The Political Scene | The New Yorker · · Transcript & summary by WhipScribe
Jill Lepore explains why voters rejected data centers and how a 1930s movement shaped today's tech elite, with a clear case for Congress to act.
Tap a chapter to play it from that moment in the show's own player; the transcript opens there too.
“I think people have a new awareness of that, of the scale of that political power, and therefore, rightly, a greater anxiety about it.” Jill Lepore · at 0:56 — ▶ hear it from here
“Sometimes regulation promotes innovation.” Jill Lepore · at 35:36 — ▶ hear it from here
Jill Lepore argues Silicon Valley has made the same promise for four decades: each new technology, from Apple's 1984 Macintosh ad to the internet, Facebook and generative AI, would strengthen democracy. Each time, the same tools also enabled surveillance, propaganda and digital persuasion. This year voters pushed back in a new way. Data centers became a ballot issue, with roughly 40 of 41 local ballot initiatives going against them. Lepore ties that backlash to environmental worries, a sense of powerlessness, and the decline of local news.
Her book, The Rise and Fall of the Artificial State, describes digital infrastructure that replaces the workings of government rather than supporting it. She contrasts the 1950 Census mainframe with Twitter's 2014 elections handbook, which called the platform a town square even though fewer than one in five Americans had an account. She traces the idea through science fiction, including E.M. Forster's 1909 story The Machine Stops, and the 1930s technocracy movement, led in Canada by Joshua Haldeman, Elon Musk's grandfather, though Lepore stresses Musk is not responsible for his grandfather's views.
Lepore says what comes next depends on whether Congress does real AI oversight, which it has largely skipped while the EU passed its AI Act in 2024. She argues AI agents fit neither legal category of person or thing, and that lawmakers have accepted the claim that regulation stifles innovation, even though 1920s radio thrived under federal planning. She calls the midterms a turning point because voters checked tech power even though Congress has not, despite Elon Musk spending close to half a billion dollars on candidates.
Jill Lepore is a staff writer for The New Yorker and a Pulitzer Prize-winning historian who teaches at Harvard. Her latest book is The Rise and Fall of the Artificial State.
Jill Lepore said voters checking Elon Musk's political money was a real turning point. Musk spent close to half a billion dollars trying to get his candidates elected, which she called pocket change for him. She argued that Congress had not checked tech power, but voters did in this election.
“That to me is a really significant turning point.”Jill Lepore · 44:40
Jill Lepore said the technocracy movement of the 1930s, led in Canada by Joshua Haldeman, held that democracy could not handle the modern industrial economy and called for rule by scientists and engineers. She said its pamphlets read almost like the techno-utopian manifestos that come out of Silicon Valley today.
“they're just incredibly similar to the techno-utopian manifestos that come out of Silicon Valley”Jill Lepore · 25:03
Jill Lepore said Twitter produced an elections handbook in 2014 that called the platform the new town square for candidates and officeholders. She argued that the claim was weak, since fewer than one in five Americans then had a Twitter account and most political tweets came from a small minority of users.
“Twitter produced an elections handbook claiming that Twitter was the new town square”Jill Lepore · 15:41
Jill Lepore said Congress failed to educate itself about the Cambridge Analytica scandal and the social media issues around the 2016 election, when Mark Zuckerberg was called to testify. She argued that this failure is significantly worse in the AI era.
“The failure of members of Congress to even educate themselves about the technological matters at hand”Jill Lepore · 27:53
Jill Lepore argues that Silicon Valley has repeatedly promised that each new technology will be good for democracy, and that each promise has failed. She says voters rejected data centers in the 2026 midterms, and that Congress must do real oversight of AI rather than leaving regulation to other countries.
Jill Lepore said voters rejected data centers across the 2026 midterms, with roughly 40 of 41 local ballot initiatives on them going against the projects. She says the backlash combines environmental concerns, NIMBYism, a feeling of powerlessness over decisions made without community knowledge, and the decline of local news that left residents unaware of projects being built nearby. She says it is not a partisan issue.
Jill Lepore said Silicon Valley has promised since Apple's 1984 Macintosh Super Bowl ad that each new technology would be good for democracy, including the internet in the 1990s, Facebook and social media in the 2000s, and early generative AI. Each time, she argues, the promise proved false, because the same tools can be used for surveillance, propaganda and digital persuasion.
Jill Lepore's idea of the artificial state is a digital infrastructure that replaces the workings of the liberal nation state instead of aiding it. She contrasts the first commercial mainframe sold to the U.S. Census Bureau for the 1950 census with Twitter's 2014 elections handbook, which called the platform the new town square. She argues the artificial state is a creature of science fiction whose stories always collapse once machinery drains finite natural resources.
Jill Lepore said Joshua Haldeman led Technocracy Incorporated in Canada during the 1930s. The movement argued democracy could not handle the modern industrial economy and proposed rule by a small elite of scientists and engineers, with people given numbers instead of names. Canada declared the group illegal in 1940, and Haldeman later moved his family to South Africa, where Elon Musk was born. Lepore says Musk is not responsible for his grandfather's views, but the 1930s pamphlets resemble Silicon Valley manifestos.
