Welcome back to The State of AI, a new collaboration between the Financial Times and MIT Technology Review. Every Monday, writers from both publications debate one aspect of the generative AI revolution reshaping global power. You can read the rest of the series here.

In this final edition, MIT Technology Review’s senior AI editor Will Douglas Heaven talks with Tim Bradshaw, FT global tech correspondent, about where AI will go next, and what our world will look like in the next five years.

(As part of this series, join MIT Technology Review’s editor in chief, Mat Honan, and editor at large, David Rotman, for an exclusive conversation with Financial Times columnist Richard Waters on how AI is reshaping the global economy. Live on Tuesday, December 9 at 1:00 p.m. ET. This is a subscriber-only event and you can sign up here.)

Will Douglas Heaven writes: 

Every time I’m asked what’s coming next, I get a Luke Haines song stuck in my head: “Please don’t ask me about the future / I am not a fortune teller.” But here goes. What will things be like in 2030? My answer: same but different. 

There are huge gulfs of opinion when it comes to predicting the near-future impacts of generative AI. In one camp we have the AI Futures Project, a small donation-funded research outfit led by former OpenAI researcher Daniel Kokotajlo. The nonprofit made a big splash back in April with AI 2027, a speculative account of what the world will look like two years from now. 

The story follows the runaway advances of an AI firm called OpenBrain (any similarities are coincidental, etc.) all the way to a choose-your-own-adventure-style boom or doom ending. Kokotajlo and his coauthors make no bones about their expectation that in the next decade the impact of AI will exceed that of the Industrial Revolution—a 150-year period of economic and social upheaval so great that we still live in the world it wrought.

At the other end of the scale we have team Normal Technology: Arvind Narayanan and Sayash Kapoor, a pair of Princeton University researchers and coauthors of the book AI Snake Oil, who push back not only on most of AI 2027’s predictions but, more important, on its foundational worldview. That’s not how technology works, they argue.

Advances at the cutting edge may come thick and fast, but change across the wider economy, and society as a whole, moves at human speed. Widespread adoption of new technologies can be slow; acceptance slower. AI will be no different. 

What should we make of these extremes? ChatGPT came out three years ago last month, but it’s still not clear just how good the latest versions of this tech are at replacing lawyers or software developers or (gulp) journalists. And new updates no longer bring the step changes in capability that they once did. 

And yet this radical technology is so new it would be foolish to write it off so soon. Just think: Nobody even knows exactly how this technology works—let alone what it’s really for. 

As the rate of advance in the core technology slows down, applications of that tech will become the main differentiator between AI firms. (Witness the new browser wars and the chatbot pick-and-mix already on the market.) At the same time, high-end models are becoming cheaper to run and more accessible. Expect this to be where most of the action is: New ways to use existing models will keep them fresh and distract people waiting in line for what comes next. 

Meanwhile, progress continues beyond LLMs. (Don’t forget—there was AI before ChatGPT, and there will be AI after it too.) Technologies such as reinforcement learning—the powerhouse behind AlphaGo, DeepMind’s board-game-playing AI that beat a Go grand master in 2016—is set to make a comeback. There’s also a lot of buzz around world models, a type of generative AI with a stronger grip on how the physical world fits together than LLMs display. 

Ultimately, I agree with team Normal Technology that rapid technological advances do not translate to economic or societal ones straight away. There’s just too much messy human stuff in the middle. 

But Tim, over to you. I’m curious to hear what your tea leaves are saying. 

Tim Bradshaw responds:

Will, I am more confident than you that the world will look quite different in 2030. In five years’ time, I expect the AI revolution to have proceeded apace. But who gets to benefit from those gains will create a world of AI haves and have-nots.

It seems inevitable that the AI bubble will burst sometime before the end of the decade. Whether a venture capital funding shakeout comes in six months or two years (I feel the current frenzy still has some way to run), swathes of AI app developers will disappear overnight. Some will see their work absorbed by the models upon which they depend. Others will learn the hard way that you can’t sell services that cost $1 for 50 cents without a firehose of VC funding.

How many of the foundation model companies survive is harder to call, but it already seems clear that OpenAI’s chain of interdependencies within Silicon Valley make it too big to fail. Still, a funding reckoning will force it to ratchet up pricing for its services.

