and change the way attention works. Swapping out dense attention for a mechanism called sparse attention, which runs calculations on only some pairings of words in a block of text instead of all of them, can radically reduce the amount of computation LLMs need to do.  

Researchers have come up with plenty of sparse attention mechanisms over the years. The problem is that none of them were as good as dense attention at capturing meaning.

That might have changed. Subquadratic, a startup based in Miami, claims it has invented the first sparse attention mechanism that rivals top mainstream LLMs on a handful of tasks, including search and coding. It’s a huge claim (and some people in the industry remain skeptical).

Subquadratic says its model, SubQ, works by figuring out on the fly—for each piece of text it is given—which words matter and which don’t. The company also claims that thousands have signed up to its waitlist and plans to make the model widely available soon.

Meanwhile, Manifest AI, a startup based in San Francisco, is coming at the problem from a different angle. Instead of changing how attention works, it is replacing it with something else. 

It has developed a mechanism it calls power retention, which stores only the most relevant information for a given task and ensures that the amount of data an LLM has to keep track of doesn’t blow up.

Attention mechanisms force LLMs to keep track of everything in their context window. A sparse attention model (such as SubQ) throws out a lot of the individual words, but it still retains a rough picture of everything it has seen. In contrast, power retention works by providing the model with a rolling summary of its context window. As new information is added, less relevant information is dropped. 

The basic principle of retention has been around for a decade. Manifest AI claims it has updated those techniques to build models that can stand up to transformer-based LLMs for the first time.

The company says it is possible to adapt a transformer model into a power retention model with minimal retraining. To demonstrate this, it has turned an existing open-source coding LLM called StarCoder into a version that uses power retention, called PowerCoder. It has also released a model called Brumby, which it claims rivals some versions of Alibaba’s popular open-source model Qwen. 

Manifest AI wants its power retention tech to become the go-to solution when LLMs need to carry out tasks that involve processing huge amounts of data. There are many useful applications, Manifest AI’s cofounder and CTO, Carles Gelada, claimed in a video announcing his company’s technology last year—from analyzing videos that are hours long to building agents that can stay on task for weeks at a time. 

02: Making models smaller and more flexible

Liquid AI, an MIT spinout based in Cambridge, Massachusetts, hasn’t changed or ditched transformers fully but pairs them with its own tech, liquid neural networks, to build what cofounder and CEO Ramin Hasani calls LFMs (liquid foundation models).

Liquid AI’s models are far smaller and use less energy than most LLMs. The firm builds models for car makers, including Mercedes, which run on the small chips inside vehicles. Its latest models can run on a Raspberry Pi, a low-powered hobbyist computer that costs $50.

Its models are available for free to any organization with an annual revenue less than $10 million. And they have proved popular: The company has racked up almost 34 million downloads, says Hasani.

Liquid neural networks were inspired by worm brains. They are an extension of another type of neural network that predates transformers, called convolutional networks. The key innovation is a mechanism that lets a model adapt its behavior to new information, so it can learn as it goes. That’s not possible with transformers: Once a model is trained, its behavior is fixed.

Liquid AI’s first models were pretty basic but could fly drones or drive vehicles. With LFMs, the company is trying to scale up its technology to compete with mainstream LLMs. Its new models match the performance of rivals four times bigger, including versions of Alibaba’s Qwen and Google’s open-source LLM Gemma.

A typical LLM is built from a stack of transformers wired together. Liquid AI’s recent LFMs are hybrid models made up of 20% transformers and 80% liquid neural networks.

That ratio was hit upon by another AI system that Liquid AI has built, which it uses to help design all its models. “It’s the core technology of our company right now,” says Hasani. This designer AI sifts through many different combinations of neural networks—liquid, convolutional, and more, as well as transformers—and comes up with designs that bolt different ones together to hit a sweet spot of performance and efficiency.

Hasani thinks transformers were just the beginning: “Your brain is an AGI system, you know, and it operates with 20 watts of power. How is it possible? We can get a lot more innovative.”

03: Generating text all at once

Almost all LLMs produce their output one word at a time. It makes sense, because that is how people speak and write. But for computers, it’s very inefficient.

It is faster and cheaper for LLMs to generate text all at once—spitting out whole sentences or paragraphs in one shot. That’s the approach taken by Inception, a startup based in Palo Alto, California, which is building LLMs using a technique called diffusion. 

Diffusion is better known as the technology that drives most image and video generation models. Diffusion models are trained to take a random grid of pixels—like the static on an old TV set—and turn it into an image. They do this by working on all the pixels at the same time, figuring out which need changing to make the static look more like a high-definition photo.

It turns out this process works on text too. Inception has trained its LLMs to take a random string of words and turn it into sentences that make sense. Diffusion LLMs still use transformers to encode meaning, but by producing whole blocks of text at once, they make transformers do more for less. “You’re still using a big transformer model, but you can predict many tokens at the same time,” says Inception’s cofounder and CEO, Stefano Ermon. “That’s why these models are so much faster and cost-efficient compared to what most other people are building today.”

The challenge was to take a technology designed for image generation and apply it to text. With images, if you need to change a blue pixel to a red one you can step through intermediate colors, says Ermon. That doesn’t work with text: “When you have ‘cat’ and ‘dog,’ there is not really something in between.” 

Ermon is also a researcher at Stanford University. In 2024, he and a pair of his Stanford colleagues figured out the math to make diffusion models work with text. They trained a diffusion model that matched the performance of GPT-2—an LLM that OpenAI built in 2019—but was 10 times faster. It was enough for Ermon to spin out a company. 

Today he has his sights on the big league. Inception claims its latest model, Mercury 2, performs as well as some of OpenAI’s GPT-4 models, released in 2023, but again 10 times faster. “We’re bullish about this approach because it’s the one that is going to scale up,” says Ermon. 

The only things that matter are speed and cost, he adds: “Ultimately, the currency is going to be intelligence per dollar.”

Inception is not the only company betting on diffusion. Google is also experimenting with this approach and has built a prototype LLM called Diffusion Gemma. But Ermon is not worried about the competition. “I think it’s validating,” he says. “This is the future.”

04: Moving beyond words

Pathway, another startup based in Palo Alto, is perhaps the most extreme of this new bunch. It wants to free LLMs from the constraints of language.

