Sep 18 2025
How to measure the returns on R&D spending

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.

Given the draconian cuts to US federal funding for science, including the administration’s proposal to reduce the 2026 budgets of the National Institutes of Health by 40% and the National Science Foundation by 57%, it’s worth asking some hard-nosed money questions: How much should we be spending on R&D? How much value do we get out of such investments, anyway? To answer that, it’s important to look at both successful returns and investments that went nowhere.

Sure, it’s easy to argue for the importance of spending on science by pointing out that many of today’s most useful technologies had their origins in government-funded R&D. The internet, CRISPR, GPS—the list goes on and on. All true. But this argument ignores all the technologies that received millions in government funding and haven’t gone anywhere—at least not yet. We still don’t have DNA computers or molecular electronics. Never mind the favorite examples cited by contrarian politicians of seemingly silly or frivolous science projects (think shrimp on treadmills).

While cherry-picking success stories help illustrate the glories of innovation and the role of science in creating technologies that have changed our lives, it provides little guidance for how much we should spend in the future—and where the money should go.

A far more useful approach to quantifying the value of R&D is to look at its return on investment (ROI). A favorite metric for stock pickers and PowerPoint-wielding venture capitalists, ROI weighs benefits versus costs. If applied broadly to the nation’s R&D funding, the same kind of thinking could help account for both the big wins and all the money spent on research that never got out of the lab.

The problem is that it’s notoriously difficult to calculate returns for science funding—the payoffs can take years to appear and often take a circuitous route, so the eventual rewards are distant from the original funding. (Who could have predicted Uber as an outcome of GPS? For that matter, who could have predicted that the invention of ultra-precise atomic clocks in the late 1940s and 1950s would eventually make GPS possible?) And forget trying to track the costs of countless failures or apparent dead ends.

But in several recent papers, economists have approached the problem in clever new ways, and though they ask slightly different questions, their conclusions share a bottom line: R&D is, in fact, one of the better long-term investments that the government can make.

This story is part of MIT Technology Review’s “America Undone” series, examining how the foundations of US success in science and innovation are currently under threat. You can read the rest here.

That might not seem very surprising. We’ve long thought that innovation and scientific advances are key to our prosperity. But the new studies provide much-needed details, supplying systematic and rigorous evidence for the impact that R&D funding, including public investment in basic science, has on overall economic growth.

And the magnitude of the benefits is surprising.

Bang for your buck

In “A Calculation of the Social Returns to Innovation,” Benjamin Jones, an economist at Northwestern University, and Lawrence Summers, a Harvard economist and former US Treasury secretary, calculate the effects of the nation’s total R&D spending on gross domestic product and our overall standard of living. They’re taking on the big picture, and it’s ambitious because there are so many variables. But they are able to come up with a convincing range of estimates for the returns, all of them impressive.

On the conservative end of their estimates, says Jones, investing $1 in R&D yields about $5 in returns—defined in this case as additional GDP per person (basically, how much richer we become). Change some of the assumptions—for example, by attempting to account for the value of better medicines and improved health care, which aren’t fully captured in GDP—and you get even larger payoffs.

While the $5 return is at the low end of their estimates, it’s still “a remarkably good investment,” Jones says. “There aren’t many where you put in $1 and get $5 back.”

That’s the return for the nation’s overall R&D funding. But what do we get for government-funded R&D in particular? Andrew Fieldhouse, an economist at Texas A&M, and Karel Mertens at the Federal Reserve Bank of Dallas looked specifically at how changes in public R&D spending affect the total factor productivity (TFP) of businesses. A favorite metric of economists, TFP is driven by new technologies and innovative business know-how—not by adding more workers or machines—and is the main driver of the nation’s prosperity over the long term.

The economists tracked changes in R&D spending at five major US science funding agencies over many decades to see how the shifts eventually affected private-sector productivity. They found that the government was getting a huge bang for its nondefense R&D buck.

The benefits begin kicking in after around five to 10 years and often have a long-lasting impact on the economy. Nondefense public R&D funding has been responsible for 20% to 25% of all private-sector productivity growth in the country since World War II, according to the economists. It’s an astonishing number, given that the government invests relatively little in nondefense R&D. For example, its spending on infrastructure, another contributor to productivity growth, has been far greater over those years.

The large impact of public R&D investments also provides insight into one of America’s most troubling economic mysteries: the slowdown in productivity growth that began in the 1970s, which has roiled the country’s politics as many people face stunted living standards and limited financial prospects. Their research, says Fieldhouse, suggests that as much as a quarter of that slowdown was caused by a decline in public R&D funding that happened roughly over the same time.

After reaching a high of 1.86% of GDP in 1964, federal R&D spending began dropping. Starting in the early 1970s, TFP growth also began to decline, from above 2% a year in the late 1960s to somewhere around 1% since the 1970s (with the exception of a rise during the late 1990s), roughly tracking the spending declines with a lag of a few years.

If in fact the productivity slowdown was at least partially caused by a drop in public R&D spending, it’s evidence that we would be far richer today if we had kept up a higher level of science investment. And it also flags the dangers of today’s proposed cuts. “Based on our research,” says Fieldhouse, “I think it’s unambiguously clear that if you actually slash the budget of the NIH by 40%, if you slash the NSF budget by 50%, there’s going to be a deceleration in US productivity growth over the next seven to 10 years that will be measurable.”

Out of whack

Though the Trump administration’s proposed 2026 budget would slash science budgets to an unusual degree, public funding of R&D has actually been in slow decline for decades. Federal funding of science is at its lowest rate in the last 70 years, accounting for only around 0.6% of GDP.

Even as public funding has dropped, business R&D investments have steadily risen. Today businesses spend far more than the government; in 2023, companies invested about $700 billion in R&D while the US government spent $172 billion, according to data from the NSF’s statistical agency. You might think, Good—let companies do research. It’s more efficient. It’s more focused. Keep the government out of it.

But there is a big problem with that argument. Publicly funded research, it turns out, tends to lead to relatively more productivity growth over time because it skews more toward fundamental science than the applied work typically done by companies.

In a new working paper called “Public R&D Spillovers and Productivity Growth,” Arnaud Dyèvre, an assistant professor at of economics at HEC Paris, documents the broad and often large impacts of so-called knowledge spillovers—the benefits that flow to others from work done by the original research group. Dyèvre found that the spillovers of public-funded R&D have three times more impact on productivity growth across businesses and industries than those from private R&D funding.

The findings are preliminary, and Dyèvre is still updating the research—much of which he did as a postdoc at MIT—but he says it does suggest that the US “is underinvesting in fundamental R&D,” which is heavily funded by the government. “I wouldn’t be able to tell you exactly which percentage of R&D in the US needs to be funded by the government or what percent needs to be funded by the private sector. We need both,” he says. But, he adds, “the empirical evidence” suggests that “we’re out of balance.”

The big question

Getting the balance of funding for fundamental science and applied research right is just one of the big questions that remain around R&D funding. In mid-July, Open Philanthropy and the Alfred P. Sloan Foundation, both nonprofit organizations, jointly announced that they planned to fund a five-year “pop-up journal” that would attempt to answer many of the questions still swirling around how to define and optimize the ROI of research funding.

