Does AI need an antitrust exemption so it doesn’t kill everyone????

Today on Decoder, we’ve got the first of a two-part series on the future of business, and I’m talking with Jonathan Kanter, the former antitrust chief for the US Department of Justice in the Biden administration. These days, he’s both a professor of law at WashU and professor of technology policy at Carnegie Mellon. The biggest story in tech right now is the spiraling debate about AI safety and regulation. Researchers at the big AI labs including Anthropic and Google DeepMind have quit in noisy ways, saying the models pose real threats and safety isn’t being taken seriously across the industry. Other researchers have said the chance of AI killing us all is greater than 10 percent, and the CEOs of all these companies have issued various calls to slow down development and develop regulation, including asking for antitrust exemptions so they can all coordinate on safety issues. That is a lot of ideas, and I am especially curious about the antitrust piece, which has led to accusations that these companies are seeking regulatory capture, attempting to form a cartel, and even somehow finding a way out from investor pressure ahead of their IPOs. Verge subscribers, don’t forget you get exclusive access to ad-free Decoder wherever you get your podcasts. Head here. Not a subscriber? You can sign up here. So to make sense of it all, I called Jonathan Kanter. Jonathan brought huge antitrust cases against Google, Apple, and Ticketmaster during his time as antitrust chief, and along with Lina Khan, led what amounted to a full reboot of American antitrust policy. As you’ll hear him say, he won those cases against Google and Ticketmaster, and that case against Apple is still live in the Trump DOJ, surprising many. That reboot has also led to some surprising alliances in the AI debate — avowed libertarian and former Trump AI czar David Sacks has been approvingly retweeting Lina Khan saying there’s no need for an antitrust exemption. This is all a lot, and you’ll hear Jonathan and me really get into the weeds of how all this regulation might work, and how it might interact with antitrust law, competition policy, and our relationship with China. Okay: Jonathan Kanter, former DOJ antitrust chief, on AI safety and regulation. Here we go. This interview has been lightly edited for length and clarity. Jonathan Kanter, you’re the former head of the DOJ Antitrust Division in the Biden administration and a current professor of law at WashU and professor of technology policy at Carnegie Mellon. Welcome back to Decoder. Always a pleasure to be back. This is great. You’re a private citizen now. You don’t have to give me the politician answers. I can say whatever I want. I feel so free. I’m very excited about this. I plan to as well. I have a million questions about the state of AI regulation in America right now. There is a swirl of ideas about how we should or should not regulate these companies. There’s the Trump administration whose posture seems to be “no,” which is fascinating. You rarely see the federal government just decline to participate in the way that the Trump administration appears to be declining to participate. And then there’s an incredible backbone of antitrust concern floating throughout all of this. Are these companies trying to form a cartel? Are they aware that maybe their IPOs aren’t going to go the way they want to, so they’re seeking regulatory capture in some way that is good or bad? Are they trying to foreclose cheaper competition from open-weight Chinese models? Give me your sense of just the state of things right now, and then I want to dive into the specifics with you. The state of the world right now is like we’ve invented cars and trucks, but we have no lines on the road, no traffic lights and no stop signs and no speed limits. We’ve invented this incredibly powerful technology that’s transforming how we live our lives, how we conduct business, how we socialize, how we seek and generate and consume information. We need some basic rules of the road. I think about it on two fronts. What are the things that companies should be doing? And what are things the government should be doing? There are responsibilities on both ends. I’m happy to dive into both. Let’s just start with the very notion that asking for regulation necessarily forms a cartel. All the big frontier AI companies — OpenAI, Anthropic, Google DeepMind, even Elon Musk — do not necessarily like each other. These are people that do not necessarily like each other. They are saying in concert that they should slow down. The phrase that Dario Amodei from Anthropic is using is to “pace the frontier.” My read of this is that they simply do not trust each other, that they are in a prisoner’s dilemma and they need an outside force, most likely the government, to enforce some kind of policy framework on them that they have to agree with. Is that a fair read of just the basics here? Kind of. Let me give you my interpretation from the most generous to the most cynical, neither of which, TL;DR, should result in any sort of antitrust exemption. The most generous interpretation is that they’re afraid of the pace of innovation and the lack of regulation means they don’t see any lines on the road. They don’t know where to drive and they’re worried about driving off the road or hitting somebody else. They’re begging for the government to step in and do something. Probably there’s a modicum of truth to that. I believe they believe it could destroy humanity. I’m not sure that that doomsday scenario is accurate, but I believe they believe that. That’s the more generous version. The more cynical version is that they’re spending so much money, they’re hemorrhaging cash and they can’t keep this up, but if they pull back, it’s going to affect their valuation as part of the IPO. So they would like somebody to give them all permission to stop spending so much money and slow the pace of innovation so that they don’t have to compete as hard so they can figure out their economics before they go public. Does an agreement among all the top players in a field like this automatically form a cartel? On that cynical side of the interpretation, there’s a lot of, “You just want regulatory capture. You want to make a cartel. You want to foreclose competition.” That is a risk certainly of that kind of arrangement. Let’s walk through what’s necessary and what’s not. Let me just be very clear, these companies do not need to coordinate in order to deliver safe and secure products to society. Think about this: imagine Boeing and Airbus. Boeing fears doors falling off the plane. It wasn’t Airbus’s fault and they didn’t need to slow down innovation with Airbus in order to prevent those doors from flying off the plane. They needed to figure out how to build planes that were safer. If you build cars that explode while you’re driving, it’s not the other car company’s fault and you don’t need to come together and figure out how to solve those problems. You need to figure out what went wrong in your manufacturing process. If you create products that create AI agents that break into somebody else’s and hack somebody else’s technology, it is no different, in my mind, than having an employee go do that. You have a responsibility to build safe and secure products and companies should be held liable when their agents, whether those agents are digital and binary or whether those agents are AI or whether those agents