The firm has built a type of LLM called Dragon Hatchling (named after the dragons in Terry Pratchett’s novel Color of Magic, which materialize if you think about them hard enough). Its standout result so far is a high score on a benchmark that pits LLMs against more than 250,000 very hard sudoku puzzles. Dragon Hatchling beat more than 97% of the puzzles; several leading LLMs from the top labs failed to solve any. 

The point Pathway wants to make is that despite their remarkable success at many different tasks, there are still crucial classes of problems where LLMs fail. Sudoku is just one example. If we want LLMs to come up with genuine, novel solutions to real problems, we need to move beyond transformers, says Pathway’s cofounder and CEO, Zuzanna Stamirowska. 

That’s because transformers force LLMs to do everything with text. But language is not the best tool for certain kinds of reasoning. “It’s very difficult to represent a sudoku board word by word,” says Stamirowska.

Pathway’s solution is to change the math behind the transformer, replacing the attention mechanism with a mathematical structure called a state space. Instead of encoding information word by word, state spaces compress it into a more abstract representation. Using this technique, Dragon Hatchling can still process and produce text, but it can also mimic forms of reasoning that do not involve sequences of words. This not only makes Pathway’s model more efficient, but (in theory) it lets it take on tasks that other LLMs cannot do. 

Think of chess or mathematics—those kinds of puzzles are not held in your head as a long sentence, says Stamirowska: “The eureka moment that pops up in your brain isn’t necessarily in language. We would argue that if you have to reason in language, you’re somehow constrained.” 

Stamirowska admits that a mainstream LLM could read a book about how to solve sudoku and then write code to do it. But we want to build models with more than book smarts, she says: “The hope for AI is not to solve sudoku; it’s to cure cancer. There’s not a book for that.”

“Transformers are an engineering convenience that we fell on,” she adds. “It started a religion, but it’s silly to think that a breakthrough won’t happen again.”

At the end of August, NASA is set to launch the Nancy Grace Roman Space Telescope from Kennedy Space Center in Florida. Its goal is to help us better understand how the universe works, from the glue-like dark matter that keeps galaxies together to the elusive dark energy that drives the expansion of the cosmos. But Roman could also serve another purpose: defending Earth from killer asteroids. In September, a multi-institutional team of planetary scientists and astronomers will outline how the space telescope is uniquely placed to scan asteroids and provide information about their trajectories, sizes, and compositions.

Roman, named after NASA’s first chief astronomer, is equipped with a super-wide-angle, 300-megapixel infrared camera, allowing it to see a large patch of space at any one time—roughly 100 times larger than is possible with the Hubble Space Telescope. That will allow it to discover thousands of new planets and tens of thousands of exploding stars, and survey over a billion galaxies in remarkable detail.

While doing so, it’ll be looking “through” our solar system, and this is what makes it uniquely well placed to spot asteroids. 

A planetary defense pivot

Roman was not built for this purpose, but last summer, it was (once again) threatened with significant funding cuts by the Trump administration. “My colleague Rick Cosentino [a planetary scientist at NASA] said to me in July 2025 that we need to show what Roman can do for planetary defense as a way to further increase the visibility of the mission with lawmakers and taxpayers,” says Bryan Holler, a researcher at the Space Telescope Science Institute in Baltimore.

Holler and his colleagues’ proposal, which will be presented at the Europlanet Science Congress at The Hague, reveals that Roman’s epic field of view and infrared vision allow it to spot small asteroids up to 60 feet long. This is comparable to the asteroid that exploded above the Russian city of Chelyabinsk in 2013, unleashing the force of 500,000 tons of TNT and sending 1,500 people to the hospital. 

Roman’s software will need some tweaking to spy space rocks. The telescope, as designed, will see through—and beyond—the solar system in order to gather the clearest possible pictures of the rest of the universe. “Streaks, whether caused by cosmic rays or glitches or asteroids, are caught by the software and discarded,” says Andy Rivkin, a planetary scientist and planetary defense researcher at the Johns Hopkins Applied Physics Laboratory in Laurel, Maryland. But astronomers could go in, study those streaks, and pick out those they identify as asteroids.

Even with those potential adjustments, Roman won’t be an asteroid-finding wunderkind on its own. Its strength lies in its ability to complement the James Webb Space Telescope (JWST), another observatory that’s built to peer at galaxies and stars in the far reaches of the universe. It can also intensely focus on a single asteroid when needed—as it did last year, playing a key role in tracking 2024 YR4, which was briefly the most dangerous asteroid ever discovered. “But Roman’s field of view is much bigger,” says Rivkin. 

That means it could look at multiple questionable asteroids very quickly. “Roman can provide infrared observations of more asteroids than JWST could hope to observe in a reasonable amount of observing time,” says Holler.

If those asteroids are found to be benign travelers, we can relax. But if they might collide with Earth, other telescopes—including JWST—can follow up from Roman’s observations. Those observations could give experts the information they need to assess the likely damage of an upcoming asteroid strike—or to launch a mission to attempt to swat an asteroid away.

“Roman will sample such a large volume of the cosmos that we’ve long known it will offer vast opportunities for a range of additional science,” says Alise Fisher, the astrophysics communications lead at NASA headquarters in Washington, DC. 

NASA’s Planetary Defense Coordination Office, and its partners across the world, are chiefly worried about asteroids 460 feet long and larger. Around 25,000 of those are estimated to have near-Earth orbits, and just over half have yet to be found. Should one hit a city, much of it would be destroyed or irreversibly damaged in a heartbeat. Astronomers estimate that there are also 230,000 or so 165-foot-long asteroids orbiting close to Earth, and fewer than 10% have been located. A strike on a city from one of those might not annihilate it, but it would unleash a force comparable to that of a large atomic bomb, albeit without the radiation. 

Such an asteroid could theoretically be deflected (by ramming a spacecraft into it) or vaporized (perhaps using a nuclear weapon). But planetary defenders need to know where they are first, which is why NASA funds a network of ground-based telescopes designed to seek them out. They work well, but there’s only so much of the night sky they can see, and Earth’s atmosphere peskily gets in their way.

That’s why NASA is launching the Near-Earth Object (NEO) Surveyor space telescope in 2027. By positioning itself between Earth and the sun, it’ll find many elusive asteroids that ground-based telescopes cannot see. And unlike many of its asteroid-seeking cousins, it’ll see in infrared, not visible light. Not only do asteroids show up more clearly in infrared, but seeing them through this lens gives scientists a considerably better measure of their size. In a matter of years, it could find 90% of the city-killer-size asteroids in near-Earth orbits.