When OpenAI was created in 2015, it pledged to “advance digital intelligence in the way that is most likely to benefit humanity as a whole.” That seems increasingly untenable. Sooner or later, the investors who bought in at a $500 billion price tag will push for returns. Those data centers won’t pay for themselves. By that point, many companies and individuals will have come to depend on ChatGPT or other AI services for their everyday workflows. Those able to pay will reap the productivity benefits, scooping up the excess computing power as others are priced out of the market.

Being able to layer several AI services on top of each other will provide a compounding effect. One example I heard on a recent trip to San Francisco: Ironing out the kinks in vibe coding is simply a matter of taking several passes at the same problem and then running a few more AI agents to look for bugs and security issues. That sounds incredibly GPU-intensive, implying that making AI really deliver on the current productivity promise will require customers to pay far more than most do today.

The same holds true in physical AI. I fully expect robotaxis to be commonplace in every major city by the end of the decade, and I even expect to see humanoid robots in many homes. But while Waymo’s Uber-like prices in San Francisco and the kinds of low-cost robots produced by China’s Unitree give the impression today that these will soon be affordable for all, the compute cost involved in making them useful and ubiquitous seems destined to turn them into luxuries for the well-off, at least in the near term.

The rest of us, meanwhile, will be left with an internet full of slop and unable to afford AI tools that actually work.

Perhaps some breakthrough in computational efficiency will avert this fate. But the current AI boom means Silicon Valley’s AI companies lack the incentives to make leaner models or experiment with radically different kinds of chips. That only raises the likelihood that the next wave of AI innovation will come from outside the US, be that China, India, or somewhere even farther afield.

Silicon Valley’s AI boom will surely end before 2030, but the race for global influence over the technology’s development—and the political arguments about how its benefits are distributed—seem set to continue well into the next decade. 

Will replies: 

I am with you that the cost of this technology is going to lead to a world of haves and have-nots. Even today, $200+ a month buys power users of ChatGPT or Gemini a very different experience from that of people on the free tier. That capability gap is certain to increase as model makers seek to recoup costs. 

We’re going to see massive global disparities too. In the Global North, adoption has been off the charts. A recent report from Microsoft’s AI Economy Institute notes that AI is the fastest-spreading technology in human history: “In less than three years, more than 1.2 billion people have used AI tools, a rate of adoption faster than the internet, the personal computer, or even the smartphone.” And yet AI is useless without ready access to electricity and the internet; swathes of the world still have neither. 

I still remain skeptical that we will see anything like the revolution that many insiders promise (and investors pray for) by 2030. When Microsoft talks about adoption here, it’s counting casual users rather than measuring long-term technological diffusion, which takes time. Meanwhile, casual users get bored and move on. 

How about this: If I live with a domestic robot in five years’ time, you can send your laundry to my house in a robotaxi any day of the week. 

JK! As if I could afford one. 

Further reading 

What is AI? It sounds like a stupid question, but it’s one that’s never been more urgent. In this deep dive, Will unpacks decades of spin and speculation to get to the heart of our collective technodream. 

AGI—the idea that machines will be as smart as humans—has hijacked an entire industry (and possibly the US economy). For MIT Technology Review’s recent New Conspiracy Age package, Will takes a provocative look at how AGI is like a conspiracy. 

The FT examined the economics of self-driving cars this summer, asking who will foot the multi-billion-dollar bill to buy enough robotaxis to serve a big city like London or New York.

A plausible counter-argument to Tim’s thesis on AI inequalities is that freely available open-source (or more accurately, “open weight”) models will keep pulling down prices. The US may want frontier models to be built on US chips but it is already losing the global south to Chinese software.

In 2017, fresh off a PhD on theoretical chemistry, John Jumper heard rumors that Google DeepMind had moved on from building AI that played games with superhuman skill and was starting up a secret project to predict the structures of proteins. He applied for a job.

Just three years later, Jumper celebrated a stunning win that few had seen coming. With CEO Demis Hassabis, he had co-led the development of an AI system called AlphaFold 2 that was able to predict the structures of proteins to within the width of an atom, matching the accuracy of painstaking techniques used in the lab, and doing it many times faster—returning results in hours instead of months.