“There is a lot of evidence consistent with a really high return to R&D, which suggests we should do more of it,” says Matt Clancy, a senior program officer at Open Philanthropy. “But when you ask me how much more, I don’t have a good answer. And when you ask me what types of R&D should get more funding, we don’t have a good answer.”

Pondering such questions should keep innovation economists busy for the next several years. But there is another mystifying piece of the puzzle, says Northwestern’s Jones. If the returns on R&D investments are so high—the kind that most venture capitalists or investors would gladly take—why isn’t the government spending more?

“I think it’s unambiguously clear that if you actually slash the budget of the NIH by 40%, if you slash the NSF budget by 50%, there’s going to be a deceleration in US productivity growth over the next seven to 10 years that will be measurable.”

Jones, who served as a senior economic advisor in the Obama administration, says discussions over R&D budgets in Washington are often “a war of anecdotes.” Science advocates cite the great breakthroughs that resulted from earlier government funding, while budget hawks point to seemingly ludicrous projects or spectacular failures. Both have plenty of ammunition. “People go back and forth,” says Jones, “and it doesn’t really lead to anywhere.”

The policy gridlock is rooted in in the very nature of fundamental research. Today’s science will lead to great advances. And there will be countless failures; a lot of money will be wasted on fruitless experiments. The problem, of course, is that when you’re deciding to fund new projects, it’s impossible to predict which the outcome will be, even in the case of odd, seemingly silly science. Guessing just what research will or will not lead to the next great breakthrough is a fool’s errand.

Take the cuts in the administration’s proposed fiscal 2026 budget for the NSF, a leading funder of basic science. The administration’s summary begins with the assertion that its NSF budget “is prioritizing investments that complement private-sector R&D and offer strong potential to drive economic growth and strengthen U.S. technological leadership.” So far, so good. It cites the government’s commitment to AI and quantum information science. But dig deeper and you will see the contradictions in the numbers.

Not only is NSF’s overall budget cut by 57%, but funding for physical sciences like chemistry and materials research—fields critical to advancing AI and quantum computers—has also been blown apart. Funding for the NSF’s mathematical and physical sciences program was reduced by 67%. The directorate for computer and information science and engineering fared little better; its research funding was cut by 66%.

There is a great deal of hope among many in the science community that Congress, when it passes the actual 2026 budget, will at least partially reverse these cuts. We’ll see. But even if it does, why attack R&D funding in the first place? It’s impossible to answer that without plunging into the messy depths of today’s chaotic politics. And it is equally hard to know whether the recent evidence gathered by academic economists on the strong returns to R&D investments will matter when it comes to partisan policymaking.

But at least those defending the value of public funding now have a far more productive way to make their argument, rather than simply touting past breakthroughs. Even for fiscal hawks and those pronouncing concerns about budget deficits, the recent work provides a compelling and simple conclusion: More public funding for basic science is a sound investment that makes us more prosperous.

Ecommerce MGMT 0 Comments
Sep 18 2025
AI-designed viruses are here and already killing bacteria

Artificial intelligence can draw cat pictures and write emails. Now the same technology can compose a working genome.

A research team in California says it used AI to propose new genetic codes for viruses—and managed to get several of these viruses to replicate and kill bacteria.

The scientists, based at Stanford University and the nonprofit Arc Institute, both in Palo Alto, say the germs with AI-written DNA represent the “the first generative design of complete genomes.”

The work, described in a preprint paper, has the potential to create new treatments and accelerate research into artificially engineered cells. It is also an “impressive first step” toward AI-designed life forms, says Jef Boeke, a biologist at NYU Langone Health, who was provided an advance copy of the paper by MIT Technology Review.  

Boeke says the AI’s performance was surprisingly good and that its ideas were unexpected. “They saw viruses with new genes, with truncated genes, and even different gene orders and arrangements,” he says.

This is not yet AI-designed life, however. That’s because viruses are not alive. They’re more like renegade bits of genetic code with relatively puny, simple genomes. 

In the new work, researchers at the Arc Institute sought to develop variants of a bacteriophage—a virus that infects bacteria—called phiX174, which has only 11 genes and about 5,000 DNA letters.

To do so, they used two versions of an AI called Evo, which works on the same principles as large language models like ChatGPT. Instead of feeding them textbooks and blog posts to learn from, the scientists trained the models on the genomes of about 2 million other bacteriophage viruses.

But would the genomes proposed by the AI make any sense? To find out, the California researchers chemically printed 302 of the genome designs as DNA strands and then mixed those with E. coli bacteria.

That led to a profound “AI is here” moment when, one night, the scientists saw plaques of dead bacteria in their petri dishes. They later took microscope pictures of the tiny viral particles, which look like fuzzy dots.

“That was pretty striking, just actually seeing, like, this AI-generated sphere,” says Brian Hie, who leads the lab at the Arc Institute where the work was carried out.

Overall, 16 of the 302 designs ended up working—that is, the computer-designed phage started to replicate, eventually bursting through the bacteria and killing them.

J. Craig Venter, who created some of the first organisms with lab-made DNA nearly two decades ago, says the AI methods look to him like “just a faster version of trial-and-error experiments.”

For instance, when a team he led managed to create a bacterium with a lab-printed genome in 2008, it was after a long hit-or-miss process of testing out different genes. “We did the manual AI version—combing through the literature, taking what was known,” he says. 

But speed is exactly why people are betting AI will transform biology. The new methods already claimed a Nobel Prize in 2024 for predicting protein shapes. And investors are staking billions that AI can find new drugs. This week a Boston company, Lila, raised $235 million to build automated labs run by artificial intelligence.

Computer-designed viruses could also find commercial uses. For instance, doctors have sometimes tried “phage therapy” to treat patients with serious bacterial infections. Similar tests are underway to cure cabbage of black rot, also caused by bacteria.

“There is definitely a lot of potential for this technology,” says Samuel King, the student who spearheaded the project in Hei’s lab. He notes that most gene therapy uses viruses to shuttle genes into patients’ bodies, and AI might develop more effective ones.

The Stanford researchers say they purposely haven’t taught their AI about viruses that can infect people. But this type of technology does create the risk that other scientists—out of curiosity, good intentions, or malice—could turn the methods on human pathogens, exploring new dimensions of lethality.

“One area where I urge extreme caution is any viral enhancement research, especially when it’s random so you don’t know what you are getting,” says Venter. “If someone did this with smallpox or anthrax, I would have grave concerns.”

Whether an AI can generate a bona fide genome for a larger organism remains an open question. For instance, E. coli has about a thousand times more DNA code than phiX174 does. “The complexity would rocket from staggering to … way way more than the number of subatomic particles in the universe,” says Boeke.

Also, there’s still no easy way to test AI designs for larger genomes. While some viruses can “boot up” from just a DNA strand, that’s not the case with a bacterium, a mammoth, or a human. Scientists would instead have to gradually change an existing cell with genetic engineering—a still laborious process.