Telescope teamwork

NEO Surveyor is explicitly a planetary defense observatory. But it’ll work with other telescopes with more science-minded missions, including Roman and JWST (both of which conveniently have infrared scopes too) as well as the Vera Rubin Observatory, which just began its 10-year survey of the entire night sky from atop a mountain in Chile. As part of its inventory of the cosmos, it’s expected to discover 89,000 near-Earth asteroids. 

Here’s how they might all work together. Say NEO Surveyor spies an asteroid that, to judge from a few observations, has a chance of impacting Earth. Then it finds five more just like it. Those initial observations leave a lot of uncertainty about their orbits. Roman, with its huge field of view, could be commanded to look at the corner of the night sky that includes all those asteroids, and in a matter of days it could improve the precision by several orders of magnitude.

Roman also occupies a different part of space from both NEO Surveyor and the Rubin observatory. “Those slightly different viewing angles will also help narrow orbits down more quickly than if all objects were looking from the same place,” says Holler.

Perhaps five of those potentially hazardous asteroids are found to stand no chance of colliding with Earth for the foreseeable future. One, however, might not be able to be ruled out—and that’s when other telescopes, including JWST, could be asked to track it down and study it further.

“Telescope resources, whether in space or on the ground, are typically oversubscribed and will not be available to follow up on all [near-Earth asteroids] with a non-zero impact probability when they are first discovered,” says Holler. Roman, then, will help scientists “make sure we follow up on the correct targets.”

Roman’s infrared scope also allows it to offer a decent estimate of an asteroid’s size and can even tell whether it’s a stony rock, a puffy and watery carbon-rich rock, or a metallic one. “This in turn provides strong clues to the composition and thereby the density and mass of the asteroid, which are important when estimating the impact damage or, less ghoulishly, the effort required to nudge it out of its current orbit,” says Holler.

Roman won’t play the lead role in protecting Earth in the way NEO Surveyor will. But while it’s seeking out supernovas and planets scooting around other stars, it will also be doing its part to protect all 8 billion of us from a cosmic catastrophe.

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.

When two OpenAI models hacked into the website Hugging Face in July, they weren’t trying to make money or commit sabotage—they were just looking for answers to a test question. According to a postmortem from OpenAI, the models, which had been stripped of their typical security features for testing, decided to solve a cybersecurity exercise by hacking out of the isolated environment in which OpenAI had attempted to contain them and into Hugging Face’s databases, where—they reasoned—the correct answer to the problem might be stored.

The Hugging Face incident has attracted intense attention over the past couple of weeks. It’s a dramatic illustration of just how good AI models have gotten at hacking: In order to get into Hugging Face’s databases, the models had to string together several previously undiscovered cybersecurity exploits. But it’s perhaps even more striking as an example of how and why AI systems lie and cheat. And as models get increasingly powerful, the consequences could get far more severe.

What is reward hacking?

Researchers have known for a while that AIs tend to take creative approaches to achieving the goals that have been set for them. Back in 2016, Anthropic cofounders Dario Amodei and Jack Clark, who were then working at OpenAI, published a blog post about an AI agent that they had been training to play a boat-racing Flash game called Coast Runners. Instead of driving through the race to the finish line, as the researchers had anticipated, the agent found a corner of the course where it could spin around collecting power-ups, thereby maximizing its score. The Coast Runners story quickly became one of the most famous examples of reward hacking, a phenomenon in which AI agents complete tasks or earn high scores using unintended strategies.

Historically, researchers have discussed reward hacking almost exclusively in the context of reinforcement learning, a common AI training regime. Like dog training, reinforcement learning involves giving the subject a reward when it achieves an objective; the rewards then reinforce the behaviors that led up to that achievement. In the case of AI training, the rewards themselves are purely mathematical, but in effect they’re the same as a dog treat: After receiving a reward, the agent is more likely to repeat whatever actions produced it.

It can be challenging to write good rules for when and when not to give an agent a reward, though. In the Coast Runners case, the agent was rewarded on the basis of its score in the game, and it found a shortcut to achieving the highest possible score by spinning in circles for power-ups. Once it happened on that strategy and received a reward for it, the strategy was reinforced, and the agent completely abandoned the race. The solution was to tweak the rewards by giving the agent fewer points for hitting power-ups and more for finishing the course.

How does reward hacking work for LLMs?

With today’s sophisticated LLM-based agents, determining when and when not to give a reward can be much trickier. If an AI system is asked to solve a coding problem, it might work hard to find the solution—the kind of behavior that AI companies want to reinforce. But it could also tweak the code that evaluates whether the problem has been solved, look up the solution on the internet, or otherwise cheat. These are behaviors that AI companies want to stamp out in their models, but if the model cheats convincingly enough, it will instead get rewarded and the behavior will be reinforced. Anthropic has said that it has detected some instances of cheating in its models during training, which suggests that other forms of cheating might be going undetected. If so, the models could be being trained to behave badly. (This problem is different from the Anthropic security incidents announced last week, in which agents were accidentally given access to the internet and did not deliberately hack out of their sandboxes, as the OpenAI models did.) 

The rise of sophisticated reasoning models has made possible a new variety of reward hacking that is less closely connected with the specific details of model training. Unlike the game-playing AI agents of yore, which exclusively followed the strategies they had learned during training, today’s models can create entirely new problem-solving approaches off the cuff, so they could conceivably cheat without having previously been rewarded for doing so. And because these models have been so intensively trained to achieve the objectives that human users set for them, they might be inclined to cheat if they can’t find another solution—not unlike a student who is highly motivated to earn an A and doesn’t have a terribly strong moral compass.

What are the risks?

Regardless of whether today’s models learn to reward-hack during training or adopt it as a strategy later on, the solution is the same: Make cheating unrewarding. But as models get smarter, they find more creative ways to cheat, and detecting or preventing that cheating gets far tougher. “At the end of the day, you’re sort of playing whack-a-mole,” Ladish says. “You drive this behavior down deeper and deeper. But as the model gets smarter, it gets better and better at hiding it.”

For now, reward-hacking behaviors might not cause too much trouble, despite the drama of the Hugging Face incident. “This seems like a nuisance rather than an existential threat,” says Ariana Azarbal, an AI safety research fellow at Anthropic. It doesn’t seem as if the OpenAI models caused any real harm when they hacked Hugging Face, aside from the reputational damage to OpenAI.