AlphaFold 2 had cracked a 50-year-old grand challenge in biology. “This is the reason I started DeepMind,” Hassabis told me a few years ago. “In fact, it’s why I’ve worked my whole career in AI.” In 2024, Jumper and Hassabis shared a Nobel Prize in chemistry.

It was five years ago this week that AlphaFold 2’s debut took scientists by surprise. Now that the hype has died down, what impact has AlphaFold really had? How are scientists using it? And what’s next? I talked to Jumper (as well as a few other scientists) to find out.

“It’s been an extraordinary five years,” Jumper says, laughing: “It’s hard to remember a time before I knew tremendous numbers of journalists.”

AlphaFold 2 was followed by AlphaFold Multimer, which could predict structures that contained more than one protein, and then AlphaFold 3, the fastest version yet. Google DeepMind also let AlphaFold loose on UniProt, a vast protein database used and updated by millions of researchers around the world. It has now predicted the structures of some 200 million proteins, almost all that are known to science.

Despite his success, Jumper remains modest about AlphaFold’s achievements. “That doesn’t mean that we’re certain of everything in there,” he says. “It’s a database of predictions, and it comes with all the caveats of predictions.”

A hard problem

Proteins are the biological machines that make living things work. They form muscles, horns, and feathers; they carry oxygen around the body and ferry messages between cells; they fire neurons, digest food, power the immune system; and so much more. But understanding exactly what a protein does (and what role it might play in various diseases or treatments) involves figuring out its structure—and that’s hard.

Proteins are made from strings of amino acids that chemical forces twist up into complex knots. An untwisted string gives few clues about the structure it will form. In theory, most proteins could take on an astronomical number of possible shapes. The task is to predict the correct one.

Jumper and his team built AlphaFold 2 using a type of neural network called a transformer, the same technology that underpins large language models. Transformers are very good at paying attention to specific parts of a larger puzzle.

But Jumper puts a lot of the success down to making a prototype model that they could test quickly. “We got a system that would give wrong answers at incredible speed,” he says. “That made it easy to start becoming very adventurous with the ideas you try.”

They stuffed the neural network with as much information about protein structures as they could, such as how proteins across certain species have evolved similar shapes. And it worked even better than they expected. “We were sure we had made a breakthrough,” says Jumper. “We were sure that this was an incredible advance in ideas.”

What he hadn’t foreseen was that researchers would download his software and start using it straight away for so many different things. Normally, it’s the thing a few iterations down the line that has the real impact, once the kinks have been ironed out, he says: “I’ve been shocked at how responsibly scientists have used it, in terms of interpreting it, and using it in practice about as much as it should be trusted in my view, neither too much nor too little.”

Any projects stand out in particular? 

Honeybee science

Jumper brings up a research group that uses AlphaFold to study disease resistance in honeybees. “They wanted to understand this particular protein as they look at things like colony collapse,” he says. “I never would have said, ‘You know, of course AlphaFold will be used for honeybee science.’”

He also highlights a few examples of what he calls off-label uses of AlphaFold—“in the sense that it wasn’t guaranteed to work”—where the ability to predict protein structures has opened up new research techniques. “The first is very obviously the advances in protein design,” he says. “David Baker and others have absolutely run with this technology.”

Baker, a computational biologist at the University of Washington, was a co-winner of last year’s chemistry Nobel, alongside Jumper and Hassabis, for his work on creating synthetic proteins to perform specific tasks—such as treating disease or breaking down plastics—better than natural proteins can.

Baker and his colleagues have developed their own tool based on AlphaFold, called RoseTTAFold. But they have also experimented with AlphaFold Multimer to predict which of their designs for potential synthetic proteins will work.    

“Basically, if AlphaFold confidently agrees with the structure you were trying to design then you make it and if AlphaFold says ‘I don’t know,’ you don’t make it. That alone was an enormous improvement.” It can make the design process 10 times faster, says Jumper.