Despite that, Jason Kelly, the CEO of Ginkgo Bioworks, a cell-engineering company in Boston, says exactly such an effort is needed. He believes it could be carried out in “automated” laboratories where genomes get proposed and tested and the results are fed back to AI for further improvement.

 “This would be a nation-scale scientific milestone, as cells are the building blocks of all life,” says Kelly. “The US should make sure we get to it first.”

Ecommerce MGMT 0 Comments
Sep 17 2025
The looming crackdown on AI companionship

As long as there has been AI, there have been people sounding alarms about what it might do to us: rogue superintelligence, mass unemployment, or environmental ruin from data center sprawl. But this week showed that another threat entirely—that of kids forming unhealthy bonds with AI—is the one pulling AI safety out of the academic fringe and into regulators’ crosshairs.

This has been bubbling for a while. Two high-profile lawsuits filed in the last year, against Character.AI and OpenAI, allege that companion-like behavior in their models contributed to the suicides of two teenagers. A study by US nonprofit Common Sense Media, published in July, found that 72% of teenagers have used AI for companionship. Stories in reputable outlets about “AI psychosis” have highlighted how endless conversations with chatbots can lead people down delusional spirals.

It’s hard to overstate the impact of these stories. To the public, they are proof that AI is not merely imperfect, but a technology that’s more harmful than helpful. If you doubted that this outrage would be taken seriously by regulators and companies, three things happened this week that might change your mind.

A California law passes the legislature

On Thursday, the California state legislature passed a first-of-its-kind bill. It would require AI companies to include reminders for users they know to be minors that responses are AI generated. Companies would also need to have a protocol for addressing suicide and self-harm and provide annual reports on instances of suicidal ideation in users’ conversations with their chatbots. It was led by Democratic state senator Steve Padilla, passed with heavy bipartisan support, and now awaits Governor Gavin Newsom’s signature. 

There are reasons to be skeptical of the bill’s impact. It doesn’t specify efforts companies should take to identify which users are minors, and lots of AI companies already include referrals to crisis providers when someone is talking about suicide. (In the case of Adam Raine, one of the teenagers whose survivors are suing, his conversations with ChatGPT before his death included this type of information, but the chatbot allegedly went on to give advice related to suicide anyway.)

Still, it is undoubtedly the most significant of the efforts to rein in companion-like behaviors in AI models, which are in the works in other states too. If the bill becomes law, it would strike a blow to the position OpenAI has taken, which is that “America leads best with clear, nationwide rules, not a patchwork of state or local regulations,” as the company’s chief global affairs officer, Chris Lehane, wrote on LinkedIn last week.

The Federal Trade Commission takes aim

The very same day, the Federal Trade Commission announced an inquiry into seven companies, seeking information about how they develop companion-like characters, monetize engagement, measure and test the impact of their chatbots, and more. The companies are Google, Instagram, Meta, OpenAI, Snap, X, and Character Technologies, the maker of Character.AI.

The White House now wields immense, and potentially illegal, political influence over the agency. In March, President Trump fired its lone Democratic commissioner, Rebecca Slaughter. In July, a federal judge ruled that firing illegal, but last week the US Supreme Court temporarily permitted the firing.

“Protecting kids online is a top priority for the Trump-Vance FTC, and so is fostering innovation in critical sectors of our economy,” said FTC chairman Andrew Ferguson in a press release about the inquiry. 

Right now, it’s just that—an inquiry—but the process might (depending on how public the FTC makes its findings) reveal the inner workings of how the companies build their AI companions to keep users coming back again and again. 

Sam Altman on suicide cases

Also on the same day (a busy day for AI news), Tucker Carlson published an hour-long interview with OpenAI’s CEO, Sam Altman. It covers a lot of ground—Altman’s battle with Elon Musk, OpenAI’s military customers, conspiracy theories about the death of a former employee—but it also includes the most candid comments Altman’s made so far about the cases of suicide following conversations with AI. 

Altman talked about “the tension between user freedom and privacy and protecting vulnerable users” in cases like these. But then he offered up something I hadn’t heard before.

“I think it’d be very reasonable for us to say that in cases of young people talking about suicide seriously, where we cannot get in touch with parents, we do call the authorities,” he said. “That would be a change.”

So where does all this go next? For now, it’s clear that—at least in the case of children harmed by AI companionship—companies’ familiar playbook won’t hold. They can no longer deflect responsibility by leaning on privacy, personalization, or “user choice.” Pressure to take a harder line is mounting from state laws, regulators, and an outraged public.

But what will that look like? Politically, the left and right are now paying attention to AI’s harm to children, but their solutions differ. On the right, the proposed solution aligns with the wave of internet age-verification laws that have now been passed in over 20 states. These are meant to shield kids from adult content while defending “family values.” On the left, it’s the revival of stalled ambitions to hold Big Tech accountable through antitrust and consumer-protection powers. 

Consensus on the problem is easier than agreement on the cure. As it stands, it looks likely we’ll end up with exactly the patchwork of state and local regulations that OpenAI (and plenty of others) have lobbied against. 

For now, it’s down to companies to decide where to draw the lines. They’re having to decide things like: Should chatbots cut off conversations when users spiral toward self-harm, or would that leave some people worse off? Should they be licensed and regulated like therapists, or treated as entertainment products with warnings? The uncertainty stems from a basic contradiction: Companies have built chatbots to act like caring humans, but they’ve postponed developing the standards and accountability we demand of real caregivers. The clock is now running out.

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

Ecommerce MGMT 0 Comments
Sep 13 2025
How do AI models generate videos?

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.

It’s been a big year for video generation. In the last nine months OpenAI made Sora public, Google DeepMind launched Veo 3, the video startup Runway launched Gen-4. All can produce video clips that are (almost) impossible to distinguish from actual filmed footage or CGI animation. This year also saw Netflix debut an AI visual effect in its show The Eternaut, the first time video generation has been used to make mass-market TV.

Sure, the clips you see in demo reels are cherry-picked to showcase a company’s models at the top of their game. But with the technology in the hands of more users than ever before—Sora and Veo 3 are available in the ChatGPT and Gemini apps for paying subscribers—even the most casual filmmaker can now knock out something remarkable. 

The downside is that creators are competing with AI slop, and social media feeds are filling up with faked news footage. Video generation also uses up a huge amount of energy, many times more than text or image generation. 

With AI-generated videos everywhere, let’s take a moment to talk about the tech that makes them work.

How do you generate a video?

Let’s assume you’re a casual user. There are now a range of high-end tools that allow pro video makers to insert video generation models into their workflows. But most people will use this technology in an app or via a website. You know the drill: “Hey, Gemini, make me a video of a unicorn eating spaghetti. Now make its horn take off like a rocket.” What you get back will be hit or miss, and you’ll typically need to ask the model to take another pass or 10 before you get more or less what you wanted. 

So what’s going on under the hood? Why is it hit or miss—and why does it take so much energy? The latest wave of video generation models are what’s known as latent diffusion transformers. Yes, that’s quite a mouthful. Let’s unpack each part in turn, starting with diffusion. 

What’s a diffusion model?