Jill Lepore said whether the new Congress acts on AI depends on whether members educate themselves about the technology and weigh it against its economic effects. She expects scrutiny of the Jensen Huang scandal around NVIDIA and of the Trump administration's ties to tech CEOs. She notes that the main AI regulation so far, the EU's AI Act passed in 2024, has been hard to implement, and that Congress has accepted tech executives' promise that deregulation would help growth.
In a recent piece for The New Yorker, Jill Lepore argued that AI is above the law and that Congress should fix that. Her core problem is that law recognizes only persons and things, so an AI agent that commits a crime while shopping for a car fits neither category. She also describes a pacing problem, where technology changes faster than law. She argues regulation can promote innovation, pointing to airplanes and to the radio conferences Herbert Hoover held as Commerce Secretary in the 1920s.
Jill Lepore said Elon Musk spent close to half a billion dollars trying to get his candidates elected in the 2026 midterms, which she called pocket change for him. She argues that voters rejected that kind of spending in a way Congress has not, making the election a significant turning point. She says Silicon Valley's power has not been impeded, but voters have checked it.
Our own transcript of the public episode, with a timestamp about every 90 seconds. Tap a timestamp to play the episode from there in the show's own player; the link button on a paragraph copies a link that opens this page there.
0:00Hi, Jill. Morning. Thank you so much for being here. Hey, good morning. Back in August, long before a single vote was cast, you characterized the midterms as the first AI election. What did you mean by that? Like AI as a campaign tool or a method of interference or as an issue that was driving people to the polls or all of those things? All of the above. I mean, recall ChatGPT was released in 2022. I don't think that it had established itself in ways that would have informed elections, certainly not along all of those different vectors by 2024. And in that piece, I was making that observation globally, really, because all around the world, this question of data centers has affected elections. The deepfake AI campaign
0:56ads have affected elections. There's been evidence, a lot of evidence of people asking chatbots who to vote for. You know, I'm a social conservative, and I want to know how to vote on the following ballot measures that I don't understand, kind of question that you prompt you might put into a chatbot. And then I just think there's sort of broadly anxiety about the tremendous and still rising political influence of Silicon Valley, of a small number of CEOs from Silicon Valley on local and national politics around the world, I think people have a new awareness of that, of the scale of that political power, and therefore, rightly, a greater anxiety about it. That's Jill Lepore, a staff writer at The New Yorker and a Pulitzer Prize-winning historian
1:42whose work explores American democracy, technology, and power. For years, the tech titans of Silicon Valley have been telling us that AI is going to change everything. It'll transform the economy, remake the nature of work, and help America beat China in a new technological race. But that vision is running into some resistance. Last week, Democrats won back Congress, in large part because of a public backlash to data centers, even as some of the most powerful people in the tech world, like Elon Musk, poured nearly half a billion dollars into midterms races. Then, just days later, a scandal erupted around NVIDIA's CEO, Jensen Huang. It's a reminder of how much the AI industry rests on the actions
2:22of a few extraordinarily wealthy men, and how far their decisions reach from the economy to our politics to our national security. Jill's latest book is called The Rise and Fall of the Artificial State. It's about how technology has eroded American democracy. And part of it argues that the ideas driving Elon Musk and the other tech bros from a political movement of the 1930s to the science fiction they grew up on are anything but new. I wanted to talk to Jill about the role that AI and backlash to AI played in the recent election, what tech billionaires might do next, and whether the new Congress has any hope of reining them in. This is The Political Scene. I'm Tyler Foggett, and I'm a senior editor at The New Yorker.
3:04It seems like every generation sees new technology that changes the way that voters get information, as well as form opinions and make decisions at the ballot box. And so I guess I just keep wondering what's actually new about our situation, even though everything about it also feels unprecedented. I do think that there is an emergence of rhetoric about the relationship between democracy and technology that comes out starting really in the 1980s, and we're still in a pattern that is established then. I think a lot about the famous Super Bowl ad that Apple put out in 1984, which was a fantastic ad. It's like really amazing. I show it in class all the time. So it's sort of this, you know, Orwellian Big Brother 1984
3:52totalitarian state. People dressed in shabby gray rags are marching in a gray sort of imprisoned world into an auditorium to watch on a giant screen Big Brother intoning how they shall live. And then it's a sort of American Olympian athlete comes running into the auditorium with a giant hammer and smashes the screen and the voiceover says, On January 24th, Apple Computer will introduce Macintosh. And you'll see why 1984 won't be like 1984. That promise that a new iteration in the computer revolution will halt the advance of totalitarianism and will always be good for democracy is one that is taken up as sort of a marketing strategy in Silicon Valley from that moment forward. And when the internet
4:52is open to the public in the 90s, the same promise is made all over again, right, that the internet is going to make the world a better place and good for democracy. The same promise, you know, we associate with Mark Zuckerberg and the release of Facebook and the 20 aughts. Social media will make the world a better place and specifically it will be good for democracy. And you saw with the release of early generative AI, that same bid being made again, this technology will be good for democracy and it will make the world a better place. But it's sort of like fourth time doesn't really fly. Because it wasn't true in any of those other cases. As much potential as there is in any of those tools to advance good citizenship, to empower the people, all those
5:31tools also contain within them the prospects of advancing authoritarianism, right? They can be used for surveillance and for propaganda, for digital persuasion, what people call computational propaganda or digital authoritarianism. In other words, I think we've been living for decades now with a world in which technology companies found it useful and sometimes sincerely to claim that their tools were going to be good for democracy and good for electoral competition, good for participation, representation, deliberation. And in each of those instances that promise has turned out to be a false one and why i think this has been the first ai election is that people weren't sort of voting on personal computers in their promise right people didn't
6:18really vote on the internet we didn't have really much of a say in the 1996 telecommunications act i mean we elected people to Congress. People didn't like recall members of Congress who voted for that piece of legislation after it passed. There was no punishment for that. It really wasn't part of our political discourse. But in this election, you really did see candidates who had supported data centers were elected out of office. And those who ran on a promise to usher in an era of some kind of restraint or at least review or regulation, or in some cases a full-on moratorium on data centers, kind of just won all over the place, right? I think almost entirely. Yeah, I think it was 40 out of 41 local ballot initiatives sort of ended in voters citing against data centers.
