CEOInterviews.AI
Start App
Get an email when Jill Lepore gives a new interview
One email per new interview. Unsubscribe in one click.
Jill Lepore
Professor of American History and Law; Staff Writer, The New Yorker, Harvard University

After Words: Jill Lepore on "The Rise and Fall of the Artificial State"

📅 Oct 24, 2026 C-SPAN 58 MIN 4258 VIEWS 83 SEGMENTS · 5 SPEAKERS
Harvard historian Jill Lepore explores how digital technology is re-shaping society. She was interviewed by WPVI-TV anchor Tamala Edwards at the Free Library of Philadelphia in September. Download the FREE C-SPAN Now App. https://www.c-span.org/c-spanNow/ Discover the C-SPAN Video Library at https://www.c-span.org/quickguide/ Explore C-SPAN's Free Educational Resources at https://www.c-span.org/classroom/ C-SPAN: Created by Cable in 1979. Offered as a public service. Support C-SPAN by Donating Today: https://donorbox.org/support-c-span?u... Subscribe to our YouTube channel:    / cspan...

What Jill Lepore said

Written from the verified transcript and checked against it. Every figure links to the moment it was said.

Jill Lepore discussed her book 'The Rise and Fall of the Artificial State', arguing that liberal democracy, founded in 1776, is fragile and being replaced by an 'artificial state' where privately owned bots dominate public discourse. She traced the history from the 1790 census through IBM's tabulating machines to modern campaign targeting, noting the first election where people ask chatbots how to vote. Lepore critiqued Silicon Valley's ideology, linking Elon Musk's grandfather's technocracy movement to current tech politics, and analyzed how science fiction like 'Hitchhiker's Guide to the Galaxy' is misread as instruction manuals. She expressed uncertainty about the danger's severity, compared it to early COVID, and offered modest advice like keeping phones out of bathrooms. She concluded that tools lack intrinsic political valence but working-class revolution won't be aided by chatbots, and emphasized personal responsibility in participating in these systems.

Key takeaways

  1. Liberal democracy is only 250 years old and fragile, being replaced by an artificial state where bots outnumber humans on platforms.
  2. Elon Musk's grandfather led the technocracy movement in the 1930s, and Musk's political ideas recycle its discredited ideology.
  3. Silicon Valley's 'California ideology' from the 1960s, combining libertarianism and counterculture, led to the 1996 Telecommunications Act.
  4. Lepore believes hope is cheap; what matters is determination, like keeping your phone out of the bathroom.

Numbers and commitments

FigureWhat it refers toTypeAt
250 years Age of liberal democracy since Declaration of Independence timeline 8:24
1790 Year of first federal census timeline 11:28
1890 Year Herman Hollerith's tabulating machine counted census timeline 12:54
1959 and 1960 Years Kennedy campaign used machine for targeting timeline 14:17
1940 Year Canada outlawed Technocracy movement timeline 23:00
1950 Year Musk's grandfather moved family to South Africa timeline 23:00
1996 Year Telecommunications Act signed timeline 28:50
2004 Year Facebook launched timeline 28:50
17 Number of Franklin children metric 40:38
24 Average age at marriage in Franklin family metric 40:38

Chapters

  1. 0:00Origin of the artificial state
  2. 10:20History of counting and campaigns
  3. 16:56Science fiction and Silicon Valley
  4. 21:35Musk's grandfather and technocracy
  5. 26:39California ideology and deregulation
  6. 30:57Data center backlash and AI president
  7. 34:26Animals, robots, and apocalypse
  8. 44:33Hope versus determination
  9. 46:42Class structure and tools
  10. 49:41Writing about the present