Imagine taking an image and adding a random spattering of pixels to it. Take that pixel-spattered image and spatter it again and then again. Do that enough times and you will have turned the initial image into a random mess of pixels, like static on an old TV set. 

A diffusion model is a neural network trained to reverse that process, turning random static into images. During training, it gets shown millions of images in various stages of pixelation. It learns how those images change each time new pixels are thrown at them and, thus, how to undo those changes. 

The upshot is that when you ask a diffusion model to generate an image, it will start off with a random mess of pixels and step by step turn that mess into an image that is more or less similar to images in its training set. 

But you don’t want any image—you want the image you specified, typically with a text prompt. And so the diffusion model is paired with a second model—such as a large language model (LLM) trained to match images with text descriptions—that guides each step of the cleanup process, pushing the diffusion model toward images that the large language model considers a good match to the prompt. 

An aside: This LLM isn’t pulling the links between text and images out of thin air. Most text-to-image and text-to-video models today are trained on large data sets that contain billions of pairings of text and images or text and video scraped from the internet (a practice many creators are very unhappy about). This means that what you get from such models is a distillation of the world as it’s represented online, distorted by prejudice (and pornography).

It’s easiest to imagine diffusion models working with images. But the technique can be used with many kinds of data, including audio and video. To generate movie clips, a diffusion model must clean up sequences of images—the consecutive frames of a video—instead of just one image. 

What’s a latent diffusion model? 

All this takes a huge amount of compute (read: energy). That’s why most diffusion models used for video generation use a technique called latent diffusion. Instead of processing raw data—the millions of pixels in each video frame—the model works in what’s known as a latent space, in which the video frames (and text prompt) are compressed into a mathematical code that captures just the essential features of the data and throws out the rest. 

A similar thing happens whenever you stream a video over the internet: A video is sent from a server to your screen in a compressed format to make it get to you faster, and when it arrives, your computer or TV will convert it back into a watchable video. 

And so the final step is to decompress what the latent diffusion process has come up with. Once the compressed frames of random static have been turned into the compressed frames of a video that the LLM guide considers a good match for the user’s prompt, the compressed video gets converted into something you can watch.  

With latent diffusion, the diffusion process works more or less the way it would for an image. The difference is that the pixelated video frames are now mathematical encodings of those frames rather than the frames themselves. This makes latent diffusion far more efficient than a typical diffusion model. (Even so, video generation still uses more energy than image or text generation. There’s just an eye-popping amount of computation involved.) 

What’s a latent diffusion transformer?

Still with me? There’s one more piece to the puzzle—and that’s how to make sure the diffusion process produces a sequence of frames that are consistent, maintaining objects and lighting and so on from one frame to the next. OpenAI did this with Sora by combining its diffusion model with another kind of model called a transformer. This has now become standard in generative video. 

Transformers are great at processing long sequences of data, like words. That has made them the special sauce inside large language models such as OpenAI’s GPT-5 and Google DeepMind’s Gemini, which can generate long sequences of words that make sense, maintaining consistency across many dozens of sentences. 

But videos are not made of words. Instead, videos get cut into chunks that can be treated as if they were. The approach that OpenAI came up with was to dice videos up across both space and time. “It’s like if you were to have a stack of all the video frames and you cut little cubes from it,” says Tim Brooks, a lead researcher on Sora.

A selection of videos generated with Veo 3 and Midjourney. The clips have been enhanced in postproduction with Topaz, an AI video-editing tool. Credit: VaigueMan

Using transformers alongside diffusion models brings several advantages. Because they are designed to process sequences of data, transformers also help the diffusion model maintain consistency across frames as it generates them. This makes it possible to produce videos in which objects don’t pop in and out of existence, for example. 

And because the videos are diced up, their size and orientation do not matter. This means that the latest wave of video generation models can be trained on a wide range of example videos, from short vertical clips shot with a phone to wide-screen cinematic films. The greater variety of training data has made video generation far better than it was just two years ago. It also means that video generation models can now be asked to produce videos in a variety of formats. 

What about the audio? 

A big advance with Veo 3 is that it generates video with audio, from lip-synched dialogue to sound effects to background noise. That’s a first for video generation models. As Google DeepMind CEO Demis Hassabis put it at this year’s Google I/O: “We’re emerging from the silent era of video generation.” 

The challenge was to find a way to line up video and audio data so that the diffusion process would work on both at the same time. Google DeepMind’s breakthrough was a new way to compress audio and video into a single piece of data inside the diffusion model. When Veo 3 generates a video, its diffusion model produces audio and video together in a lockstep process, ensuring that the sound and images are synched.  

You said that diffusion models can generate different kinds of data. Is this how LLMs work too? 

No—or at least not yet. Diffusion models are most often used to generate images, video, and audio. Large language models—which generate text (including computer code)—are built using transformers. But the lines are blurring. We’ve seen how transformers are now being combined with diffusion models to generate videos. And this summer Google DeepMind revealed that it was building an experimental large language model that used a diffusion model instead of a transformer to generate text. 

Here’s where things start to get confusing: Though video generation (which uses diffusion models) consumes a lot of energy, diffusion models themselves are in fact more efficient than transformers. Thus, by using a diffusion model instead of a transformer to generate text, Google DeepMind’s new LLM could be a lot more efficient than existing LLMs. Expect to see more from diffusion models in the near future!

Ecommerce MGMT 0 Comments
Sep 12 2025
Texas banned lab-grown meat. What’s next for the industry?

Last week, a legal battle over lab-grown meat kicked off in Texas. On September 1, a two-year ban on the technology went into effect across the state; the following day, two companies filed a lawsuit against state officials.

The two companies, Wildtype Foods and Upside Foods, are part of a growing industry that aims to bring new types of food to people’s plates. These products, often called cultivated meat by the industry, take live animal cells and grow them in the lab to make food products without the need to slaughter animals.

Texas joins six other US states and the country of Italy in banning these products. These legal challenges are adding barriers to an industry that’s still in its infancy and already faces plenty of challenges before it can reach consumers in a meaningful way.

The agriculture sector makes up a hefty chunk of global greenhouse-gas emissions, with livestock alone accounting for somewhere between 10% and 20% of climate pollution. Alternative meat products, including those grown in a lab, could help cut the greenhouse gases from agriculture.

The industry is still in its early days, though. In the US, just a handful of companies can legally sell products including cultivated chicken, pork fat, and salmon. Australia, Singapore, and Israel also allow a few companies to sell within their borders.

Upside Foods, which makes cultivated chicken, was one of the first to receive the legal go-ahead to sell its products in the US, in 2022. Wildtype Foods, one of the latest additions to the US market, was able to start selling its cultivated salmon in June.

Upside, Wildtype, and other cultivated-meat companies are still working to scale up production. Products are generally available at pop-up events or on special menus at high-end restaurants. (I visited San Francisco to try Upside’s cultivated chicken at a Michelin-starred restaurant a few years ago.)