7:07So the numbers are just staggering. I mean, have there been other moments in history when voters delivered this kind of rebuke to an industry? I think, you know, this is quite distinctive. And it's important to remember that during the era of electrification, you know, in the 1910s and 1920s, rural farmers were desperate for electrification, getting telephones out to the countryside, even railroads. You know, people didn't want railroads to—they didn't want railroad companies to seize their land or the state to seize it by eminent domain. But cities vied for the railroad coming to their cities. People were desperate for the railroad to come through. People were desperate for the radio in the 1920s, you know, to have broadcast stations within the power of voters to make these decisions, that these are not, that decisions about data centers are not to be made by mayors or aldermen and women, you know,
8:10in backdoor deals. Yeah, I mean, there's so many ways to take the votes against data centers and kind of use them to sort of represent a larger movement. Like, there's the resistance to technology, which is pretty clear. There's also just like this fight over economic power with data centers being a very tangible thing that citizens can try to put a stop to. And so I guess I'm wondering, like, how much do you think we should actually make of these outcomes in the elections and how much are data centers a stand in for something big versus something really small. Yeah, I think we won't know that yet. And I think certainly candidates and officeholders have been surprised by the resistance to data centers
8:52and also by a kind of broader rejection of the next iteration of this technological moment, which is the coming of the kind of humanoid robots or physical AI. You know, you see stories about people taking baseball bats to DoorDash robots, you know, people sabotaging Waymo cars in cities. There is a sort of rising tide of a kind of fury at tech that I think is really destabilizing that's related to, but certainly not the same as the data center pushback. I mean, last summer, I was finishing writing this book, The Rise and Fall of the Artificial State, and I was really trying to figure out what to say in the epilogue. And I thought, you know, I really think there's going to be a pushback against the data center build out once people
9:38realize we're talking about these places, they're just giant. And I kind of wrote it into my epilogue, figuring out by the time the book comes out, maybe that would have taken giant. And I kind of wrote it into my epilogue, figuring out by the time the book comes out, maybe that would have taken place. And it had, right? It did sort of seem, if you were paying attention, that opposition to data centers would rise. And honestly, I think it is all of those things. It's environmental concerns. It's nimbyism. It's just a kind of sense of powerlessness because these decisions were, you know, made without people in communities knowing about them. It's also a theory about the decline of local news, right?
10:10The reason people didn't know about data centers being built in their backyards is because there are such a diminishment of local news over the last 20 years. this is a quite intangible thing, but I certainly sense it, a kind of yowl of frustration that people have found to voice exasperation with being tethered to machinery in a way that, you know, you have your iPhone in your pocket and you take it out and there's pictures of your kids and your dogs and it's cute and you get a text from your husband and there's some email from work that's disgruntling. But the phone itself is adorable and it's essential to you in so many ways. And yet you also feel that you have been hitched to something big, something very large
10:59and very powerful over which you have very little control. And so I think the materiality of the data centers, being able to picture this giant campus of, you know, warehouses for machinery is a kind of vindication of your suspicion, you know, that actually your phone is also the thing that is making your, you know, teenage daughter upset all the time. That is also the thing that is making your teenage daughter upset all the time. That is also the thing that is actually making your relationship to the people that you work with kind of poisonous. That is also the thing that means your ability to understand what's going on in the world politically has been impeded by distortion, right? So I think there's just a lot behind that
11:41that hasn't really quite found expression yet and doesn't really fit within our existing political arrangements. just a lot behind that that hasn't really quite found expression yet and doesn't really fit within our existing political arrangements. The parties don't have real platforms about, so surprisingly, right, this is not a partisan issue, which is a good thing, but it's a very fast emerging set of concerns that I think if there are powerful spokespeople to organize those concerns into meaningful dissent, we could see something really interesting happening and also something possibly quite threatening and terrifying. Let's take a quick break. When we come back, we'll get into Jill's idea of the artificial state
12:15and how a political movement from the 1930s helps explain Silicon Valley today. More with historian Jill Lepore after the break. This is the political scene from The New Yorker. Wired has always put a microscope on the people, power, and forces shaping our world. Uncanny Valley brings that same fearless reporting straight to your feed. Is Doge finally over? Will AI actually democratize American healthcare? Each week, Wired journalists from across the newsroom are going to unpack where politics, technology, and Silicon Valley collide. From conversations with tech leaders across Silicon Valley, internet fandom investigations, and government crackdowns on rigged gambling were taking you all over the news cycle, going straight inside the priorities,
13:09pressures, and power plays driving today's biggest decisions. Uncanny Valley tackles the questions keeping you up at night and helps make sense of the future taking shape right now. Listen to new episodes every Thursday, wherever you get your podcasts. the movie everybody's seeing this weekend, and we kind of go galaxy brand with it. Right, we take the ideas at the center of these things and trace them across history, across mediums, all to get to the bottom of why this particular text is important right now in politics and culture and also in your everyday life. We go highbrow, we go lowbrow, we go everything in between. Food culture to meme culture, Charlie XCX to Toni Morrison. K-pop to the Kennedys.