Questions asked in this interview

12
  1. 2:22And I also wonder because there's so much out there about AI, why did you want to do it and why did you say there's something to be done that hasn't been done yet?
  2. 10:20I'm going to start with the census, which goes back to a debate that stretches into the Bible: should you be able to count people and what happens when you do?
  3. 16:56So why would you look to those stories as a guiding light, Sam Altman or anybody else, Peter Thiel?
  4. 25:37What is it that you go from don't be evil to what is evil?
  5. 30:57Like, where does this end up?
  6. 34:26And as I read it, I mean, I literally thought to myself, what is in us?
  7. 36:07How do we stop, and will we do it in time?
  8. 39:47But I'm interested in knowing then, what does it mean to you to tell history in a way that's accountable to evidence, and what does this accountability look like for you when you perform this inquiry on a historical subject?
  9. 44:26A question from the center or this side, anybody?
  10. 46:36Do we have it? Okay, right there?
  11. 46:42So, can it be used by us, the working people, to build a better society?
  12. 49:13Where's one of those — Where's there any of the students who had a hand up up there?
Narrator 0:00 ↗
This fall, Democratic US Senate candidate Roy Cooper often starts his campaign events with this line.
Roy Cooper 0:07 ↗
And I had my campaign sent to me personally every text and email that my campaign sends out.
So, my back pocket is constantly buzzing with me asking myself for money, with me asking myself to volunteer for my campaign. I feel your pain.
Narrator 0:34 ↗
And also this one.
Roy Cooper 0:36 ↗
And one of the things I do sometimes, my campaign, I tell my campaign people, "Give me the list of people who have given $5, $10, $25, and I'll go down the list and I will call people." And they pick up the phone and we have to get by, "Is this really you? Is this AI?" And all of that.
Narrator 0:58 ↗
Robocalls, AI-generated campaign ads, polling models that churn through massive data sets.
Technology plays a huge role in both campaigns and governing. But is there a point at which government by the people, with the help of machines, becomes government by machine? That's the question posed by historian Jill Lepore in her book The Rise and Fall of the Artificial State. That's coming up on Afterwards.
This is Afterwards from C-SPAN. This week, a conversation on the role of technology and democracy with Harvard historian Jill Lepore. She was interviewed by WPVI-TV anchor Tamala Edwards at the Free Library of Philadelphia in September.
Tamala Edwards 1:45 ↗
Welcome, welcome, welcome. It is wonderful to see a full house tonight, but we do want the young folks from Temple University to stand up. Let us see you.
Truly, the owls are in the house. Guys, thank you for being here. Just so you know, my understanding is they get a point, they get some credit for coming up with a question and extra credit if they get to ask the question. So, we're going to let them get in some questions tonight. Get ready, guys.
Jill, it is wonderful to be with you here tonight. And I wanted to start with, as was pointed out in the introduction, you have done so many different kinds of essays and books looking at the interactions between Native Americans and the settlers that came to this land, the enslaved, all kinds of — Jane Franklin, you know, we think we love Ben Franklin, but did we realize the story? What took you to AI? And I also wonder because there's so much out there about AI, why did you want to do it and why did you say there's something to be done that hasn't been done yet?
Jill Lepore 2:59 ↗
Um, that's a great and kind of difficult question. Um, but I first just want to say it's such an honor to share a stage with you. Thank you. And thanks everyone for coming out to this beautiful, really special place. Um, I'm just — I wish I lived in Philly and could go to all these events that Daniel was just announcing to all of you. So, buy your tickets soon. I'm like, I'll buy a ticket. I'll come down.
Um, you know, I think I'm a little bit disciplined by working for a weekly magazine. It just — as you're asking me that question, I'm like, I think it's my job to kind of keep an eye out for what's going on and what, you know, for a magazine, you have to have a news peg. This was something that was really hard for me as a scholar to accept. I'd always be pitching pieces and I'll be like, you should do a piece about this fragment of hair found, you know, in a field in western Pennsylvania because it changes what we know about the 18th century or something. And my editor would be like, "Nobody — what's the news peg?" You know, what's going on in the world that we can use to bother to tell this story, to share the story with our readers. So, I think I've — I'm not really principally a journalist, but I have to think about what's newsy in a different way.
So, I think my beat has kind of changed where I really am trying to look for ways to explain the present by way of the past. So, often I'll do stories that are just, you know, I really wanted to write about Jane Franklin. There is no breaking news about Benjamin Franklin's sister. Um, but I have written a lot of pieces about the history of technology because I'm really interested in it. I also happen to be married to a computer scientist. So, all of our, you know, kitchen table conversation is about the latest whatever. Um, so, for someone who's not a computer scientist, I happen to kind of have a front seat to that stuff.