Until recently, the only place you could reliably find lab-grown meat in Texas was a sushi restaurant in Austin. Otoko featured Wildtype’s cultivated salmon on a special tasting menu starting in July. (The chef told local publication Culture Map Austin that the cultivated fish tastes like wild salmon, and it was included in a dish with grilled yellowtail to showcase it side-by-side with another type of fish.)

The as-yet-limited reach of lab-grown meat didn’t stop state officials from moving to ban the technology, effective from now until September 2027.

The office of state senator Charles Perry, the author of the bill, didn’t respond to requests for comment. Neither did the Texas and Southwestern Cattle Raisers Association, whose president, Carl Ray Polk Jr., testified in support of the bill in a March committee hearing.

“The introduction of lab-grown meat could disrupt traditional livestock markets, affecting rural communities and family farms,” Perry said during the meeting.

In an interview with the Texas Tribune, Polk said the two-year moratorium would help the industry put checks and balances in place before the products could be sold. He also expressed concern about how clearly cultivated-meat companies will be labeling their products.

“The purpose of these bans is to try to kill the cultivated-meat industry before it gets off the ground,” said Myra Pasek, general counsel of Upside Foods, via email. The company is working to scale up its manufacturing and get the product on the market, she says, “but that can’t happen if we’re not allowed to compete in the marketplace.”

Others in the industry have similar worries. “Moratoriums on sale like this not only deny Texans new choices and economic growth, but they also send chilling signals to researchers and entrepreneurs across the country,” said Pepin Andrew Tuma, the vice president of policy and government relations for the Good Food Institute, a nonprofit think tank focused on alternative proteins, in a statement. (The group isn’t involved in the lawsuit.) 

One day after the moratorium took effect on September 1, Wildtype Foods and Upside Foods filed a lawsuit challenging the ban, naming Jennifer Shuford, commissioner of the Texas Department of State Health Services, among other state officials.

A lawsuit wasn’t necessarily part of the scale-up plan. “This was really a last resort for us,” says Justin Kolbeck, cofounder and CEO of Wildtype.

Growing cells to make meat in the lab isn’t easy—some companies have spent a decade or more trying to make significant amounts of a product that people want to eat. These legal battles certainly aren’t going to help. 

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

Ecommerce MGMT 0 Comments
Sep 12 2025
We can’t “make American children healthy again” without tackling the gun crisis

Note for readers: This newsletter discusses gun violence, a raw and tragic issue in America. It was already in progress on Wednesday when a school shooting occurred at Evergreen High School in Colorado and Charlie Kirk was shot and killed at Utah Valley University. 

Earlier this week, the Trump administration’s Make America Healthy Again movement released a strategy for improving the health and well-being of American children. The report was titled—you guessed it—Make Our Children Healthy Again.

Robert F. Kennedy Jr., who leads the Department of Health and Human Services, and his colleagues are focusing on four key aspects of child health: diet, exercise, chemical exposure, and overmedicalization.

Anyone who’s been listening to RFK Jr. posturing on health and wellness won’t be surprised by these priorities. And the first two are pretty obvious. On the whole, American children should be eating more healthily. And they should be getting more exercise.

But there’s a glaring omission. The leading cause of death for American children and teenagers isn’t ultraprocessed food or exposure to some chemical. It’s gun violence. 

Yesterday’s news of yet more high-profile shootings at schools in the US throws this disconnect into even sharper relief. Experts believe it is time to treat gun violence in the US as what it is: a public health crisis.

I live in London, UK, with my husband and two young children. We don’t live in a particularly fancy part of the city—in one recent ranking of London boroughs from most to least posh, ours came in at 30th out of 33. I do worry about crime. But I don’t worry about gun violence.

That changed when I temporarily moved my family to the US a couple of years ago. We rented the ground-floor apartment of a lovely home in Cambridge, Massachusetts—a beautiful area with good schools, pastel-colored houses, and fluffy rabbits hopping about. It wasn’t until after we’d moved in that my landlord told me he had guns in the basement.

My daughter joined the kindergarten of a local school that specialized in music, and we took her younger sister along to watch the kids sing songs about friendship. It was all so heartwarming—until we noticed the school security officer at the entrance carrying a gun.

Later in the year, I received an email alert from the superintendent of the Cambridge Public Schools. “At approximately 1:45 this afternoon, a Cambridge Police Department Youth Officer assigned to Cambridge Rindge and Latin School accidentally discharged their firearm while using a staff bathroom inside the school,” the message began. “The school day was not disrupted.”

These experiences, among others, truly brought home to me the cultural differences over firearms between the US and the UK (along with most other countries). For the first time, I worried about my children’s exposure to them. I banned my children from accessing parts of the house. I felt guilty that my four-year-old had to learn what to do if a gunman entered her school. 

But it’s the statistics that are the most upsetting.

In 2023, 46,728 people died from gun violence in the US, according to a report published in June by the Johns Hopkins Bloomberg School of Public Health. That includes both homicides and suicides, and it breaks down to 128 deaths per day, on average. The majority of those who die from gun violence are adults. But the figures for children are sickening, too. In 2023, 2,566 young people died from gun violence. Of those, 234 were under the age of 10.

Gun death rates among children have more than doubled since 2013. Firearms are involved in more child deaths than cancer or car crashes.

Many other children survive gun violence with nonfatal—but often life-changing—injuries. And the impacts are felt beyond those who are physically injured. Witnessing gun violence or hearing gunshots can understandably cause fear, sadness, and distress.  

That’s worth bearing in mind when you consider that there have been 434 school shootings in the US since Columbine in 1999. The Washington Post estimates that 397,000 students have experienced gun violence at school in that period. Another school shooting took place at Evergreen High School in Colorado on Wednesday, adding to that total.

“Being indirectly exposed to gun violence takes its toll on our mental health and children’s ability to learn,” says Daniel Webster, Bloomberg Professor of American Health at the Johns Hopkins Center for Gun Violence Solutions in Baltimore.

The MAHA report states that “American youth face a mental health crisis,” going on to note that “suicide deaths among 10- to 24-year-olds increased by 62% from 2007 to 2021” and that “suicide is now the leading cause of death in teens aged 15-19.” What it doesn’t say is that around half of these suicides involve guns.

“When you add all these dimensions, [gun violence is] a very huge public health problem,” says Webster.

Researchers who study gun violence have been saying the same thing for years. And in 2024, then US Surgeon General Vivek Murthy declared it a public health crisis. “We don’t have to subject our children to the ongoing horror of firearm violence in America,” Murthy said in a statement at the time. Instead, he argued, we should tackle the problem using a public health approach.

Part of that approach involves identifying who is at the greatest risk and offering support to lower that risk, says Webster. Young men who live in poor communities tend to have the highest risk of gun violence, he says, as do those who experience crisis or turmoil. Trying to mediate conflicts or limit access to firearms, even temporarily, can help lower the incidence of gun violence, he says.

There’s an element of social contagion, too, adds Webster. Shooting begets more shooting. He likens it to the outbreak of an infectious disease. “When more people get vaccinated … infection rates go down,” he says. “Almost exactly the same thing happens with gun violence.”

But existing efforts are already under threat. The Trump administration has eliminated hundreds of millions of dollars in grants for organizations working to reduce gun violence.