14:08From horror movies to heated rivalry. From the Oscars to the Met Gala. The Golden Bachelor to Wuthering Heights. Great expectations to desperate housewives. That's what we're talking about over here. We're having fun. It's Critics at Large from The New Yorker. Every Thursday, wherever you get your podcasts. From Vanity Fair, I'm Mark Aduchi, and I want to tell you about our new show, Fair Game. On every episode of Fair Game, I'll sit down with a leading character in our culture, from the worlds of Hollywood, politics, style, Silicon Valley, and beyond. Fair Game is a deep dive into some of the personalities that shape our world, and the stories that you need to hear. Think everyone and everything that you'd read about in the pages of Vanity Fair.
14:50So join me, Mark Aduchi, on YouTube, VanityFair.com, and wherever you listen to podcasts. It's all Fair Game. Your latest book is called The Rise and Fall of the Artificial State. Walk us through your idea of the artificial state. What exactly is that? Yeah, so I'm thinking about a digital infrastructure that replaces the workings of the liberal nation state rather than aiding it. There would be a difference then between the first commercially sold mainframe computer was sold to the U.S. Census Bureau and to tabulate the results of the 1950 census, right? That's not an artificial, like that's a digital infrastructure that is aiding the functioning of the state, right? We need a census. It's
15:41important to count. But if you think about the difference between that and, you know, in 2014, Twitter produced an elections handbook claiming that Twitter was the new town square and that since everyone was on Twitter, candidates running for office or elected officials holding office should use Twitter to consult with their constituents, to communicate with their constituents, to broadcast their own ideas in place of a town square. And this was at a time, just to give you some perspective, when I think fewer than one in five Americans had a Twitter account and 90% of all political tweets were tweeted by less than 10% of Twitter users, and anybody who tweeted about politics was politically extreme, generally more to the left than to the right.
16:31But in other words, there was nothing representative about the Twitterate. It had very little to do with the electorate, right? It's a minority of a minority of a minority, and then not even a representative minority. But there's considerable evidence that politicians and officeholders really did use Twitter in those ways, right? Did kind of use it as to get a sense of the pulse of the people and it informed their decisions and therefore contributed to our polarization. But from our vantage today in 2026, that era was kind of a golden age because Twitter or X, you know, the majority of traffic on X as on every other social media platform is bots. It's not humans. So you see this sort of slow replacement of the deliberative spaces in which we participate in our democracy. the basketball court sidelines, the PTA meeting, the city council meeting, the union hall,
17:27with an entirely artificial corporate-owned influence machine, right? So that all predates artificial intelligence, right? It's much worse in an era of artificial intelligence. But I think you see the kind of gradual seizure of power over the public square by corporations and citizens' slow surrender of their own political agency to that space and to the corporations that own that space and profit from it. That would be an example of what I mean when I talk about the half-built artificial state. So I guess just to clarify, like the title of your book, The Rise and Fall of the Artificial State, that title promises a fall. And you argue in the book that this fall is kind of inevitable. I'm wondering if you can explain that argument a little bit, why you think the artificial state
18:21is kind of destined to fall, and if you think the midterms were maybe the beginning of that fall, or if you're talking more about regulation and stuff that's going to have to happen at the congressional level. Yeah, so partly I'm talking about, narratively, the artificial state is destined to fall in that it itself is sort of a creature of science fiction. You know, for nearly two centuries, really since the beginnings of the Industrial Revolution, writers have imagined the possible future in which machines rule humans. It's sort of like an inevitable thought experiment once machines, once the Industrial Revolution gets going and machines are getting faster and stronger and faster and stronger and faster and stronger. And even by the 1820s, you have Charles Babbage thinking about machines that can do things like thinking. You know, by the 1870s,
19:09you have the telephone, which is a talking machine, right? So people are already thinking machines are one, they're going to be able to talk and think, and they're already stronger and more powerful than we are. One day they will rule us. And so if you read that body of science fiction that I think of as the parable of the artificial state, it always collapses. It's an inevitability, a narrative inevitability, and it has to do in science fiction with the machinery of the artificial state relying on the resources of the natural world, which are finite and not renewable. So ultimately, there's sort of one of two possible endings to every parable of an artificial state. Either people escape into the natural world and
19:52stop the machine before the natural world is essentially destroyed, or they don't escape in time and everybody dies. But either way, the artificial state collapses. I mean, my favorite version of this story is a story by E.M. Forster from 1909 called The Machine Stops. It's a short story and kind of a near future in which everyone lives sort of in their own little cells. You don't need to leave home for anything because the machine supplies everything to you. You have kind of screens. You can sort of FaceTime with your friends. There are screens for entertainment. There's a kind of 1909 version of DoorDash. All your food is delivered. You don't ever see family or friends except through screens.