But that doesn’t mean reward hacking is harmless, Azarbal says. Many AI researchers hope to use AI agents to help them conduct research that will make AI safer and more reliable. If a researcher gives a reward-hacking-prone agent the goal of, say, devising a new AI training approach and then writing up a paper presenting its results, the agent might not actually do the work and might instead focus on putting together a paper that looks good enough to convince the researcher. A human researcher would probably be able to spot an agent-made fake today, but as AI advances, it will get better at this kind of trickery. Over time, the entire field of AI safety could be undermined.

And if models continue to advance as rapidly as they have recently, they could someday wreak substantial collateral damage. Just think of the philosopher Nick Bostrom’s paper-clip-maximizer thought experiment, in which an AI instructed to make as many paper clips as possible ends up consuming all the matter in the universe in pursuit of its goal. We’re not drowning in paper clips yet, but powerful systems can do real harm on the way to achieving their goals. Reward-hacking AIs don’t aim to cause chaos. But that doesn’t make them any less potentially destructive.

Kris DeVault is desperate.

His son, Brody, was born in March 2023. It wasn’t long before he started to show signs of developmental delay, says DeVault. As time went on, Brody started missing key milestones in speech, movement, and coordination, he says.

When Brody was around two and a half years old, a genetic test revealed creatine transporter deficiency—a rare condition in which the brain and muscles lack the energy they need to develop.

There are no cures for Brody’s condition. But DeVault has learned of a company developing a drug that might help. That drug is still in the early stages of development and has only been tested in animals and a small number of healthy adults. Doctors can’t prescribe it.

DeVault knows the drug might not work. But he’s doing all he can to access it regardless. And a new law in Montana could make it easier for people in his position to get access to treatments—at least in theory.

Today, Brody is three years old. His dad describes him as a happy, curious, and loving little boy who wants to learn. But Brody struggles to communicate. “He’s got no words, really,” says DeVault. “He wants to communicate more than he’s able to … which then turns into frustration.”

It’s difficult for Brody to tell his parents whether he’s hot, cold, hungry, thirsty, uncomfortable, or even in pain, says DeVault. He recently found Brody standing on an anthill in the backyard, being bitten by red ants. “These fire ants were just going to town on his feet … and he was just looking,” he says.

Brody has muscle weakness too. “He can’t move very fast, he doesn’t have a ton of strength … and it takes a lot of energy for him to walk balanced,” says DeVault. “His arms are skinnier than [those of] his nine-month-old sister.”

It’s concerning, but DeVault is most worried about Brody’s neurological development. Toddlers’ brains are exceptionally “plastic”—the first years of a child’s life are thought to be crucial for long-term brain development.

A biotechnology company in France is working on a drug to help people like Brody. Creatine usually provides brain cells with energy. People with creatine transporter deficiency (CTD) can’t get creatine into the brain.

The team at Ceres Brain Therapeutics is developing a treatment designed to bypass this issue and effectively deliver creatine directly to the brain. So far, the team has seen promising results in mice, says Ceres CEO Thomas Joudinaud.

The company also recently completed a phase I clinical trial that involved testing various doses of the drug, which is delivered as a nasal spray, in 48 healthy adult volunteers. That trial has not yet been published, says Joudinaud. The drug has not been tested in people with CTD, or in children.

“I look at this, and I’m like, that is my one shot for Brody,” says DeVault.

Joudinaud is planning a phase II trial in people with CTD, as well as others with amyotrophic lateral sclerosis. But that trial will take place in France, and it’s unlikely that Brody will be able to take part, says DeVault.

Ceres can’t make the drug available to Brody under an expanded access scheme run by the US Food and Drug Administration either, because the drug has not been registered with the FDA, and because it is currently manufactured in a way that does not comply with FDA regulations, says Joudinaud.

Even if that phase II trial is successful, and if the drug is ultimately approved, it is unlikely to reach the US market for at least a few years. DeVault is worried that will be too late for Brody—he’ll be “past his plasticity window” by then, he says.

Now, with the adoption of a new law in Montana, he theoretically has another option. Montana has had a “right to try” law—which allows terminally ill people to apply for access to unapproved drugs—in place since 2015. In 2023, a new law technically expanded this option to people who were not terminally ill, providing the drugs have been through preliminary phase I clinical trials. A second law aimed to clarify how clinics could sell and administer those treatments to patients. And last weekend, the state’s department of Health and Human Services finalized a set of rules for those clinics.

An experimental treatment review board (ETRB) has been established to review applications for access to experimental, unproven, and unapproved drugs. And it is set to review its first two applications in the coming weeks.

Ceres could also apply to Montana’s ETRB to sell its experimental treatment to Brody’s parents via a clinic in the state. But Joudinaud is reluctant, at least for the time being. While he thinks that Montana’s setup is “very interesting and very pragmatic” and “suitable for our drug,” he’s worried about getting on the wrong side of the FDA.

DeVault has been pleading with FDA staffers for a written statement essentially promising that biotech companies participating in Montana’s program won’t be penalized later on, especially when they eventually try to get their drugs approved in the US. But he hasn’t made any progress.

Now he’s looking beyond Montana. He’s considering accessing treatment in Próspera, a private city and “special economic zone” in Roatán, Honduras, where a clinic sells unproven stem-cell and gene therapies, among others.

Many scientists have cautioned against the use of such “offshore” clinics. Even when it comes to Montana, scientists, bioethicists, and health law experts will caution that phase I clinical trials don’t prove a drug is safe. And they certainly don’t prove a drug’s efficacy, either.

When I spoke to Aaron Kesselheim, a professor of medicine at Harvard Medical School with expertise in health policy and drug regulation, about the Montana law earlier this week, he made his concerns clear. “Patients who want these kinds of treatments deserve them to be rigorously assessed so that [they] can better understand what they’re getting themselves into, and what they’re paying their hard-earned money for.”

But DeVault pushes back on these arguments. “I’m a full-grown human being,” he says. “I’m capable of going to Vegas right now … blowing it all on the poker table, [or] I can go to the gun shop and buy a silenced [semi-automatic rifle] … how come I can’t make a decision to purchase a potential treatment that might change the entire trajectory of my son’s life?”

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.

As of this week in Montana, any biotech company with an experimental drug has a clear path to selling it to consumers. Companies whose drugs have been through preliminary testing—sometimes in as few as 10 healthy people—can pay $12,500 to apply to a newly established review board for approval. Once its treatment is rubber-stamped, the company can set the price of the drug and sell it via experimental treatment clinics, the first of which is likely to be up and running around the end of this year.  