Webster thinks the MAHA report has “missed the mark” when it comes to the health and well-being of children in the US. “This document is almost the polar opposite to how many people in public health think,” he says. “We have to acknowledge that injuries and deaths from firearms are a big threat to the health and safety of children and adolescents.”

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.

Ecommerce MGMT 0 Comments
Sep 10 2025
Help! My therapist is secretly using ChatGPT

In Silicon Valley’s imagined future, AI models are so empathetic that we’ll use them as therapists. They’ll provide mental-health care for millions, unimpeded by the pesky requirements for human counselors, like the need for graduate degrees, malpractice insurance, and sleep. Down here on Earth, something very different has been happening. 

Last week, we published a story about people finding out that their therapists were secretly using ChatGPT during sessions. In some cases it wasn’t subtle; one therapist accidentally shared his screen during a virtual appointment, allowing the patient to see his own private thoughts being typed into ChatGPT in real time. The model then suggested responses that his therapist parroted. 

It’s my favorite AI story as of late, probably because it captures so well the chaos that can unfold when people actually use AI the way tech companies have all but told them to.

As the writer of the story, Laurie Clarke, points out, it’s not a total pipe dream that AI could be therapeutically useful. Early this year, I wrote about the first clinical trial of an AI bot built specifically for therapy. The results were promising! But the secretive use by therapists of AI models that are not vetted for mental health is something very different. I had a conversation with Clarke to hear more about what she found. 

I have to say, I was really fascinated that people called out their therapists after finding out they were covertly using AI. How did you interpret the reactions of these therapists? Were they trying to hide it?

In all the cases mentioned in the piece, the therapist hadn’t provided prior disclosure of how they were using AI to their patients. So whether or not they were explicitly trying to conceal it, that’s how it ended up looking when it was discovered. I think for this reason, one of my main takeaways from writing the piece was that therapists should absolutely disclose when they’re going to use AI and how (if they plan to use it). If they don’t, it raises all these really uncomfortable questions for patients when it’s uncovered and risks irrevocably damaging the trust that’s been built.

In the examples you’ve come across, are therapists turning to AI simply as a time-saver? Or do they think AI models can genuinely give them a new perspective on what’s bothering someone?

Some see AI as a potential time-saver. I heard from a few therapists that notes are the bane of their lives. So I think there is some interest in AI-powered tools that can support this. Most I spoke to were very skeptical about using AI for advice on how to treat a patient. They said it would be better to consult supervisors or colleagues, or case studies in the literature. They were also understandably very wary of inputting sensitive data into these tools.

There is some evidence AI can deliver more standardized, “manualized” therapies like CBT [cognitive behavioral therapy] reasonably effectively. So it’s possible it could be more useful for that. But that is AI specifically designed for that purpose, not general-purpose tools like ChatGPT.

What happens if this goes awry? What attention is this getting from ethics groups and lawmakers?

At present, professional bodies like the American Counseling Association advise against using AI tools to diagnose patients. There could also be more stringent regulations preventing this in future. Nevada and Illinois, for example, have recently passed laws prohibiting the use of AI in therapeutic decision-making. More states could follow.

OpenAI’s Sam Altman said last month that “a lot of people effectively use ChatGPT as a sort of therapist,” and that to him, that’s a good thing. Do you think tech companies are overpromising on AI’s ability to help us?

I think that tech companies are subtly encouraging this use of AI because clearly it’s a route through which some people are forming an attachment to their products. I think the main issue is that what people are getting from these tools isn’t really “therapy” by any stretch. Good therapy goes far beyond being soothing and validating everything someone says. I’ve never in my life looked forward to a (real, in-person) therapy session. They’re often highly uncomfortable, and even distressing. But that’s part of the point. The therapist should be challenging you and drawing you out and seeking to understand you. ChatGPT doesn’t do any of these things. 

Read the full story from Laurie Clarke. 

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

Ecommerce MGMT 0 Comments
Sep 10 2025
AI is changing the grid. Could it help more than it harms?

The rising popularity of AI is driving an increase in electricity demand so significant it has the potential to reshape our grid. Energy consumption by data centers has gone up by 80% from 2020 to 2025 and is likely to keep growing. Electricity prices are already rising, especially in places where data centers are most concentrated. 

Yet many people, especially in Big Tech, argue that AI will be, on balance, a positive force for the grid. They claim that the technology could help get more clean power online faster, run our power system more efficiently, and predict and prevent failures that cause blackouts. 

This story is a part of MIT Technology Review’s series “Power Hungry: AI and our energy future,” on the energy demands and carbon costs of the artificial-intelligence revolution.

There are early examples where AI is helping already, including AI tools that utilities are using to help forecast supply and demand. The question is whether these big promises will be realized fast enough to outweigh the negative effects of AI on local grids and communities. 

A delicate balance

One area where AI is already being used for the grid is in forecasting, says Utkarsha Agwan, a member of the nonprofit group Climate Change AI.

Running the grid is a balancing act: Operators have to understand how much electricity demand there is and turn on the right combination of power plants to meet it. They optimize for economics along the way, choosing the sources that will keep prices lowest for the whole system.

That makes it necessary to look ahead hours and in some cases days. Operators consider factors such as historical data (holidays often see higher demand) and the weather (a hot day means more air conditioners sucking up power). These predictions also consider what level of supply is expected from intermittent sources like solar panels.

There’s little risk in using AI tools in forecasting; it’s often not as time sensitive as other applications, which can require reactions within seconds or even milliseconds. A grid operator might use a forecast to determine which plants will need to turn on. Other groups might run their own forecasts as well, using AI tools to decide how to staff a plant, for example. The tools also can’t physically control anything. Rather, they can be used alongside more conventional methods to provide more data.  

Today, grid operators make a lot of approximations to model the grid, because the system is so incredibly complex that it’s impossible to truly know what’s going on in every place at every time. Not only are there a whole host of power plants and consumers to think about, but there are considerations like making sure power lines don’t get overloaded.

Working with those estimates can lead to some inefficiencies, says Kyri Baker, a professor at the University of Colorado Boulder. Operators tend to generate a bit more electricity than the system uses, for example. Using AI to create a better model could reduce some of those losses and allow operators to make decisions about how to control infrastructure in real time to reach a closer match of supply and demand.

She gives the example of a trip to the airport. Imagine there’s a route you know will get you there in about 45 minutes. There might be another, more complicated route that could save you some time in ideal conditions—but you’re not sure whether it’s better on any particular day. What the grid does now is the equivalent of taking the reliable route.

“So that’s the gap that AI can help close. We can solve this more complex problem, fast enough and reliably enough that we can possibly use it and shave off emissions,” Baker says. 

In theory, AI could be used to operate the grid entirely without human intervention. But that work is largely still in the research phase. Grid operators are running some of the most critical infrastructure in this country, and the industry is hesitant to mess with something that’s already working, Baker says. If this sort of technology is ever used in grid operations, there will still be humans in the loop to help make decisions, at least when it’s first deployed.  