Uh, and I think that this particular project, which really began as an op-ed I was supposed to be writing for the New York Times and then it never ran in the New York Times. I felt like we were paying a lot of attention, um, after the ChatGPT release, which was just 2022, to just the quick uptake of that without looking back toward the uptake of social media as a cautionary tale that might inform how we think about the quick uptake of this new thing, or without looking back to the promises made about the opening of the internet in the 1990s. Oh, who was going to save the world, end all wars, bring peace and harmony, democracy would go around the world. Or that same claim that was made in the 1980s about the personal computer. The personal computer was going to bring democracy around the world, we'd all be happy, you know, prosperity, no one would ever have to work a day. Like the hype — we were on the fourth cycle of hype and it was like everybody had forgotten the other three.
So, just at a time when like even my kids' public school was like, "We think we're going to be banning social media in the classroom, but we're bringing in AI." Do you know what I mean? You're like, "But what?" In a few years you're just going to have to ban the AI because they won't have brains. Like it just — like the lack of a historical vantage on the ChatGPT moment really troubled me.
Um, so, I wrote this thing that was an opinion piece and then I made it a little bit longer, wrote it, it was published in The New Yorker, and then my book editor at the time — I was working on this book about the Constitution, We the People, and my book editor said, "Can you just really quickly turn that article about the artificial state into a very short book? We want to release it in like 2 months." And I was like, "I'm still finishing this book about the Constitution that also feels really urgent to me." Like that felt like a really urgent project. Um, so, I convinced him that I needed a few more than a couple months. Um, but yeah, it was that sense that we were kind of — we've lost our bearings in the present moment and there was probably something in offering up a history that would help us get our bearings.
Tamala Edwards 7:27 ↗
I mean, it really is like when you're watching a television show and somebody's got all the different plot points and you're like, "Oh, they're going to find the guy in the end." Like you've got all the different plot points of things and you're like, "Oh, it all leads here." I wanted you to start also by explaining — you talk about the liberal democracy as a state which you say actually started here in 1776. We don't realize how young it is, that it changed our understanding of what existed. And then when you talk about the artificial state, that's different than AI. You're talking about something else. Explain those two things.
Jill Lepore 7:59 ↗
Yeah, I did want to give a name to what I think we're kind of in or in the middle of becoming. Um, so I kind of got committed to this idea of this term, the artificial state. But what I mean by that, like all states are artificial. Like chimpanzees don't live in states, right? It's a human thing. We form governments. And it's different than having a society.
And one of the things that really affected me in working on this book about the US Constitution was just how fragile a liberal constitutional democracy is. I mean, I think that's a large part of our public discourse now. People think a lot about how fragile our constitutional system seems. Um, but part of its fragility is really its novelty. It really, you know, we're celebrating the 250th anniversary of the Declaration of Independence that declares as self-evident truths these principles on which our liberal democracies rest. And that's the first time they were declared in that way, right? So only 250 years ago.
Um, so what I wanted to call attention to was that it is not a permanent state. There's no reason to believe the liberal nation-state will last forever or much longer. But then the question is what is going to replace it? And it seems to me we're sort of well on the way of replacing it with an artificial state. And by that I mean a state in which, in lieu of government by consent, right? A government of the people, by the people, for the people. We're really building a government by machinery in which the infrastructure of our relationships with one another is a privately owned infrastructure. It's actually dominated not by humans but by bots. There are more bots online anywhere on any platform online than there are humans. And those bots are owned by private actors and they have motives that have nothing to do with our public sphere. And yet they're determining how we engage with one another.
So, I wanted to call attention both to the role of corporations in taking over the powers of elected representatives and nation states, and also to our own individual surrendering of our autonomy and our obligations as citizens to machinery.
Tamala Edwards 10:20 ↗