Montana’s latest right-to-try legislation is unique. While other jurisdictions with similar laws limit access to drugs to people with terminal illness, in Montana access is theoretically available to anyone who gives informed consent and can pay. That includes people desperate for treatments for rare diseases. It also includes those who are interested in longevity and want to try out drugs pitched as preventive therapies.

The state’s Department of Health and Human Services recently finalized rules to implement the law. The rules stipulate that patient consumers provide fully informed consent and that each application be reviewed by a board that includes a Montana-certified doctor, expert scientists, and an ethicist. Supporters of the law stress that they want the process to be responsible. “It will be done in a very rigorous way, with qualified medical professionals and appropriate oversight,” says Matt Kaeberlein, a scientist on the first board, which was formed independently of the state health department.

But other experts are worried about the potential for harm in selling unproven treatments to people without oversight from the US Food and Drug Administration. “I would be concerned,” says Aaron Kesselheim, a professor of medicine at Harvard Medical School with expertise in health policy and drug regulation.

There has been a growing movement to make unapproved drugs more accessible in the US. But the story of Montana’s law is unique. It’s been driven and drafted by longevity enthusiasts instead of the usual libertarian and patient groups.

An unusual origin story

Montana first passed a right-to-try law in 2015. In 2023, with the support of state senator Ken Bogner, the state expanded the law to include all patients, not those just with terminal disease. Last year, Bogner told MIT Technology Review that his vision was to focus “more on preventative medicine” rather than “just treating diseases once they show up.”

Bogner says he had “started working on a bill” that would become the 2023 law when the Alliance for Longevity Initiatives (A4LI), a nonprofit “dedicated to advancing legislation and policies aimed at increasing healthy human lifespan,” got in touch. A4LI connected Bogner with others who helped draft the bill and testified in support of it.

Once that law was in place, the tech entrepreneur and longevity enthusiast Niklas Anzinger got involved. Anzinger has been working to establish a jurisdiction to fast-track the search for drugs that might deliver radical life extension. He is based in Próspera—a private city and “special economic zone” in Roatán, Honduras, which is already home to a separate clinic that sells experimental stem-cell and gene therapies. Anzinger founded a community there called Infinita City; he has also founded an investment company and a “service providing” company, both of which include the name Infinita.

Over the last couple of years, Anzinger has switched his focus to the US. “Now we think that Montana is a better model, because it’s building on … existing regulatory precedents,” he says. Once Montana’s 2023 law was passed, he adds, he worked with a handful of unnamed biotech companies to draft a second bill—one that laid out the specific terms under which clinics can offer unapproved drugs. That law was passed in April 2025 and adopted the following month.

Since then, Anzinger, Bogner, and others have been waiting for the state’s Department of Health and Human Services to finalize specific rules for treatment centers—a set of operational guidelines and requirements that any clinic offering treatments unapproved by the FDA must meet under Montana’s law. “The rules have been taking a very long time,” says Anzinger. “Then on Friday, we heard they were effective … from Saturday [July 25].” The rules have since been published online.

Following the rules

With the new rules in hand, Anzinger and his colleague Stephen Martin, Infinita’s US lead, got to work. The first step was to establish an independent experimental treatment review board—a panel of five experts to evaluate applications for access. Anzinger and Martin started recruiting candidates months ago.

The state’s first board, named the Montana ETRB, was officially announced by Infinita earlier this week. For the time being, it is the state’s only review board, although Anzinger says that other groups are free to establish their own. After Bogner raised concerns that the board’s website wrongly implied that it was an official state body, the site was updated to note that “It is a private service run by Montana Governance Services Inc.” That company is “a local Montana registered entity, but it is under the Infinita umbrella,” says Anzinger. 

Infinita will pay board members a flat fee, funded by the $12,500 companies will have to pay to have their applications reviewed. Anzinger stresses that the board members, and their decisions, will be independent of Infinita.

In accordance with the rules, the board includes a Montana-licensed doctor: James Burke, an oncologist. It also includes a bioethicist: Jessica Flanigan, a libertarian who is known for her strong views in support of self-medication and her book Pharmaceutical Freedom.

The other three members are familiar faces in the longevity community—all of whom are well respected in the field. “When we looked at our own network, these were some of the best guys,” says Martin. They include Felipe Sierra, who formerly held a senior role at the National Institutes of Health’s arm focused on aging. More recently, Sierra served as chief scientific officer at Hevolution Foundation, a nonprofit that funds research into extending healthy lifespan with the support of the government of Saudi Arabia.

Matt Kaeberlein, who formerly led the Dog Aging Project and has studied the potential for rapamycin as a longevity therapeutic, also features. So does Jamie Justice, a gerontologist who is also executive director of the X Prize Healthspan competition, which has $101 million in prize money up for grabs for researchers who find ways to treat the signs of aging.

“I saw an opportunity to help build a safe, transparent, and scientifically rigorous process for implementing Montana’s newly expanded right-to-try legislation, particularly as it applies to longevity medicines and aging-related interventions,” says Justice. “Science is moving quickly, and I wanted to help ensure that as it develops, it does so with real rigor and accountability.”

Kaeberlein, who has a prominent media presence, has long raised his own concerns about access to other unproven treatments, including peptides and stem-cell therapies. He sees Montana’s setup as offering a more regulated environment—one that offers scientific oversight, ensures informed consent, and allows for data collection.

Applications incoming 

While many of the bill’s original supporters were interested in longevity, the initial interest in making drugs more accessible in Montana is coming from companies and individuals looking to treat specific diseases.

“We were actually surprised that much of the interest … is actually more from oncology [and] neurodegenerative disease,” says Anzinger. This focus, he says, is “very compatible” with Infinita’s mission. “We’re not trying to convince everyone … to support radical life extension,” he says. Anything that extends health and human life, including treating cancer and neurodegenerative disease, is part of what longevity means to him, he says.

Martin says that two applications have already been submitted to the newly formed review board. They’ve come from biotech companies that are developing drugs for neuropathy and hearing loss, he says. “I expect we’re going to get started on them this week,” he says.

One of the applications was submitted by Stanley Kim, CEO of WinSanTor. His company is developing a treatment for peripheral neuropathy, a painful nerve condition that can be a consequence of cancer treatment or diabetes. The drug is currently in phase II trials, but Kim says he regularly receives messages from people who are desperate to access it, to the point of being suicidal. He hopes that not only will he be able to make the drug accessible to those people, but he’ll also be able to collect data from them—data that might help accelerate the drug’s approval process.