Another off-label use that Jumper highlights: Turning AlphaFold into a kind of search engine. He mentions two separate research groups that were trying to understand exactly how human sperm cells hooked up with eggs during fertilization. They knew one of the proteins involved but not the other, he says: “And so they took a known egg protein and ran all 2,000 human sperm surface proteins, and they found one that AlphaFold was very sure stuck against the egg.” They were then able to confirm this in the lab.

“This notion that you can use AlphaFold to do something you couldn’t do before—you would never do 2,000 structures looking for one answer,” he says. “This kind of thing I think is really extraordinary.”

Five years on

When AlphaFold 2 came out, I asked a handful of early adopters what they made of it. Reviews were good, but the technology was too new to know for sure what long-term impact it might have. I caught up with one of those people to hear his thoughts five years on.

Kliment Verba is a molecular biologist who runs a lab at the University of California, San Francisco. “It’s an incredibly useful technology, there’s no question about it,” he tells me. “We use it every day, all the time.”

But it’s far from perfect. A lot of scientists use AlphaFold to study pathogens or to develop drugs. This involves looking at interactions between multiple proteins or between proteins and even smaller molecules in the body. But AlphaFold is known to be less accurate at making predictions about multiple proteins or their interaction over time.

Verba says he and his colleagues have been using AlphaFold long enough to get used to its limitations. “There are many cases where you get a prediction and you have to kind of scratch your head,” he says. “Is this real or is this not? It’s not entirely clear—it’s sort of borderline.”

“It’s sort of the same thing as ChatGPT,” he adds. “You know—it will bullshit you with the same confidence as it would give a true answer.”

Still, Verba’s team uses AlphaFold (both 2 and 3, because they have different strengths, he says) to run virtual versions of their experiments before running them in the lab. Using AlphaFold’s results, they can narrow down the focus of an experiment—or decide that it’s not worth doing.

It can really save time, he says: “It hasn’t really replaced any experiments, but it’s augmented them quite a bit.”

New wave  

AlphaFold was designed to be used for a range of purposes. Now multiple startups and university labs are building on its success to develop a new wave of tools more tailored to drug discovery. This year, a collaboration between MIT researchers and the AI drug company Recursion produced a model called Boltz-2, which predicts not only the structure of proteins but also how well potential drug molecules will bind to their target.  

Last month, the startup Genesis Molecular AI released another structure prediction model called Pearl, which the firm claims is more accurate than AlphaFold 3 for certain queries that are important for drug development. Pearl is interactive, so that drug developers can feed any additional data they may have to the model to guide its predictions.

AlphaFold was a major leap, but there’s more to do, says Evan Feinberg, Genesis Molecular AI’s CEO: “We’re still fundamentally innovating, just with a better starting point than before.”

Genesis Molecular AI is pushing margins of error down from less than two angstroms, the de facto industry standard set by AlphaFold, to less than one angstrom—one 10-millionth of a millimeter, or the width of a single hydrogen atom.

“Small errors can be catastrophic for predicting how well a drug will actually bind to its target,” says Michael LeVine, vice president of modeling and simulation at the firm. That’s because chemical forces that interact at one angstrom can stop doing so at two. “It can go from ‘They will never interact’ to ‘They will,’” he says.

With so much activity in this space, how soon should we expect new types of drugs to hit the market? Jumper is pragmatic. Protein structure prediction is just one step of many, he says: “This was not the only problem in biology. It’s not like we were one protein structure away from curing any diseases.”

Think of it this way, he says. Finding a protein’s structure might previously have cost $100,000 in the lab: “If we were only a hundred thousand dollars away from doing a thing, it would already be done.”

At the same time, researchers are looking for ways to do as much as they can with this technology, says Jumper: “We’re trying to figure out how to make structure prediction an even bigger part of the problem, because we have a nice big hammer to hit it with.”

In other words, make everything into nails? “Yeah, let’s make things into nails,” he says. “How do we make this thing that we made a million times faster a bigger part of our process?”

What’s next?

Jumper’s next act? He wants to fuse the deep but narrow power of AlphaFold with the broad sweep of LLMs.  

“We have machines that can read science. They can do some scientific reasoning,” he says. “And we can build amazing, superhuman systems for protein structure prediction. How do you get these two technologies to work together?”