Planning ahead

Another fertile area for AI is planning future updates to the grid. Building a power plant can take a very long time—the typical time from an initial request to commercial operation in the US is roughly four years. One reason for the lengthy wait is that new power plants have to demonstrate how they might affect the rest of the grid before they can connect. 

An interconnection study examines whether adding a new power plant of a particular type in a particular place would require upgrades to the grid to prevent problems. After regulators and utilities determine what upgrades might be needed, they estimate the cost, and the energy developer generally foots the bill. 

Today, those studies can take months. They involve trying to understand an incredibly complicated system, and because they rely on estimates of other existing and proposed power plants, only a few can happen in an area at any given time. This has helped create the years-long interconnection queue, a long line of plants waiting for their turn to hook up to the grid in markets like the US and Europe. The vast majority of projects in the queue today are renewables, which means there’s clean power just waiting to come online. 

AI could help speed this process, producing these reports more quickly. The Midcontinent Independent System Operator, a grid operator that covers 15 states in the central US, is currently working with a company called Pearl Street to help automate these reports.

AI won’t be a cure-all for grid planning; there are other steps to clearing the interconnection queue, including securing the necessary permits. But the technology could help move things along. “The sooner we can speed up interconnection, the better off we’ll be,” says Rob Gramlich, president of Grid Strategies, a consultancy specializing in transmission and power markets.

There’s a growing list of other potential uses for AI on the grid and in electricity generation. The technology could monitor and plan ahead for failures in equipment ranging from power lines to gear boxes. Computer vision could help detect everything from wildfires to faulty lines. AI could also help balance supply and demand in virtual power plants, systems of distributed resources like EV chargers or smart water heaters. 

While there are early examples of research and pilot programs for AI from grid planning to operation, some experts are skeptical that the technology will deliver at the level some are hoping for. “It’s not that AI has not had some kind of transformation on power systems,” Climate Change AI’s Agwan says. “It’s that the promise has always been bigger, and the hope has always been bigger.”

Some places are already seeing higher electricity prices because of power needs from data centers. The situation is likely to get worse. Electricity demand from data centers is set to double by the end of the decade, reaching 945 terawatt-hours, roughly the annual demand from the entire country of Japan. 

The infrastructure growth needed to support AI load growth has outpaced the promises of the technology, “by quite a bit,” says Panayiotis Moutis, an assistant professor of electrical engineering at the City College of New York. Higher bills caused by the increasing energy needs of AI aren’t justified by existing ways of using the technology for the grid, he says. 

“At the moment, I am very hesitant to lean on the side of AI being a silver bullet,” Moutis says. 

Correction: This story has been updated to correct Moutis’s affiliation.

Ecommerce MGMT 0 Comments
Sep 10 2025
Three big things we still don’t know about AI’s energy burden

Earlier this year, when my colleague Casey Crownhart and I spent six months researching the climate and energy burden of AI, we came to see one number in particular as our white whale: how much energy the leading AI models, like ChatGPT or Gemini, use up when generating a single response. 

This fundamental number remained elusive even as the scramble to power AI escalated to the White House and the Pentagon, and as projections showed that in three years AI could use as much electricity as 22% of all US households. 

The problem with finding that number, as we explain in our piece published in May, was that AI companies are the only ones who have it. We pestered Google, OpenAI, and Microsoft, but each company refused to provide its figure. Researchers we spoke to who study AI’s impact on energy grids compared it to trying to measure the fuel efficiency of a car without ever being able to drive it, making guesses based on rumors of its engine size and what it sounds like going down the highway.

This story is a part of MIT Technology Review’s series “Power Hungry: AI and our energy future,” on the energy demands and carbon costs of the artificial-intelligence revolution.

But then this summer, after we published, a strange thing started to happen. In June, OpenAI’s Sam Altman wrote that an average ChatGPT query uses 0.34 watt-hours of energy. In July, the French AI startup Mistral didn’t publish a number directly but released an estimate of the emissions generated. In August, Google revealed that answering a question to Gemini uses about 0.24 watt-hours of energy. The figures from Google and OpenAI were similar to what Casey and I estimated for medium-size AI models. 

So with this newfound transparency, is our job complete? Did we finally harpoon our white whale, and if so, what happens next for people studying the climate impact of AI? I reached out to some of our old sources, and some new ones, to find out.

The numbers are vague and chat-only

The first thing they told me is that there’s a lot missing from the figures tech companies published this summer. 

OpenAI’s number, for example, did not appear in a detailed technical paper but rather in a blog post by Altman that leaves lots of unanswered questions, such as which model he was referring to, how the energy use was measured, and how much it varies. Google’s figure, as Crownhart points out, refers to the median amount of energy per query, which doesn’t give us a sense of the more energy-demanding Gemini responses, like when it uses a reasoning model to “think” through a hard problem or generates a really long response. 

The numbers also refer only to interactions with chatbots, not the other ways that people are becoming increasingly reliant on generative AI. 

“As video and image becomes more prominent and used by more and more people, we need the numbers from different modalities and how they measure up,” says Sasha Luccioni, AI and climate lead at the AI platform Hugging Face. 

This is also important because the figures for asking a question to a chatbot are, as expected, undoubtedly small—the same amount of electricity used by a microwave in just seconds. That’s part of the reason AI and climate researchers don’t suggest that any one individual’s AI use creates a significant climate burden. 

A full accounting of AI’s energy demands—one that goes beyond what’s used to answer an individual query to help us understand its full net impact on the climate—would require application-specific information on how all this AI is being used. Ketan Joshi, an analyst for climate and energy groups, acknowledges that researchers don’t usually get such specific information from other industries but says it might be justified in this case.

“The rate of data center growth is inarguably unusual,” Joshi says. “Companies should be subject to significantly more scrutiny.”

We have questions about energy efficiency

Companies making billion-dollar investments into AI have struggled to square this growth in energy demand with their sustainability goals. In May, Microsoft said that its emissions have soared by over 23% since 2020, owing largely to AI, while the company has promised to be carbon negative by 2030. “It has become clear that our journey towards being carbon negative is a marathon, not a sprint,” Microsoft wrote.

Tech companies often justify this emissions burden by arguing that soon enough, AI itself will unlock efficiencies that will make it a net positive for the climate. Perhaps the right AI system, the thinking goes, could design more efficient heating and cooling systems for a building, or help discover the minerals required for electric-vehicle batteries. 

But there are no signs that AI has been usefully used to do these things yet. Companies have shared anecdotes about using AI to find methane emission hot spots, for example, but they haven’t been transparent enough to help us know if these successes outweigh the surges in electricity demand and emissions that Big Tech has produced in the AI boom. In the meantime, more data centers are planned, and AI’s energy demand continues to rise and rise. 

The ‘bubble’ question

One of the big unknowns in the AI energy equation is whether society will ever adopt AI at the levels that figure into tech companies’ plans. OpenAI has said that ChatGPT receives 2.5 billion prompts per day. It’s possible that this number, and the equivalent numbers for other AI companies, will continue to soar in the coming years. Projections released last year by the Lawrence Berkeley National Laboratory suggest that if they do, AI alone could consume as much electricity annually as 22% of all US households by 2028.