20:36You never need to leave the room. And the last thing you would do would be go out into the natural world. It's a little bit like a precursor to, say, Silo. And you only have one book called The Book of the Machine, which is about how the machine will provide you with everything. I mean, it's a totalitarian state that is also a machine state. And the way the story unfolds is our main character's son keeps warning her, we have to leave because the machine is going to stop, because I went out and you could see the natural world is being destroyed and our main character fails to leave and so everybody dies. But so in the context of the book, right, I'm partly drawing from the world of fiction that anticipates, which I think actually, you know, people, it's a commonplace now to
21:20say we're living in a science fiction future, which is a good reason to pay attention to what the next chapter of that science fiction future is. Yeah. I mean, I feel like we have so many tech founders who were explicitly influenced by science fiction that it's actually, we're at a point where it's like a useful source text. I wonder if you can talk a little bit more about the roots of the artificial state or just the things that kind of helped build to this moment. Like I'm thinking about some of what you've written about with the technocracy movement of the 1930s, which Elon Musk's grandfather led in Canada. I mean, I guess maybe it makes sense to talk about technocracy and how we got from there to here. Think about the beginning of the Great
21:57Depression, right? The stock market crash in 1929. The kind of suffering across the United States and of course across much of the world is really unfathomable to us today and it led to very sincere discussions of whether or not democracy was a system of government that could address the challenges of the modern industrial economy and strongmen all over the world said no it's it's not. We need to have authoritarian rule. And that's among the reasons that authoritarians came to power. That discussion took place in the United States as well, where some Americans were drawn to communism, some were drawn to fascism, which had quite a bit of strength in the United States. Many people, of course, defended democracy
22:43and exerted their efforts in the defense of democracy, heroically so. But there was also a lesser-known movement called Technocracy Incorporated that had really millions of supporters in the United States and Canada. Technocrats believed that democracy was indeed incapable of handling the challenges of a modern industrial economy, And indeed, all existing systems of politics and the nation state itself were unable to rise to those challenges. And that in lieu of a democracy or a nation state, people should live in a technate and be ruled by a small elite of scientists and engineers. People would have numbers instead of names. They would have something like a universal basic income. Their exertions would be measured in a unit of energy and they
23:30would be compensated in a unit of energy. There would be no elections. There would be no currency system. There would be no banks. You know, there were technocracy mass meetings, places like the Hollywood Bowl that drew, you know, tens of thousands of people. They dressed in identical gray uniforms and drove in identical gray cars in what was known as the gray fleet. And it fell apart in the U.S. after FDR was elected on the promise of a new deal, right? FDR was promising a third way, a compromise between a wholly state-regulated economy and a laissez-faire economy. Of course, indeed, the New Deal was about the federal government taking on new powers that have been contested ever since. So the following for technocracy fell off almost immediately after FDR's election in
24:161932, but it continued in Canada under the leadership of its national chairman, an amateur aviator who actually had been born in the U.S. named Joshua Haldeman. He was a true believer throughout the 1930s, even after Canada declares technocracy incorporated illegal in 1940 as subversive of the government. He's briefly put in prison for his advocacy of technocracy. He's denied entry to the United States in 1940 when he wants to come on a speaking tour here. And he eventually, over the course of the 1940s, after leading a failed anti-Semitic party, decides to move his family to South Africa after it declares the policy of apartheid in 1948. That's why Elon Musk is born in South Africa, because his grandfather had migrated there, where he becomes
25:03a very, very vocal proponent of the apartheid regime. So I think what you can see, I mean, it's kind of this sort of family history is just objectively interesting. I mean, that's a whole episode in itself. I mean, Haldeman died when Musk was a toddler, like he's not responsible for his grandfather's views. But there is a family likeness. And if you read the pamphlets of the technocracy movement from the 1930s, they're just incredibly similar to the techno-utopian manifestos that come out of Silicon Valley. They really are recycling ideas from the 1930s, which for me as a historian is striking. It's arresting. It's remarkable. Well, then what does history suggest might happen next? Because we have this recycling of ideas that failed. And yet, you know, despite
25:51some of the more positive outcomes of this most recent election, I feel like, you know, we're looking ahead to the anthropic IPO next week. It doesn't necessarily seem like Silicon Valley has significantly less power than it did before, even if there are more data center moratoriums. And so I guess I'm just wondering if there's any argument for the idea that history might repeat itself with technocracy or if we're just kind of in a new moment. And yeah, I'm just wondering if you can kind of draw on your expertise to predict what we might see next from, you know, the modern day version of that movement and just the tech elite as they face like an extreme public opposition? Yeah, I think, you know, a lot depends on the U.S. Congress.
26:32And one of the reasons that people around the world so closely followed the U.S. midterms was partly because they're eager for a verdict on Trump and the tariffs. And, you know, they have huge consequences, of course, around the world. But people are really desperate for a verdict on whether Congress was going to take any initiative with regard to these tech companies, because the regulation has all come from other countries, especially from the EU, passed the AI Act in 2024. That hasn't even really been implemented and has proven very difficult to implement. But the fact that Congress has said, again, kind of looking back at that pattern that begins in the 1980s with that Macintosh ad from 1984, you know, the promise that tech executives have made to the U.S. government is what we're building is going to be great for the economy.