That makes me think of a system called AlphaEvolve, which is being built by another team at Google DeepMind. AlphaEvolve uses an LLM to generate possible solutions to a problem and a second model to check them, filtering out the trash. Researchers have already used AlphaEvolve to make a handful of practical discoveries in math and computer science.    

Is that what Jumper has in mind? “I won’t say too much on methods, but I’ll be shocked if we don’t see more and more LLM impact on science,” he says. “I think that’s the exciting open question that I’ll say almost nothing about. This is all speculation, of course.”

Jumper was 39 when he won his Nobel Prize. What’s next for him?

“It worries me,” he says. “I believe I’m the youngest chemistry laureate in 75 years.” 

He adds: “I’m at the midpoint of my career, roughly. I guess my approach to this is to try to do smaller things, little ideas that you keep pulling on. The next thing I announce doesn’t have to be, you know, my second shot at a Nobel. I think that’s the trap.”

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.

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!

Last week, the technology companies Anthropic and Meta each won landmark victories in two separate court cases that examined whether or not the firms had violated copyright when they trained their large language models on copyrighted books without permission. The rulings are the first we’ve seen to come out of copyright cases of this kind. This is a big deal!

The use of copyrighted works to train models is at the heart of a bitter battle between tech companies and content creators. That battle is playing out in technical arguments about what does and doesn’t count as fair use of a copyrighted work. But it is ultimately about carving out a space in which human and machine creativity can continue to coexist.

There are dozens of similar copyright lawsuits working through the courts right now, with cases filed against all the top players—not only Anthropic and Meta but Google, OpenAI, Microsoft, and more. On the other side, plaintiffs range from individual artists and authors to large companies like Getty and the New York Times.

The outcomes of these cases are set to have an enormous impact on the future of AI. In effect, they will decide whether or not model makers can continue ordering up a free lunch. If not, they will need to start paying for such training data via new kinds of licensing deals—or find new ways to train their models. Those prospects could upend the industry.

And that’s why last week’s wins for the technology companies matter. So: Cases closed? Not quite. If you drill into the details, the rulings are less cut-and-dried than they seem at first. Let’s take a closer look.

In both cases, a group of authors (the Anthropic suit was a class action; 13 plaintiffs sued Meta, including high-profile names such as Sarah Silverman and Ta-Nehisi Coates) set out to prove that a technology company had violated their copyright by using their books to train large language models. And in both cases, the companies argued that this training process counted as fair use, a legal provision that permits the use of copyrighted works for certain purposes.  

There the similarities end. Ruling in Anthropic’s favor, senior district judge William Alsup argued on June 23 that the firm’s use of the books was legal because what it did with them was transformative, meaning that it did not replace the original works but made something new from them. “The technology at issue was among the most transformative many of us will see in our lifetimes,” Alsup wrote in his judgment.

In Meta’s case, district judge Vince Chhabria made a different argument. He also sided with the technology company, but he focused his ruling instead on the issue of whether or not Meta had harmed the market for the authors’ work. Chhabria said that he thought Alsup had brushed aside the importance of market harm. “The key question in virtually any case where a defendant has copied someone’s original work without permission is whether allowing people to engage in that sort of conduct would substantially diminish the market for the original,” he wrote on June 25.

Same outcome; two very different rulings. And it’s not clear exactly what that means for the other cases. On the one hand, it bolsters at least two versions of the fair-use argument. On the other, there’s some disagreement over how fair use should be decided.

But there are even bigger things to note. Chhabria was very clear in his judgment that Meta won not because it was in the right, but because the plaintiffs failed to make a strong enough argument. “In the grand scheme of things, the consequences of this ruling are limited,” he wrote. “This is not a class action, so the ruling only affects the rights of these 13 authors—not the countless others whose works Meta used to train its models. And, as should now be clear, this ruling does not stand for the proposition that Meta’s use of copyrighted materials to train its language models is lawful.” That reads a lot like an invitation for anyone else out there with a grievance to come and have another go.   

And neither company is yet home free. Anthropic and Meta both face wholly separate allegations that not only did they train their models on copyrighted books, but the way they obtained those books was illegal because they downloaded them from pirated databases. Anthropic now faces another trial over these piracy claims. Meta has been ordered to begin a discussion with its accusers over how to handle the issue.