But this summer also saw signs of a slowdown that undercut the industry’s optimism. OpenAI’s launch of GPT-5 was largely considered a flop, even by the company itself, and that flop led critics to wonder if AI may be hitting a wall. When a group at MIT found that 95% of businesses are seeing no return on their massive AI investments, stocks floundered. The expansion of AI-specific data centers might be an investment that’s hard to recoup, especially as revenues for AI companies remain elusive. 

One of the biggest unknowns about AI’s future energy burden isn’t how much a single query consumes, or any other figure that can be disclosed. It’s whether demand will ever reach the scale companies are building for or whether the technology will collapse under its own hype. The answer will determine whether today’s buildout becomes a lasting shift in our energy system or a short-lived spike.

Ecommerce MGMT 0 Comments
Sep 9 2025
2025 Innovator of the Year: Sneha Goenka for developing an ultra-fast sequencing technology

Sneha Goenka is one of MIT Technology Review’s 2025 Innovators Under 35. Meet the rest of this year’s honorees. 

Up to a quarter of children entering intensive care have undiagnosed genetic conditions. To be treated properly, they must first get diagnoses—which means having their genomes sequenced. This process typically takes up to seven weeks. Sadly, that’s often too slow to save a critically ill child.

Hospitals may soon have a faster option, thanks to a groundbreaking system built in part by Sneha Goenka, an assistant professor of electrical and computer engineering at Princeton—and MIT Technology Review’s 2025 Innovator of the Year. 

Five years ago, Goenka and her colleagues designed a rapid-sequencing pipeline that can provide a genetic diagnosis in less than eight hours. Goenka’s software computations and hardware architectures were critical to speeding up each stage of the process. 

“Her work made everyone realize that genome sequencing is not only for research and medical application in the future but can have immediate impact on patient care,” says Jeroen de Ridder, a professor at UMC Utrecht in the Netherlands, who has developed an ultrafast sequencing tool for cancer diagnosis. 

Now, as cofounder and scientific lead of a new company, she is working to make that technology widely available to patients around the world.

Goenka grew up in Mumbai, India. Her mother was an advocate for women’s education, but as a child, Goenka had to fight to persuade other family members to let her continue her studies. She moved away from home at 15 to attend her final two years of school and enroll in a premier test-­preparation academy in Kota, Rajasthan. Thanks to that education, she passed what she describes as “one of the most competitive exams in the world,” to get into the Indian Institute of Technology Bombay. 

Once admitted to a combined bachelor’s and master’s program in electrical engineering, she found that “it was a real boys’ club.” But Goenka excelled in developing computer architecture systems that accelerate computation. As an undergraduate, she began applying those skills to medicine, driven by her desire to “have real-world impact”—in part because she had seen her family struggle with painful uncertainty after her brother was born prematurely when she was eight years old. 

While working on a PhD in electrical engineering at Stanford, she turned her focus to evolutionary and clinical genomics. One day a senior colleague, Euan Ashley, presented her with a problem. He said, “We want to see how fast we can make a genetic diagnosis. If you had unlimited funds and resources, just how fast do you think you could make the compute?”

Streaming DNA

A genetic diagnosis starts with a blood sample, which is prepped to extract the DNA—a process that takes about three hours. Next that DNA needs to be “read.” One of the world’s leading long-read sequencing technologies, developed by Oxford Nanopore Technologies, can generate highly detailed raw data of an individual’s genetic code in about an hour and a half. Unfortunately, processing all this data to identify mutations can take another 21 hours. Shipping samples to a central lab and figuring out which mutations are of interest often leads the process to stretch out to weeks. 

Goenka saw a better way: Build a real-time system that could “stream” the sequencing data, analyzing it as it was being generated, like streaming a film on Netflix rather than downloading it to watch later.

Sneha Goenka

To do this, she designed a cloud computing architecture to pull in more processing power. Goenka’s first challenge was to increase the speed at which her team could upload the raw data for processing, by streamlining the requests between the sequencer and the cloud to avoid unnecessary “chatter.” She worked out the exact number of communication channels needed—and created algorithms that allowed those channels to be reused in the most efficient way.

The next challenge was “base calling”—converting the raw signal from the sequencing machine into the nucleotide bases A, C, T, and G, the language that makes up our DNA. Rather than using a central node to orchestrate this process, which is an inefficient, error-prone approach, Goenka wrote software to automatically assign dozens of data streams directly from the sequencer to dedicated nodes in the cloud.

Meet the rest of this year’s 
Innovators Under 35.

Then, to identify mutations, the sequences were aligned for comparison with a reference genome. She coded a custom program that triggers alignment as soon as base calling finishes for one batch of sequences while simultaneously initiating base calling for the next batch, thus ensuring that the system’s computational resources are used efficiently.

Add all these im­­prove­­ments together, and Goenka’s approach reduced the total time required to analyze a genome for mutations from around 20 hours to 1.5 hours. Finally, the team worked with genetic counselors and physicians to create a filter that identified which mutations were most critical to a person’s health, and that set was then given a final manual curation by a genetic specialist. These final stages take up to three hours. The technology was close to being fully operational when, suddenly, the first patient arrived. 

A critical test

When 13-year-old Matthew was flown to Stanford’s children’s hospital in 2021, he was struggling to breathe and his heart was failing. Doctors needed to know whether the inflammation in his heart was due to a virus or to a genetic mutation that would necessitate a transplant.  

His blood was drawn on a Thursday. The transplant committee made its decisions on Fridays. “It meant we had a small window of time,” says Goenka.

Goenka was in Mumbai when the sequencing began. She stayed up all night, monitoring the computations. That was when the project stopped being about getting faster for the sake of it, she says: “It became about ‘How fast can we get this result to save this person’s life?’”

The results revealed a genetic mutation that explained Matthew’s condition, and he was placed on the transplant list the next day. Three weeks later, he received a new heart. “He’s doing great now,” Goenka says.

So far, Goenka’s technology has been tested on 26 patients, including Matthew. Her pipeline is “directly affecting the medical care of newborns in the Stanford intensive care units,” Ashley says.

Now she’s aiming for even broader impact—Goenka and her colleagues are laying the groundwork for a startup that they hope will bring the technology to market and make sure it reaches as many patients as possible. Meanwhile, she has been refining the computational pipeline, reducing the time to diagnosis to about six hours.

The demand is clear, she says: “In an in-depth study involving more than a dozen laboratory directors and neonatologists, every respondent stressed urgency. One director put it succinctly: ‘I need this platform today—preferably yesterday.’”

Goenka is also developing software to make the technology more inclusive. The reference genome is skewed toward people of European descent. The Human Pangenome Project is an international collaboration to create reference genomes from more diverse populations, which Goenka aims to use to personalize her team’s filters, allowing them to flag mutations that may be more prevalent in the population to which a patient belongs.

Since seeing her work, Goenka’s extended family has become more appreciative of her education and career. “The entire family is very proud about the impact I’ve made,” she says. 

Helen Thomson is a freelance science journalist based in London.

Ecommerce MGMT 0 Comments