As I read the book, there were often things that I just sort of titled, "That wasn't on my bingo card." Because something would start in a place and you don't expect where it ends up. I'm going to start with the census, which goes back to a debate that stretches into the Bible: should you be able to count people and what happens when you do? And you write, "The building of the artificial state was set into motion by the development of machines that can count people, sort them into categories, turn them into numbers, and measure, predict, and influence their behavior." Way back when, we start, you know, it's the king trying to figure out how to tax people. Then we get to these massive computers that TV stations are like, "We might be able to predict the vote." Which as a news reporter and a political reporter, you're like, "That makes sense." Data operations by Romney and Obama. And I remember thinking, "This is sweet. They're getting to small voters. They're figuring out how to target their message." But then it ends up in machines that seem to track our every breath and click. And we don't realize from the king counting sheep to being tracked all the time, there's a through line.
Jill Lepore 11:28 ↗
Yeah, I think it — for me, I need to see that there's a history. Like, in order to understand something, I actually need to look backwards. I'm just like — if I'm walking down the street, it's as if I'm walking backwards. I have to kind of always see where something came from. And I think in trying to lay out for readers how we got here, it was really important for me to say like, we got here innocently. Do you know what I mean? Like it's this — I don't mean like the deep state. It's not a conspiracy. It's not like there's a bunch of, you know, guys in a room who are like, "We're going to take over democracy." I mean, there are those guys. Let's be clear. There are those guys and I get to them. But like, you know, the idea that there should be a census, right? That's in the Constitution of 1787. The first federal census was done in 1790. We do one every 10 years. We were the first nation to declare that we needed to do one. Other nations followed.
And the reason that we count the people is so that we can be mathematically represented in the House of Representatives. This is why gerrymandering is a problem. Like the math becomes a problem if you don't do it fairly. But the idea that we should be counted, that's what democracy rests on. Like that's a great idea. We should be counted. Our interests should — how our political power should have some relationship to our numbers. And it should be proportionate to our numbers. And there can be other mechanisms and levers of power, but that who gets the most votes wins — that's counting. That's fine.
So that you would want to do that more efficiently, more accurately, more quickly, more cheaply, that also makes perfect sense. So you're going to — you know, IBM started with this guy Herman Hollerith who invented the first tabulating machine to count the census in like 18 — in the census of 1890. Um, and then when the general purpose computer is invented during the Second World War by the Allies, well, what are you going to do with that? Like when the war is over, we don't have to like track missiles, we hope. We don't have to crack codes, we hope. Oh, but we could use that for things like counting the census. It's a really powerful machine. So, the first commercial general purpose computer is built for the US Census Bureau. That seems perfectly reasonable.
Then, okay, well, we could use that to process public opinion polls. We could use that — CBS News is like, we're going to use one of these to predict the outcome of the election on election night. It'll be really cool. It'll make television night on election night more fun to watch because otherwise it was kind of boring. And then, you know, by 1958 these ad men in Madison Avenue get the idea, well, we could use this — if we put into a computer all the census data, all the public opinion data, all the election data, and we know a lot about voting behavior, we could make a machine that could tell a candidate what to say and to whom and tell them what their reaction would be.
So, they sell their services to the John F. Kennedy campaign in 1959 and 1960, and the Kennedy campaign asks the machine, you know, if I talk more about my Catholicism, what will the effect be on Detroit housewives who are Protestants, you know, and they get results back. They write reports and it influences the campaign. And you'd think like, well, that seems reasonable. Like it's just more efficient, it's cheap, it's more accurate, it's an interesting way.
Um, and we are in a very different place now, right, where political candidates — they know who you're going to vote for. They know what they're going to decide is whether to try to get you to vote or not, whether to try to get you out of your house. Like you're going to get specific messages, right? They're going to be very targeted to you and your demographic. But what I think is complicated about that, each step along that trajectory was perfectly reasonable, but at this point, like this is the first election where there's a lot of evidence that people are asking their chatbots how to vote.