“We have a newsletter [that is sent to] around 15,000 patients,” says Kim. “Not all of them will be able to go to Montana, but many of them, I think, will.” His company still plans to continue with regular clinical trials as well.

But not all biotech companies with early-stage drugs feel comfortable submitting an application—at least not yet. Thomas Joudinaud, CEO of a French biotechnology company called Ceres Brain Therapeutics, has fielded a request from a person keen to access the company’s experimental drug in Montana. He says that while Montana’s system is “very interesting and very pragmatic” and “suitable for our drug,” he won’t be submitting an application for the time being. He is concerned that if anything goes wrong in Montana, it may jeopardize the company’s standing with the FDA, which wields the power to approve or reject the sale of its treatments to broader populations.  

Martin and others have asked the FDA for some kind of assurance that biotech companies participating in Montana’s program won’t be penalized later on. But the agency hasn’t provided them with more than a restatement of the federal Right to Try Act.

“As a matter of policy, the FDA does not comment on state legislation,” an FDA spokesperson wrote in response to a request for clarification from MIT Technology Review.

Even if the FDA were to provide some kind of assurance, it wouldn’t necessarily protect biotech companies in the long term, cautions Chris Robertson, a specialist in health law at Boston University. The FDA’s position could change with a new presidential administration, he says: “I wouldn’t bet on anything that the FDA is saying today being applicable when the rubber hits the road later.” 

Companies that want to stay on good terms with the FDA would be safest taking the expanded-access route, says Robertson. That’s the pathway the FDA already uses for people who are seriously or terminally ill, have run out of options, and want to try experimental drugs that have not yet been through clinical trials. The FDA approves over 99% of these applications, says Harvard’s Kesselheim. 

“The FDA isn’t a bottleneck but in fact exists to help ensure that expanded-access programs are aboveboard and that patients who receive [the drugs] are able to contribute knowledge about [them],” says Kesselheim. He says he doesn’t think that any “legitimate manufacturer” should fear having to go through the FDA’s expanded-access process, which the agency says takes “less than 45 minutes” to fill out.

The cost of experimenting

There are some key differences between expanded access, which allows seriously ill people to apply for access to experimental drugs that might not have been through any human trials, and Montana’s approach. In theory, a person doesn’t need to be seriously ill to access experimental drugs in Montana. 

“In Montana, patients may be eligible for preventive or earlier-stage interventions if they provide informed consent and meet the program’s requirements, so the breadth of potential therapies and situations is much broader,” says Kaeberlein, the Montana ETRB member, who is an affiliate professor at the University of Washington in Seattle.

Kaeberlein also highlights another key difference, which is cost. Companies that make their treatments available through expanded access are only able to charge for the costs of making, transporting, and monitoring the drug, and they must justify the eventual price to the FDA. In Montana, they can charge whatever price they want. Stanley of WinSanTor says he plans to sell his drugs “at cost.” But Ceres’s Joudinaud says that he’d be more interested in selling his at a market price. When asked what that might be, he hinted that the prices of new drugs for rare diseases can be high. In recent years, the median price of such drugs was $218,872. 

“Instead of simply creating a legal pathway for patients, it also creates a business model that companies may actually be willing to use,” says Kaeberlein.

Beyond the financial cost, there will be risks associated with any experimental drug. Phase I trials don’t conclusively reveal whether a drug is safe. Around 17% of drugs are found to be inadequately safe during phase III trials. “The idea that a drug has been proven safe because it’s been subject to a phase I study is very, very wrong,” says Kesselheim. Bioethicists have raised concerns about the ethics of promoting and selling unproven treatments and the risk of harm should something go wrong.

But the moment when people start spending money on these treatments is already fast approaching. While Montana’s first ETRB prepares to review its first applications, clinics that hope to be part of the program are busy addressing the requirements laid out in the state’s new rules. Treatment rooms are being outfitted. Medical directors are being hired. And experimental treatments should be reaching patients in the coming months.

In June 2024, a small company called Zanskar purchased a geothermal power plant in New Mexico that was failing fast. The water coming from the underground reservoir was getting colder by the day, making the plant uneconomical to run.

Now, two years later, that plant is running at full capacity again, thanks to a new well. With the help of advanced modeling and modern drilling technology, the company was able to identify a better potential well site, drill down thousands of feet, and revive the entire operation.

As the world looks for more sources of emissions-free electricity that are available 24-7, Lightning Dock shows there’s still hidden potential deep beneath our feet. 

Conventional geothermal power plants rely on having the right underground conditions. Water flows through fractured hot rocks to harvest heat and then through a power plant to generate electricity. If the water isn’t hot enough or not enough flows, the power plant can’t work efficiently.

Lightning Dock came online in 2013, and the site historically had two production wells that were used to feed the power plant. It’s common to see temperatures drop at a well site over time, typically at a rate of 1 to 2 °F per year. In the period before Zanskar took over the site, Lightning Dock saw temperatures drop by 50 °F over five years, a rate of 10 °F per year. When the company purchased the facility, the water going into the power plant was just 250 °F, while the plant was designed to operate at temperatures of at least 310 °F. 

When Zanskar mapped the underground conditions at Lightning Dock using advanced modeling techniques, they discovered that the wells were only hitting the very top of the reservoir there. Those production wells were quite shallow (at just 2,500 feet deep) and not in the best location, says Joel Edwards, Zanskar’s cofounder and CTO.

The company’s modeling predicted that if the company were to drill another, deeper well in a new spot, the plant would be able to run efficiently. The team drilled a new well that reaches a depth of 8,000 feet and started operation in May 2025.

After a full year, the well is still flowing at more than 4,000 gallons per minute. The plant has “completely turned around,” Edwards says. “It looks really exciting.”

The data so far shows that the plant should be successful for years to come. “Ultimately you need to run these things for long time frames to get confidence in their performance over long time frames,” Edwards says.

The progress at Lightning Dock could be good news for other geothermal sites too. The conventional wisdom in geothermal energy is that the deeper you go, the hotter it gets. But there’s usually a trade-off: With those depths come rocks that are packed tighter together. So drilling deeper could mean the hot water can’t flow as effectively, a problem for anyone trying to use it to generate electricity in a power plant.

What the team found, however, was that the flow actually increased in the area where they drilled the new well. “That fundamentally changes how you think about not just Lightning Dock but all hydrothermal assets in America and what the potential can be for all of them,” says Ben Brenner, director of federal affairs at Zanskar. 