27:21It's going to be great for economic growth. We make no promises about shared prosperity. Indeed, we're not living in an era of to be great for economic growth. It's not going to, we make no promises about shared prosperity. Indeed, we're not living in an era of shared prosperity, but economic growth. This would be great for economic growth. And we will say it's going to be good for democracy. And we ask you to trust us on that score. What you need to do is just not regulate our industry at all. And Congress has just said, sure, fine. It's great. It's been great for economic growth, great for economic growth. And to an absurd effect under AI, right, where there's been no congressional scrutiny whatsoever. And I think people look back at congressional inaction
27:53with regard to social media. I mean, even just thinking back to, I remember in 2015 with Brexit and 2016 with the Trump election and the Cambridge Analytica scandal and Zuckerberg was called to Congress and Warren Hatch sort of famously said, I don't even understand how do you make money? You know, and Zuckerberg says, you know, Senator, we sell ads. The failure of members of Congress to even educate themselves about the technological matters at hand, I think is significantly worse in the AI era. So in terms of what happens next, it really depends whether Congress's, whether new members of Congress and sitting members of Congress do the work to think about what the implications of this technology or this set of
28:32technologies are and to measure those against the effects on the economy, which are not all necessarily positive, right? There's still a great deal of talk about how vulnerable the economy is to these stocks tumbling, to these companies falling apart. You know, the Jensen Wang story is a cautionary tale about that kind of vulnerability as well, right, to foreign meddling. Absolutely. The lack of scrutiny, kind of Trump waving that along. Oh, we're not going to, we're not even going to talk about that. That's not going to be a concern. It's a new moment. Well, let's take a quick break. And then when we come back, I want to talk exactly about that. You know, Democrats are about to take control of Congress.
29:09Will they reign in AI? Can they? And what might that look like? So we'll have more with Jill Lepore after the break. This is the political scene from The New Yorker. Thank you. that shape our world and the stories that you need to hear. Think everyone and everything that you'd read about in the pages of Vanity Fair. So join me, Mark Aduchi, on YouTube, VanityFair.com, and wherever you listen to podcasts. It's all fair game. One day last year, police entered a mansion in L.A. County and discovered the couple that lived there had almost two dozen children. Nearly every one of them born through surrogates. But those surrogates had been deceived. I started crying. I was like, what the heck did I just get myself into?
30:20I'm Ava Kaufman. The Journey is a new series from In the Dark and The New Yorker. Find it now in the magazine, I've written dozens of cover stories and traveled the world to talk to and write about some of the most interesting people in the GQ universe. From Martin Scorsese to George Clooney and Brad Pitt to outcast Andre 3000. And now I'll be doing it weekly. I'm going to sit down with some of the most interesting people in the world and figure out what makes them, well, them. New episodes of One More Question will be available each week wherever you get podcasts, or on YouTube if seeing people in chairs is more your thing. So Jill, you recently wrote a piece for the magazine where you asked if AI is above the law.
31:28And your answer was yes. And you argued that it's Congress's job to fix that. And I want to talk a little bit about the terminology you used in this piece because I found it so interesting. I think that the word you used for these machines after describing the Hugging Face incident was outlaws, which is something that, of course, refers to people normally. And so just how you're thinking about these machines and how we should go about regulating them. You also make a really interesting connection in that piece by invoking slavery laws. And I'm wondering if you could talk a little bit about that and how that could be instructive. Yeah, I think it's a category problem with AI, right? Who's responsible for what AI does has been a question.
32:08Those lawsuits are underway and will continue. My chatbot convinced me to take the wrong medication, and I'm permanently disfigured due to it kind of case, and some with much more dire outcomes. I don't mean to be glib about that. But one of the ways in which it's a category problem has to do with AI agents, right? Bots that are doing your bidding, your own personal bidding, right? Your personal agent. If you send an AI agent out to shop for a new car and it commits a crime while doing that, are you responsible? Is the software company that's responsible or the agent responsible? You can't hold the AI agent responsible because it's not a person under the law. Things in the law are either persons or things, right? Humans are
32:58natural persons. Corporations are artificial persons. Your computer, your laptop is a thing. Your piece of software is a thing. There are some scholars more like activists and tech files who think some AIs should be considered persons, persons who bear rights even. So there's that whole morass. But just in general, there's a kind of a higher level problem, which is the law likes to move slowly. You want laws to be stable. You want them to be predictable and knowable. If laws keep changing, people have a hard time knowing what's legal and what's not legal. Like it's not fair to people, right? And law can overreact to change. So you want to have broad laws whose interpretation can be flexible. And technology
33:48moves very quickly. So legal scholars just talk about this as a pacing problem, right? The technological change always outpaces legal change. So you could say, okay, well, Congress hasn't acted on AI yet. That's appropriate. We haven't really begun to get our arms around what AI is and what its legal implications might be, and it's going to require a lot of careful study. And in a way, that makes sense. But there's also a preference in the law to understand all new technology as essentially the same, but only faster than earlier technology. So what's the railroad? It's a faster horse. What's the automobile? It's a faster horse. What's AI? It's a faster, better search engine. So you should take the existing body of law that you have and just find a way to use it to
34:38interpret a new technology rather than writing a new law, which would be confusing, which might be overreacting, which would be hard for people to know about. I think that caution in the law has been conflated with a political campaign by tech companies to insist that regulation would stifle technological change and innovation and stifle economic growth so that there's a kind of capitulation on the part of lawmakers and judges, too, to say, well, we shouldn't actually put roadblocks here because these technologies need to grow as quickly as they can, as opposed to just saying the law needs to be conservative and move slowly. It's sort of like a deference to the idea of innovation. And I think that, I mean, one of the things I was trying to make a claim about in that article is that the sort of refrain regulation stifles innovation is not some timeless truth.