And you're like, okay, well, there's so much information. If I ask OpenAI, I'll get a lot of information about the candidates and I say like, here are my political views. I'm a social conservative but economically — you can get really good information. But what I think we've not tallied in that whole trajectory is what we lose by no longer talking to one another about these things. Even if you think about those candidates, even think about the Kennedy campaign, the time that they didn't go to talk to housewives at playgrounds or PTA meetings in Detroit to kind of test out, well, what do you think about Kennedy's Catholicism? They didn't send their campaign workers out to have those conversations.
And now campaigns don't do that same kind of door knocking as they used to do. If you're asking Claude how to vote, you're not asking your next door neighbor. You're not going to a community meeting at your kid's high school. You're not asking someone at the basketball court. But the daily conversations that we have with one another as humans are what forges and maintains our bonds as a civil society. So every time you ask a machine to do something for you that you wouldn't otherwise have done by talking to another person, you're dismantling our polity.
Narrator 16:42 ↗
We'll get right back to Afterwards from C-SPAN. Now back to Afterwards from C-SPAN. Historian Jill Lepore is talking about her book The Rise and Fall of the Artificial State.
Tamala Edwards 16:56 ↗
There's also a circular logic that you write about in Silicon Valley. You know, I'm afraid of machines, so we need to build machines faster to save us from the machines that will inevitably kill us. Um, and you point out, you know, a lot of these fears go back into science fiction and all of these stories that are written in the 20th century that influenced many of the people — the names that we know now, a Musk, a Bezos, a Zuckerberg, so many different people you can draw the line back to a Lovecraft or an Asimov or somebody else. But I thought to myself, none of those stories end well. So why would you look to those stories as a guiding light, Sam Altman or anybody else, Peter Thiel?
Jill Lepore 17:38 ↗
Yeah, I got really fascinated by this some years ago when I was working on a project about Musk, whom I at the time, this was like in 2020, I knew very little about. Um, and I spent a lot of time watching interviews with Musk back to the mid-90s when he first, you know, made his first big paycheck. Um, and bought a DeLorean and was, you know, interviewed.
And he consistently from his 20s to his 50s talks about the influence of Douglas Adams' Hitchhiker's Guide to the Galaxy that really influenced him when he read it when he was like 11 or 12 or something. You know, he was a very familiar — very familiar to me kid, like, you know, the certain boy who goes from like dinosaurs to Legos to comic books to Dungeons and Dragons to girls or boys, but — and like not — they don't always make it the whole way.
Um, like, you know, he invented his first video game when he was 12. He's really into comic books. Um, he was really into science fiction. And, you know, most kids who have that — I was really into science fiction when I was a kid, and I still am, but like the world changes and they change as they grow up, but like he all through his adulthood would talk about this one book, how much it changed his life. And I was like, I know that Douglas Adams' Hitchhiker's Guide to the Galaxy like nearly by heart. I love that book. How can you possibly like the same book that I like?
Um, then what, you know, it was a radio play in the 1970s. I went back and listened to it again and again and again and it's always struck me that, you know, it's an indictment of vast wealth and it's an indictment of colonialism. And um, I went and looked up — I don't know how I even decided to do this. I was really interested in exactly trying to find out exactly when Adams had written it. And um, I found his typewriter was being auctioned, his Hermes manual typewriter, the typewriter which he had written the Hitchhiker's Guide. And on the side of it was a big sticker that said end apartheid.
And then it was clear to me, oh Douglas — Douglas was writing in the middle of the 1970s when um the UK was really involved in calling attention to the apartheid regime. And you can read all of Hitchhiker's Guide as an indictment of apartheid. It's about these luxury planet builders who just keep going around destroying other people's planets and taking other people's land to do whatever they want. Maybe they want a new planet that's slightly pinker than the last one that has two moons instead of three.
Um, and I became just fascinated. Musk grew up in South Africa. And it's not surprising that as a boy he wouldn't have understood this book as an indictment of colonialism or specifically of apartheid, but as a grown man that it doesn't communicate itself to him as a rejection of who he is, as an attack on who he is and what he represents. But for him it's somehow an instruction manual that you should go colonize other planets.