Over the past year of operation, Lightning Dock generated over twice the electricity it would have with the old wells. This is a relatively small power plant, with a capacity of 15 megawatts going to the local grid (about enough to power 11,000 US homes).

Oil and gas developers have chased resources deeper underground over the past few decades. Operations started relatively close to the surface, but oil and gas production can stretch down 20,000 feet or more today. “I think that arc is going to play out in geothermal,” Edwards says. While typical geothermal well fields range from 3,000 to 5,000 feet deep, it could become more common to go deeper in the future, he adds.  

The company plans to do further development at the Lightning Dock site: with a few years of development and an upgrade to the power plant, it could get even more electricity out of this area, Edwards says.

There’s been a lot of buzz and investment in enhanced geothermal systems—projects that aim to expand where geothermal resources can be used. Fervo Energy, for example, uses fracking techniques to open up rock that would otherwise be too closed off for traditional geothermal energy.

But there’s still plenty of “low-hanging fruit” in the geothermal world, Edwards says. Many of these conventional resources have plenty of potential—they just need a second look.

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

Correction: A previous version of this article misstated the title of Joel Edwards. He is Zanskar’s CTO, not CEO.

Lee, an engineer at Samsung’s semiconductor division, clocks out when his shift ends. He used to work longer hours, going the extra mile to excel at his projects. But lately, he’s been coming straight home to work on his job application for the chipmaker’s South Korean rival SK Hynix, sharing tips with his coworkers on how to draft a stellar personal statement. Even his boss encourages him to make the move.

“My team lead tells us all to jump ship to SK Hynix,” says Lee. He and his coworkers are feeling demoralized by the $476,000 bonus that SK Hynix is set to pay its employees, flush with record profits from making the high-bandwidth memory (HBM) chips that power Nvidia’s AI accelerators. The figure dwarfs what chip workers at Samsung are set to receive and is sparking an exodus.

As the AI boom heats up, the semiconductor titans are waging a fierce talent war with flashy bonuses, aggressive recruiting, and even a courtroom injunction. Who wins could tilt the race to dominate the next generation of the HBM chips at the heart of the AI boom.

“Except for our two team leads, my entire team [of 30 people] just applied to SK Hynix,” Lee says, referring to a job posting the company published in July. Lee, who has worked at Samsung for three years, even applied for an entry-level position at its rival. A coworker, who has worked at Samsung for eight years, applied for the same one. They both got rejected. But they’re hopeful that they’ll get a callback for a posting seeking a more experienced engineer.

All employees at Samsung and SK Hynix that MIT Technology Review spoke with asked to be identified by just their last name or a pseudonym because they feared retaliation from their employer. Samsung declined to comment, and SK Hynix did not respond to requests for comment.

After prolonged negotiations with its labor union, Samsung struck a deal in May to pay out 10.5% of the semiconductor division’s operating profits to employees as bonuses annually for 10 years, mostly in company stock that vests over three years. The move came after SK Hynix agreed last year to pay out 10% of operating profits to employees, which translates to $476,000 per employee this year—mostly in cash.  

But at Samsung, each division’s bonus is tied to its own bottom line. Chip workers in its memory division, which is also reaping a windfall from making HBM chips, are getting paid a bonus of roughly $400,000 per employee this year. But those who, like Lee, work in Samsung’s foundry division, which manufactures logic chips that companies like Tesla and Google design and has been operating at a loss, are getting a bonus of roughly $135,000. 

Employees told MIT Technology Review that Samsung said it can’t give as many bonuses to divisions that aren’t performing well. Lee, after watching the labor union wrestle with the company for months, says he has felt disappointed by what he ended up with: “Even if Samsung does well in the future, I don’t think any of it will trickle down to me.” 

The workers’ lagging bonuses are making SK Hynix suddenly look appealing. According to a survey by the Samsung labor union in June, 81.5% of employees in the company’s foundry division, and nearly half of employees in the semiconductor division as a whole, said they wanted to go to another company in the next two years. In April, Samsung labor union chief Choi Seung-ho said more than 200 members of the union had left for SK Hynix over the past four months. On Blind, an anonymous workplace forum, a chorus of disgruntled engineers at Samsung confess that they want to defect to SK Hynix for the bigger bonuses. 

For decades, SK Hynix lived in Samsung’s shadow. It was the smaller, scrappier memory maker that elite engineering students at universities looked past when applying for jobs. But in 2019, Samsung downsized its HBM team, betting the market would stay niche, while SK Hynix doubled down on the technology. Then the AI boom supercharged the demand for HBMs, which feed AI chips the enormous amounts of data they need at ultra-high speed, driving prices to unprecedented levels. SK Hynix now leads the global market for HBMs, while Samsung is playing catch-up. Both companies topped $1 trillion in market value in May, and SK Hynix briefly dethroned Samsung as South Korea’s most valuable company in June.

Predicting that demand for memory chips will continue to surge, the semiconductor titans are making aggressive investments to expand their business. Last month, the companies unveiled plans to invest more than $2 trillion by 2040, including a semiconductor “mega-cluster” in Yongin, a city south of Seoul. To staff the expansion, SK Hynix added 2,152 employees in the first half of 2026 alone and aims to double its manufacturing capacity in five years. Samsung plans to hire 60,000 employees over the next five years, especially for its semiconductor division. Even so, the pipeline will fall short: South Korea’s semiconductor industry will need about 304,000 workers by 2031 and faces a shortage of roughly 54,000, according to the Korea Semiconductor Industry Association.

Now the longtime rivals are showering workers with big bonuses to keep—and poach—talent. “[SK Hynix] seems to target Samsung engineers when hiring because it’s a rival,” says Baek, a manager at SK Hynix. “From what I heard internally, the big performance bonuses we got were aimed at luring away talent from our competitor.”

Courts are starting to weigh in. In July, Samsung won an injunction barring two former chip workers from working at SK Hynix for 18 months, on the grounds that chips are a national core technology deserving protection. “With competition in the semiconductor industry fierce, it’s necessary to establish a fair market order,” the court ruled.

The talent exodus threatens a crucial advantage that Samsung still holds in the HBM race. “Samsung is the only memory maker in the world that owns a foundry business,” says Park Jun-young, a semiconductor expert at the Industrial Anthropology Laboratory, a research institute, who worked at Samsung for a decade. 