35:36Like sometimes that's true and sometimes that's not true. Sometimes regulation promotes innovation. Look at airplanes. Or the radio, right? In the 1920s, there were all kinds of regulations about radio broadcasting that made it possible for radio to really thrive and grow. It was the result of really careful planning and study on the part of Herbert Hoover, who was Secretary of Commerce for four years. Every year he had a radio conference at the White House just to bring in people who knew about radio to say, I think this is going to be really important. How do we want this to be set up in a way that people can really take advantage of this great tool and not be taken advantage of by it? There just really hasn't been anything like that.
36:12For the internet, for social media, it's all kind of after the fact adjustment. Like, that's why Met is dealing with all these lawsuits, right? That's why Anthropic is dealing with all these copyright lawsuits. It's all kind of after the fact. And I don't think it needs to be that way. The fact that it is after the fact is a consequence of Congress deferring to Silicon Valley rather than doing its job. Yeah, these like ex post remedies, right? Seems like that's the kind of the norm. I mean, it seems like part of the thing that makes the AI regulation in particular so difficult is just, as you mentioned earlier, this technology race that we're in with China. And I guess like I'm wondering if you can
36:52talk a little bit about how global technology races in the past have affected or limited democratic oversight. Like if there's anything to take away from like Sputnik and the Cold War, like how have we dealt with this kind of thing in the past when we're like, you know, actively competing with another nation? There was a great piece on The New Yorker a month or so ago arguing that there really is, that the race is itself a fiction. I think it was Gideon Lewis Crowe's piece, that it's a political fiction that's useful for all parties, the parties being the U.S. and China. And I think that's a really interesting claim. The easiest analogy historically is the arms race, the Cold War arms race. And, you know, remembering that John F. Kennedy ran for president in 1960, claiming that Eisenhower was responsible for a missile gap. And this was a consequence of the panic over Sputnik, right? So Russia launched the Sputnik in 1957. Eisenhower, in reaction, establishes NASA to try to generate research that will allow the U.S. to enter the space race, which was, of course, always an arms race.
37:58It was about getting missiles that could travel greater distances. Later research has shown that the so-called missile gap really never existed. It really was a campaign idea that proved very useful for Kennedy. And we lived with an arms race. We still live in the shadow of that arms race. But that era also saw the test ban treaties, nuclear nonproliferation agreements. That is to say that the State Department apparatus, the treaty apparatus around atomic weapons and the development of nuclear weapons was intact and worked mightily to reduce those risks. I'm not sure that it really is the best metaphor because everybody has AI in their kitchen, at their kitchen table. It's just a really different technology, right? It was sort of released
38:50into the wild without any real moral or ethical or political reflection. And I do wonder whether people will look back at OpenAI's decision in 2022 to release chat GPT to the public in order to gather data to fuel the development of newer models and get people addicted to the idea that they could ask their computer what to make for dinner and what to name their new puppy and what to say to their kid if they didn't do their homework, that this kind of massive rollout to a universe of users was quite ill-judged. I really do wonder if that'll be how that'll look shortly, because the whole kind of rhetoric now is the genie is out of the bottle, everything is inevitable. But then, you know, you look at the public's reaction.
39:50I think it's, you know, well above 70% of Americans don't want data centers in their neighborhoods. It's not elicited the reaction that OpenAI hoped it would in the same way that the rollout of social media, which kind of generated an ever-growing user base, it did not lead immediately to moral and ethical reflection about whether that had been wise. Weirdly enough, I feel like the thing that might actually prompt that reflection is the release of the Luca Guadagnino movie Artificial, which is about the rise of open AI. I was able to see an early screening of it. And my immediate takeaway was like, oh, this is going to prompt like a massive conversation about AI safety. And I think for a lot of people who just like use chat GPT every day, it's going to be
40:35kind of their first sort of real opportunity to go back to the beginning of that story and kind of see what happened. Obviously, it's a, you know, movie based off of a true story that, you know, has, you know, a lot of fictional elements. But I wonder if it'll actually be like a mass kind of see what happened. Obviously, it's a, you know, movie based off of a true story that, you know, has, you know, a lot of fictional elements. But I wonder if it'll actually be like a mass kind of like pop culture moment like that, that does something. Is it about OpenAI board hesitation over Altman's recklessness? And is that what the movie, the plot, is it about that moment when the board says, you're out, we don't trust you?
41:05That's in the movie. But yeah, I mean, this isn't even much of a spoiler, but it's basically like the social network, but for open AI. And so it like it's from the very beginnings and sort of like, you know, all these people that create this thing, and they're so excited about it. And then some are more inclined than others to release it to the public. And even though they know that there's all these issues with it, and that it it's hallucinating and that it's having these negative effects. So it really does like trace the entire story. And then it does give, you know, sort of the waived away because immediately there was so much interest and potential profit. But I mean, along the lines of like kind of odd or sort of like unexpected things actually changing public opinion or even impacting regulation, I want to go back to Jensen Huang and NVIDIA just because it does seem like this is a scandal that might actually prompt more of the proactive AI oversight and regulation that we're looking for as opposed to just this kind of like after the fact people have been harmed.