And it turns out that's the pattern with all of these guys. Like if then you spend some time reading what they talk about their favorite books and well, they have a very juvenile kind of reading habit. Like I don't know what they read now, but what they read was like — so then you think about like they're building our world. Like what they loved as kids were world building games like Dungeons and Dragons, and then they grew up to become world builders. Um, and it's a kind of literalness, um, but in a way it's a kind of rejection of the act of reading, which really bothered me. So the whole second half of the book is about how to read science fiction.
Tamala Edwards 21:35 ↗
I want to stay on Musk because there's an echo of history when it comes to him. I have read about Haldeman's grandfather in terms of race because his grandfather who starts out in Canada and eventually ends up in South Africa had some really virulent views. I hadn't realized the through line on tech. Um, and as I got to the end of that chapter, all the hairs on the back of my neck stood up. This is not a great through line.
Jill Lepore 22:00 ↗
Yeah, so Musk's grandfather is a really interesting guy named Joshua Haldeman who's a chiropractor. He was born in Minnesota and moved to Canada as a young man. He was a rodeo cowboy. He was an amateur aviator, a big adventurer. Uh, very colorful character. He lost his farm during the Great Depression. In the 1930s, he joined and then became the national leader of a movement called Technocracy Incorporated, which was huge in Canada and the US in the early 1930s. Like huge — hundreds of thousands of people. Um, tens of thousands of people would show up for these Technocracy events. These guys dressed in gray. They drove identical gray cars. They did not believe in democracy. They wanted scientists and engineers to take over the society. There would be no politics, no banks, no currency. In a way, kind of no jobs. It'd be something more like universal basic income. Um, people wouldn't have names. They would have numbers. It would be more efficient if we just had numbers.
Um, and it became a joke after it was revealed that the American leader of the movement was a total fraud. Um, but Haldeman continued to lead the movement in Canada until it was outlawed by Canada in 1940. He was briefly put in prison. He was denied entry to the US when he wanted to come here on a Technocracy speaking tour. Then he led an anti-Semitic party in Canada. And then in 1948 when South Africa announced the policy of apartheid, he became interested in immigrating to South Africa and brought his family there in 1950. And then he became a very ardent advocate of apartheid. Um, he was a conspiracy theorist. He was opposed to vaccination. He was an anti-communist and thought all universities were full of communists. Um, mainly he believed in an international Jewish conspiracy.
Um, and the virulence of his racism is even hard to describe. Um, but I managed to find — he's been really scrubbed from the historical record, but I managed to find in a collection of rare radical pamphlets in Michigan his unpublished writings. Um, a handful of them, which are just shocking. But what's really interesting to me — I mean, Musk's grandfather died when he was a toddler. I mean, he wasn't directly influenced by him. I don't even know how much he knows about the family history, but it does seem to be somewhat deliberately scrubbed.
Um, there's a couple things that's interesting. One is that he didn't have a big following in South Africa. Um, but if he were living in the age of Twitter, he would have had a huge following, right? He'd publish these rants and they would be everywhere. But the other thing, the thing that really struck me, is Musk really presents himself to the world as a futurist, you know, he's ushering — he's leading us into — he's bringing the light of human consciousness to the stars. Um, he's reinventing government, he's reinventing the energy infrastructure. Um, but his political ideas are the worst failed, most discredited ideas of the 20th century.
Why anyone should think that someone who is a great entrepreneur also has good political ideas, I don't know. Or someone who's a great engineer should also be determining the future of our politics. Um, but there's a very specific way in which the craziest part of the ideology of Silicon Valley really is a recycling of the technocracy movement of the 1930s.
Tamala Edwards 25:37 ↗
Even though you say Musk was a toddler, I mean, it's hard not to hear what you've heard 2 million times: history doesn't echo, but it rhymes. Like so many of these ideas, even if you never met the man, how and why anti-vaccine, so many different things echo — or I should say rhyme. You know, Silicon Valley for so often loved being held up as a place that loved to say don't be evil, move fast and break things. All these things that seemed cool and wonderful. You know, you talk about Marc Andreessen on the cover of Time magazine barefoot with a crown on his head laughing. Um, you know, Jeff Bezos the valedictorian of his class in New Mexico. Like these guys seem like good guys who were going to do good things. And then it turns. And you wonder, is there — what takes them through the looking glass? What is it? Is it the money? Is it the power? Is it really, you know, I've often felt that they really do feel they are that much smarter than everybody else. What is it that you go from don't be evil to what is evil?
Jill Lepore 26:39 ↗