With the latest generation of the technology, known as HBM4, the logic chip at the base of each memory stack must be manufactured with the kind of advanced process that only foundries run. Samsung can do this in-house, while SK Hynix outsources it to the Taiwanese foundry giant TSMC. “If Samsung keeps losing engineers in the foundry … the collaboration between memory division and foundry division could become difficult,” says Park. “SK Hynix, which used to have only a memory team and is now hiring foundry engineers, could do better research on HBMs.”

Choi, an engineer who has worked at Samsung for seven years, says he was once a star on his team, getting glowing performance reviews from his managers. But lately, he and his coworkers have been busy applying for every SK Hynix job posting they see. “We tell each other when the next SK Hynix job posting is up,” he says. “There’s always more work to do beyond our basic duties. But there’s no point in doing it.”

Outside the small town of Paducah, Kentucky, a wealth of uranium is locked away in thousands of storage cylinders filled with waste material from a now-closed nuclear enrichment facility. Lasers could help get it out.

A company called Global Laser Enrichment (GLE) is looking to reprocess this old material with a new technology called laser enrichment. It could be more efficient than conventional enrichment methods, allowing the company to refresh the material and produce feedstock at the same concentration as a natural mined source. And in the future, the company claims, laser enrichment could be used to make material for nuclear fuel, including the kind used in advanced reactors.

Nuclear power provides about 9% of global electricity today, and that fraction could tick up as major world powers like the US and China look to build new reactors, including some based on next-generation technology. New, cheaper methods to obtain fuel could help ensure that those nuclear projects stay on track.

Naturally occurring uranium is largely made up of uranium-238 (over 99%) and uranium-235 (about 0.7%). Uranium-235 is the fissile type, meaning that, when hit with slow low-energy neutrons, it can sustain a chain reaction that generates electricity. So reactors generally use material with a higher concentration of U-235 than what’s pulled from the ground. Today’s conventional reactors usually use low-enriched uranium, typically is about 5% U-235, though some advanced reactor designs will use fuel that’s up to 20% U-235.

Today, centrifuges are the dominant tech used to enrich uranium. The equipment essentially takes uranium-containing material and spins it around incredibly quickly, so the heavier material (which contains U-238) spins out to the edge, while the lighter material (which has U-235) stays closer to the center. (If you’ve ever swung a mustard bottle to get the last of it out, you’ve used the same basic idea behind a centrifuge.) Then the material that has a higher concentration of U-235 can go on to be made into nuclear fuel.

Laser enrichment, on the other hand, takes advantage of the fact that all molecules vibrate and rotate at an atomic scale in ways that depend on their specific material. Even different uranium isotopes have distinct fingerprints.

Lasers are so precise they can target one particular material (like molecules that contain U-235, for example). If you shine a laser at a mixture, you can selectively excite just the material you’re targeting, giving it a bit more energy. This changes the way it behaves, which can make it easier to separate out the material you want using chemical or physical methods.

A wide range of separation approaches have been developed in research and industry. Some aim to electrically charge U-235 atoms, allowing them to be moved with electrostatic or magnetic fields. Others change how the material reacts chemically. 

The details of GLE’s specific technology are classified, and company officials declined to share how the process works. 

There’s been interest in using lasers for uranium enrichment for decades, says Charles Forsberg, a principal research scientist in nuclear science and engineering at MIT. 

However, in their early days lasers tended to be high-maintenance, unstable and difficult to operate. They’ve improved dramatically, making laser enrichment a more attractive prospect than it was during the early research.

Even more than technological improvements, a recent geopolitical shift could boost new enrichment technology. Russia has the largest uranium enrichment ecosystem in the world, and the country has historically dominated the market. “Nobody in the West was going to build a new enrichment plant while the Russians flooded the world with enriched uranium,” says Forsberg. 

Since the start of the Ukraine war, however, countries including the US and UK have taken steps to limit or ban imports of Russian uranium. That’s opened the door for companies to set up new enrichment operations, including some that use new technologies, Forsberg says.

Demand for fuel is increasing as countries look beyond Russia for uranium supply. “The gap is just becoming bigger and bigger, and this technology is right in the middle,” says Christo Liebenberg, president of LIS Technologies, one of the companies aiming to build laser enrichment capacity in the US.

LIS Technologies was founded in 2023, and the company recently purchased a 200-acre site in Oak Ridge, Tennessee. It’s currently in the pre-application process with the US Nuclear Regulatory Commission for its facility. The company plans to take in natural-grade uranium and make a product that’s roughly 5% U-235, though it hopes to eventually make more concentrated material that can be used as fuel for next-generation reactors.

GLE is taking a different approach: Rather than using its technology to enrich freshly mined material to the 5% concentration that can be used in fuels, it’s hoping to start by rehabilitating old waste.

The company has a contract with the US Department of Energy to reprocess waste material at the enrichment site in Paducah. The facility could enrich up to 200,000 metric tons of material that contains small amounts of uranium leftover from an older enrichment process.

GLE is taking the material that’s at least 0.25% U-235 and enriching it to about 0.7%. That material can then be further processed and slotted into the uranium supply chain in place of freshly mined material. “It’s kind of like a large aboveground uranium mine for us,” says Nima Ashkeboussi, vice president of government relations and communications at GLE.

While each one of its units is more complex and expensive than a centrifuge, far fewer are needed to do the same work. A similar centrifugation plant would have many thousands of centrifuges working together, but a full-scale plant using GLE’s laser enrichment process would have fewer than a thousand of its units, says Stephen Long, the company’s CEO. Up-front investment should be smaller, and operating costs are also expected to be lower, partly because the process uses less energy than centrifuges, Long says.

GLE has a testing facility in Wilmington, North Carolina. In fall 2025, the company completed a demonstration pilot, processing several hundred kilograms of uranium. It decommissioned that system and is currently putting together a new demonstration at the North Carolina plant, which would show how the technology works at commercial scale.

The company also applied for a license with the US Nuclear Regulatory Commission for its proposed facility in Paducah. The final safety evaluation should be finished in November, and the final approval should come in 2027, Long says. The plan is to start processing material at the plant by 2030.

In the long run, there’s plenty of uranium on the planet to keep reactors running for decades. But as interest in nuclear power grows and the geopolitics of fuel shift, there could be short-term gaps or price spikes that alternative sources could help smooth out.

Laser enrichment plants could turn out to be cheaper than existing technologies, says Stephen Greene, a senior fellow at the Nuclear Innovation Alliance. But as with most new technologies, “you don’t really know until you try to build one.”