So where does that leave us? As the first rulings to come out of cases of this type, last week’s judgments will no doubt carry enormous weight. But they are also the first rulings of many. Arguments on both sides of the dispute are far from exhausted.

“These cases are a Rorschach test in that either side of the debate will see what they want to see out of the respective orders,” says Amir Ghavi, a lawyer at Paul Hastings who represents a range of technology companies in ongoing copyright lawsuits. He also points out that the first cases of this type were filed more than two years ago: “Factoring in likely appeals and the other 40+ pending cases, there is still a long way to go before the issue is settled by the courts.”

“I’m disappointed at these rulings,” says Tyler Chou, founder and CEO of Tyler Chou Law for Creators, a firm that represents some of the biggest names on YouTube. “I think plaintiffs were out-gunned and didn’t have the time or resources to bring the experts and data that the judges needed to see.”

But Chou thinks this is just the first round of many. Like Ghavi, she thinks these decisions will go to appeal. And after that we’ll see cases start to wind up in which technology companies have met their match: “Expect the next wave of plaintiffs—publishers, music labels, news organizations—to arrive with deep pockets,” she says. “That will be the real test of fair use in the AI era.”

But even when the dust has settled in the courtrooms—what then? The problem won’t have been solved. That’s because the core grievance of creatives, whether individuals or institutions, is not really that their copyright has been violated—copyright is just the legal hammer they have to hand. Their real complaint is that their livelihoods and business models are at risk of being undermined. And beyond that: when AI slop devalues creative effort, will people’s motivations for putting work out into the world start to fall away?

In that sense, these legal battles are set to shape all our futures. There’s still no good solution on the table for this wider problem. Everything is still to play for.

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

This story has been edited to add comments from Tyler Chou.

If you want to know where AI is headed, this year’s Google I/O has you covered. The company’s annual showcase of next-gen products, which kicked off yesterday, has all of the pomp and pizzazz, the sizzle reels and celebrity walk-ons, that you’d expect from a multimillion-dollar marketing event.

But it also shows us just how fast this still experimental technology is being subsumed into a lineup designed to sell phones and subscription tiers. Never before have I seen this thing we call artificial intelligence appear so normal.

Yes, Google’s roster of consumer-facing products is the slickest on offer. The firm is bundling most of its multimodal models into its Gemini app, including the new Imagen 4 image generator and the new Veo 3 video generator. That means you can now access Google’s full range of generative models via a single chatbot. It also announced Gemini Live, a feature that lets you share your phone’s screen or your camera’s view with the chatbot and ask it about what it can see.

Those features were previously only seen in demos of Project Astra, a “universal AI assistant“ that Google DeepMind is working on. Now, Google is inching toward putting Project Astra into the hands of anyone with a smartphone.

Google is also rolling out AI Mode, an LLM-powered front end to search. This can now pull in personal information from Gmail or Google Docs to tailor searches to users. It will include Deep Search, which can break a query down into hundreds of individual searches and then summarize the results; a version of Project Mariner, Google DeepMind’s browser-using agent; and Search Live, which lets you hold up your camera and ask it what it sees.

This is the new frontier. It’s no longer about who has the most powerful models, but who can spin them into the best products. OpenAI’s ChatGPT includes many similar features to Gemini’s. But with its existing ecosystem of consumer services and billions of existing users, Google has a clear advantage. Power users wanting access to the latest versions of everything on display can now sign up for Google AI Ultra for $250 a month.  

When OpenAI released ChatGPT in late 2022, Google was caught on the back foot and was forced to jump into higher gear to catch up. With this year’s product lineup, it looks as if Google has stuck its landing.

On a preview call, CEO Sundar Pichai claimed that AI Overviews, a precursor to AI Mode that provides LLM-generated summaries of search results, had turned out to be popular with hundreds of millions of users. He speculated that many of them may not even know (or care) that they were using AI—it was just a cool new way to search. Google I/O gives a broader glimpse of that future, one where AI is invisible.

“More intelligence is available, for everyone, everywhere,” Pichai told his audience. I think we are expected to marvel. But by putting AI in everything, Google is turning AI into a technology we won’t notice and may not even bother to name.