That is nicely put. Um, I mean, there's been a lot of really fantastic historical research on the ideology of Silicon Valley and how it came to be. Um, people talk about it as the California ideology. It starts in the 1960s and it's this interesting marriage of libertarianism and counterculturalism. So, like I don't believe government should exist, but then I really like surfing. Um, sorry, that I like California. I'm not there to mean, but like there's a specific electronics and early computer industry version of that.
Um, that comes to associate building personal computers and the internet with like a radical individualism that's libertarian, but also emancipatory. Like in a kind of sexual revolution emancipatory. Like, you know, the first big story about the internet was from 1971 and it was in Rolling Stone. And it was like how cool this is going to be, we'll all be hooked up. And it's like really was associated with the drug culture of California. Like it really was a countercultural thing for a long, in spite of the fact that it's like a military project.
Um, so, by the time you get to the 1980s, um, there is more money, right? People are really making money. Um, and there's a kind of Milton Friedman economic libertarianism that's huge there. You know what no one can do? No government should stand in our way. We're building the most amazing things and you people in the government don't even understand what we're building, which is true. So, you know, regulation will stifle innovation. We have to be free.
Um, and you know what also? We're going to free the human spirit. Everyone's going to have their own personal computer. You do you. It's going to be awesome. And then with the internet, that kind of vibe really finds a purchase in an alliance with the resurging conservative movement. So, Newt Gingrich with his Contract with America...
America is really behind the 1996 Telecommunications Act that Bill Clinton signed, but that says the internet is basically beyond the realm of law. This guy John Perry Barlow, who was a very important counterculture figure, writes this Declaration of the Independence of Cyberspace. He's like, "You guys don't understand what we're building here. This cyberspace, it is lawless. It is beautiful because it is beyond your laws." And that's the internet that we get. They win that argument because they joined forces with this new conservative movement in Congress. And once they win that, they're like, "That's the way we win. We just have to be allowed to build what we want to build. The government should have nothing to do with it." And one way we will make this case is by arguing that whatever we're building is going to be good for democracy, because that's a good marketing campaign. And so they say that, and the same thing happens with social media, with Facebook in 2004. "Government should have nothing to do with this. This is going to be great for democracy. We're going to bring the world together. Everyone's going to be happy. It's going to promote democratic revolutions around the world." And also, Congress clearly doesn't even understand what social media is. So, like, we've — this is the, you know, this is the — that's why I say this is the fourth time, because now Congress, you know, Bernie Sanders is holding hearings next week about the data center moratorium idea and the possible AI takeover of the entire universe within the next month. And it's like, "How are we at that point?" But you really have to go back to the 1980s and see that first set of claims about — And it's not all hooey. Like, a personal computer is this amazing thing. Like, people that don't have access to the internet and people remote from libraries, from the world of knowledge, remote from healthcare — like, the iPhone is this incredible tool. Like, the tools are amazing. AI is just this unbelievable marvel. But the political packaging in which they're put is just a box of lies.
Tamala Edwards 30:57 ↗
When you talk to people about the concern and the threat, where are we? And where does this end up if we don't — because we'll get to the end of your book where you talk about what we can do, but where do we end up if we don't do anything? Like, I'm thinking about the points where you talk about Sam Altman saying, you know, maybe Earth becomes literally nothing but data centers and those of us who survive are living on Mars. Like, where does this end up?
Jill Lepore 31:21 ↗
Yeah, I mean, he also notably said it would be great to have an AI president, because — because an AI president, and I think he meant this seriously. Like, I think it's a worldview that's this — like, foreign to me, but I also approximate to it. The idea would be that would be the perfect — for Altman, and I think this is worth taking seriously — that would be the perfect democracy, because he believes an AI president would be a machine that could constantly, in real time, assess public opinion. Every single citizen in the United States who has a view, they'd be expressing it constantly, tracked what you think about an issue, and the president could constantly be adjusting to make sure that the majority of people were getting what they wanted. And so it would be the most responsive, the most perfect form of democracy. I mean, anybody in this room that knows anything about the democracy that we have is like, that's actually not what democracy is. We elect representatives to represent our interests. We believe in deliberation. We believe in representation. We believe in participation. We believe in deliberation. These are the elements of a democratic government. It's not like click, do, click. But that he could fathom that is, I think, a measure of how much of our obligations as citizens we have already surrendered to these tools. So, in terms of like what you can, you know, what we can all do — well, mainly we need to get our legislators to do some things and actually pay some — care more careful attention and not say, "Oh, I don't understand your computers, dear." Like, sorry.