42:15So let's find a way to now compensate them. I mean, what kind of effect do you think that the scandal might have when we look to Congress coming in with this Democratic majority and maybe actually passing some regulatory laws? Yeah, I mean, Congress's role is oversight. So I do expect there to be a lot of scrutiny into that and also more broadly into the relationship between the Trump administration and tech CEOs over the course of this second term. But I also think we've kind of turned on a dime from worshipping these figures as superheroes toward turning them into supervillains, right? So it just really was not that long ago that Elon Musk was on the cover of Rolling Stone as the real world Tony Stark.
43:00And, you know, Marc Andreessen was on the cover of Time magazine in A Golden Throne. And these boy wonders were worshipped And, you know, Marc Andreessen was on the cover of Time magazine in A Golden Throne. And these boy wonders were worshipped as the superheroes of the modern day. If something that was, of course, simultaneous with the real height of the Marvel cinematic universe and its cultural power, I think not coincidentally, but also the height of kind of like lifestyles of the rich and famous of the 90s and, you know, elevating someone like Donald Trump as someone we should pay attention to and admire. I'm just not so sure that the depiction of these guys as superheroes is that different from the depiction of them as supervillains. So I think, you know, you go from seeing Musk. Remember, he was on Saturday Night Live in 2021,
43:46and he did all these weird skits as Wario in a Mario skit. And it was so freaky, but it was also, you know, he was so elevated by that in his own way. And Grimes was the musical act that night. And now he's the subject of parodies on SNL, and the same thing will be true of Jensen Wang. But I'm just not sure that's different. It's the same elevation in the cultural firmament. It just kind of has a different cast to it. And I think until we make these people answerable to the laws, including to tax laws, and refuse to allow them to position themselves to be above the authority of any national government, right? One of the claims of these guys, really, again, like it's a handful of people, but that they hold in their hands the future of humanity itself,
44:40the future of the planet. And that rhetoric, right, like that's been coming out of like pop culture depictions of tech titans for decades now, right? And so they're not, they don't consider themselves to be really answerable to the U.S. Congress. That's kind of what the press has been saying about them, right? They're put in this very different place and they're happy to accrue that power. So I think the American voters check on that power, especially with Musk in Texas, you know, who spends something close to a half a billion dollars, admittedly pocket change for him, you know, trying to get his various candidates elected. That to me is a really significant turning point. That is a rejection.
45:22It's not that Silicon Valley didn't win in many ways in this election. It's not that its power has really been impeded. But Congress hasn't been a check on this power. But voters have in this election. I think that's really worth remarking upon as a turning point. Well, I am constantly looking for a, you know, an optimistic way to end these podcast conversations, just because we often talk about whether AI is going to kill us all and, you know, just what's going to happen to American democracy. And so I think I'm going to take advantage of that optimistic note that you just gave us and end there on the power of ordinary people to actually play a role in some of this. Even though we're not in Congress and we're not tech billionaires,
46:05you know, we can still go out and vote. And the results were pretty resounding. So thank you so much, Jill. I really appreciate the time. Yeah, thanks a lot, Tyler. Take care. Jill Lepore is a staff writer for The New Yorker and a professor at Harvard. You can find her latest piece at newyorker.com, and her latest book is The Rise and Fall of the Artificial State. This has been The Political Scene from The New Yorker. I'm Tyler Foggett. This episode is produced by Nora Ritchie, with mixing by Mike Kutchman, and engineering by Pran Bandy. Our executive producer is Stephen Valentino. Our theme music is by Alison Leighton Brown. Thanks so much for listening. We'll be back next Wednesday. From Vanity Fair, I'm Mark Aduchie, and I want to tell you about our new show, Fair Game. On every episode of Fair Game, I'll sit down with a leading character in our culture,
47:08from the worlds of Hollywood, politics, style, Silicon Valley, and beyond. Fair Game is a deep dive into some of the personalities that shape our world, and the stories that you need to hear. Think everyone and everything that you'd read about in the pages of Vanity Fair. So join me, Mark Aduchie, on YouTube, VanityFair.com, and wherever you listen to podcasts. It's all fair game. Hi, I'm Nicole Phelps, Global Fashion News and Features Director at Vogue and co-host of the Run Through podcast. Each week on the show, our listeners get an all-access pass into the world of Vogue. On Tuesdays, join me as I interview influential designers like Calvin Klein, Rachel Scott, and Simone Bellotti. On Thursdays, join Chloe Mao, head of editorial content at Vogue
47:57U.S., and Choma Nadi, British Vogue's head of editorial content, as they explore fashion through the lens of culture with guests like Doja Cat and Margot Robbie. Listen and watch The Run-Through with Vogue wherever you get your podcasts and Vogue's YouTube channel. One day last year, police entered a mansion in L.A. County and discovered the couple that lived there had almost two dozen children, nearly every one of them born through surrogates. But those surrogates had been deceived. I started crying. I was like, what the heck did I just get myself into? I'm Ava Kaufman. The Journey is a new series from In the Dark and The New Yorker. Find it now in the In the Dark podcast feed.