are humans. If they start doing bad things, the company that employs them can and should be held responsible. Those are the basic rules of the road. There are areas where there is legitimate collaboration in the context of safety and security. For example, having a clearinghouse to properly share threats, malicious bots, or other things that you can create a repository for so that everyone can build safer products. Those are the kinds of collaboration that happen in other industries and you don’t need an antitrust exemption because the antitrust laws don’t prohibit that kind of work. If two companies say, “We’re competing too hard and we need to slow down,” yes, that could implicate the antitrust laws, but that’s not what we should be doing here. We should be innovating, but companies should have the responsibility to build safe and secure products and the government should be responsible for putting lines on the road that say, “If you don’t build a safe and secure product, here’s what the consequences are. If you don’t build a safe and secure product, here’s what liability might look like. If you don’t build a safe and secure product, then perhaps there should be robot jail for your agents and it should be taken off the market.” I want to dig into robot jail specifically. in the amount of pressure we’re putting on what feels like product liability law. There’s something there that I want to talk about more specifically, but my wife’s a lawyer. I told her I was talking to you this morning and I tried to explain that there are fault lines here and she just shot me that look and said, “This is just a nightmare law school hypothetical.” What you’re describing is, “Does tort law create a regulatory framework for these companies?” And somewhere in there is, “If my agent goes and hacks you, my directive is that Anthropic’s at fault. Are we going to create some kind of pass-through liability? Does that happen already or not? Do we need another law or clarification law?” What’s your view on this? If I direct my Meta Muse agent to go attack you and it does it, is that Meta’s fault? Both/and. Companies employ people and they build technology. When those people and those technologies do things that are problematic, including stealing or breaking into someone else’s property on behalf of their employer, there’s liability and that can be products liability, but it could be more than products liability. You’re just people doing bad things or companies creating technology that’s doing bad things. And the products liability framework actually did work most recently in the case against Meta for Facebook and Instagram and child safety and security and mental health. Wait, hold on. That is really fascinating to me. This is a thing I’m really interested in talking about, because it took a decade for it to work. That’s the problem! This is where the government needs to step in. Companies have an obligation today to build safe and secure products and the pace of innovation isn’t an excuse not to go do that. Let’s be very clear about that. At the same time, we’ve invented cars and trucks and trains and we have no crossings, we have no lights, we have no lines on the road. Congress is completely incapable of pretty much doing anything. While these tech companies are out there saying that they want regulation, they’re actively lobbying against it and have been for a very long time. The whole thing is a complete mess and it’s a symptom of our system, which is a complete mess. And frankly, if the political scientist in me would trace that back to things like gerrymandering and Citizens United where our process has gotten out of control and Congress, which should be doing its basic job of putting basic safety standards and responsibilities in place, hasn’t done that. That having been said, the lack of action by Congress is not an excuse to deliver dangerous products. If you believe your product is going to destroy humanity, then don’t build it. If you believe your product is going to break into someone else’s stuff, then go back to the drawing board and figure out how you build it in a way that’s safer. Guess what? That’s innovation. These are innovation problems, not just regulatory problems. We need to do a better job incentivizing our companies to build safer and secure products. When they do that, those products will perform better and we will see better innovation. Right now, innovation is asymmetric because it’s done without the proper regard for safety and security, just like we saw in social media. As a result, now we have products that are really good and really effective at letting people communicate, but really bad when it comes to childhood addiction and mental health. Perhaps we should learn our lesson and encourage people to build safer, more secure products from the get-go, at the outset of this technology. Let me make the comparison to social media regulation a little more tightly. We just had the big Meta case; that is a products liability case. Their products were designed in a way that caused harm to teenagers. Meta settled the case because I think Mark Zuckerberg would prefer to build AI than to care about moderating Facebook. But they did lose one to a jury. They usually lose one to a jury, and this case wasn’t going well, but fundamentally they settled. They gave up. There’s a view that if they had fought and appealed, they would have won on some first amendment grounds, that the product liability case maybe wasn’t as strong as we all thought it was. And the other reason I’m saying that is Meta gave in to the regulator and their posture at the end was a settlement. They gave in and they accepted some regulation. It took a long time for the product liability version of this to get a company like Meta to concede some regulation, to concede some controls— To get a company like Meta to concede some regulation, to concede some controls. If we’re saying today, “OpenAI, you are probably liable for what ChatGPT does. Anthropic, you are probably liable for what Claude does,” isn’t this the moment to actually tighten the screw and pass the regulation that says, “You are definitely liable”? Yes. This is the job of Congress, which is to clarify the law. But I think the environment today is noticeably different than the environment was 15–20 years ago. 15–20 years ago, we had Section 230, which actually granted massive amounts of immunity to online companies, which actually made it harder to sue them and gave them more flexibility to innovate without regard for consequences. We had — and this is not partisan because it can go back to the Obama administration — a very hands off, “Oh my goodness, our economy depends on these wonderful companies from Silicon Valley. Let’s let them do whatever they want and essentially allow them to set up camp at the White House and in Congress and tell us how to write or not write our rules.” The results of that have eroded or destroyed any trust the public has in these technology companies. One of the reasons we’re having this conversation a lot earlier in the process now is because people are saying, “Fool me once, shame on you. Fool me twice, shame on me.” The good news is there’s greater public awareness and appreciation for the problems and a desire to do something meaningful. The problem is we have a system in Congress because of things like Citizens United, because of things like gerrymandering that have a government that is not responsive to the needs of the people. That i