I don't know who I was even mocking. Maybe — but like, just to not defer to Silicon Valley. But I think, you know, I wrote this book last summer, and it seemed to me very unlikely that any of this was going to be slowed down anytime soon. But it actually feels very different this summer, because all spring and summer there's been this growing backlash against data centers that's nonpartisan, that I think some of which is about NIMBYism, some of which is about environmental concerns and kind of informed environmental considerations. I honestly think that most of it is about most of us feel like we are hooked up to some scary big machine that's invisible. And the data centers are so big and so material and visible that it's like, "Oh, I now see how big and scary this machine is that I am tethered to. That I have surrendered, you know, I don't — you know, my toddler got in bed with me this morning and I pushed him away so that I could check my phone. That's how tethered I am to this machine instead of being the mammal that I am." Like, that I think people have a sense of something's gone wrong. Something's gone too far. And the kind of helpful thing about the data centers is like, "Oh, now I can see what that is."
Tamala Edwards 34:26 ↗
In the book you make a case that, you know, how we treat animals and nature is a precursor perhaps to how the machines will treat us, and that we should take a look at that. In some brutal sections you write about how terribly we've treated the natural world, what we've done to the trees, and the brutalization of animals. And there's so many data points we don't even remember. Like, you talk about the way people used to treat elephants — that if an elephant, you know, being beaten by somebody in a circus stampeded or did something, there was a public execution often of the elephant. Like, they hang an elephant from a crane at one point. And as I read it, I mean, I literally thought to myself, what is in us? Perhaps we're not capable of being better and doing better.
Jill Lepore 35:09 ↗
Yeah, I think one argument about the fear of the kind of robot apocalypse is the reason that we expect, if there is a species that emerges that is smarter than we are, that it will wipe us all out, is because that is what we do. And we kind of can't picture a smarter species that would do anything else. And I think that's why it seems so inevitable. And in a way, if it were to happen, that's why it would be inevitable, because the only thing these AIs have to learn about how to act is us.
I like the snapping. Thanks so much. I'm feeling so validated. It kind of makes the "Oh my god, that's disturbing" feel a little bit more jazzy. Now again, woo. But you know, we can scat. Little scat.
Tamala Edwards 36:07 ↗
The book ends — in the book, you know, where you call the fall of the AI state inevitable. In the beginning you say the exhale that follows every inhale, the day that follows night, Icarus's plunge into the ocean. And in the end you say, "Oh, it's a wall of mirrors that can be smashed with the slightest hammer or the nearest rock." The rock is the right people to elect a government by consent and not by automation. And my question is, can we, and will we, and will we in time? We are addicted to this automation, as you just pointed out. We can read books that talk about how companies are sacrificing our data, they're sacrificing our children. The algorithm we've been told prizes disgust and rage. And we know that China is watching us all right now. And yet we want our TikTok. We cannot stop. How do we stop, and will we do it in time? Somebody tonight in the pre-event said, "Two years from now, will it be too late?"

24 more exchanges in this transcript

Sign in free to read the rest of this interview. No card required.

Sign in to read the full transcript

Cite this transcript

APA, MLA, BibTeX
APA

Lepore, J. (2026, October 24). After Words: Jill Lepore on "The Rise and Fall of the Artificial State" [Interview transcript]. C-SPAN. CEOInterviews.AI. https://ceointerviews.ai/interview/3173283/

MLA

Jill Lepore. "After Words: Jill Lepore on "The Rise and Fall of the Artificial State"." C-SPAN, 24 Oct. 2026. Transcript, CEOInterviews.AI, https://ceointerviews.ai/interview/3173283/.

BibTeX
@misc{lepore2026_3173283,
  author       = {Jill Lepore},
  title        = {After Words: Jill Lepore on "The Rise and Fall of the Artificial State"},
  howpublished = {Interview transcript, C-SPAN. CEOInterviews.AI},
  year         = {2026},
  month        = {oct},
  url          = {https://ceointerviews.ai/interview/3173283/},
  note         = {Speaker-attributed transcript with timestamps}
}