← Lunch with Jamie
Lunch with Jamie
The CEO of The Atlantic on Signalgate, The White House, and the Future of AI
01:03:25
Loading summary
Transcript108 lines
- [00:04]
A
Today I'm joined by Nick Thompson, CEO of the Atlantic, former editor in chief of Wired, arguably the smartest person I know. We talked about his book the Running Ground, his experience with thyroid cancer, his eccentric father who owned a brothel at one point, and Nick's running. And it's safe to say Nick runs more than almost anybody else in the world. And that might not be mile for mile, but Nick, as a busy man, as a CEO of the Atlantic and a father and husband, has limited times on his hand. So his solution to find times for running is to run everywhere. He runs to work, he runs to meetings, he runs to dinner, he travels around with clothes in his backpack. He even told me he's run to the airport before. I'm just going to take his word on him. He seems like an honest guy. Nick is doing so many things. He does his daily bite sized, the most interesting things in tech conversation, which he puts on LinkedIn and all social media, which at this point is primarily about AI, but it's also about many other topics. Diving into that led to just a very in depth conversation about AI, which I hope you will appreciate as much as I did. He gave his prediction for which company thinks going to win the arms race it, and also his hopefulness for an open source AI winner. But that doesn't sound too likely. We spoke about the risks of AI videos and photos and altering reality and being used as persuasion tools during the elections. We spoke a lot about the Atlantic, which is just an extraordinary organization, started in 1857 and it continues on. And while you keep hearing the reports of legacy media dying, profits are going up at the Atlantic since Nick took over as CEO. And it's not through podcasts and video, it's through great journalism which increase subscribers. They actually just hired 25 of the Washington Post journalists who were fired at the same time. Nick is concerned a bit about how AI is going to affect his industry and he feels like it probably is a net negative, but he's hopeful. He talked a little bit about professions that are going to be replaced by AI versus which are safe. He also gave some suggestions for those with younger kids. And of course we talked about Signal Gate, which, you know, without a doubt is one of the biggest stories when it comes to the Atlantic over the past few years. We talked about how he wants to be a bit more like the Devil Rays than just the Yankees or the Dodgers. Well, actually to be clear, he wants to be less like the Yankees and more like the Dodgers. So you have to listen to understand what that means. And it was just. Overall, it was a great conversation. Nick is just a wonderful man and a great CEO for the Atlantic. And at a time when media is getting hit from every side, he is just doing a wonderful job. So here's the conversation with Nick Thompson. Hey, everyone, welcome to another lunch with Jamie. Nice to see all of you. First, I need to shout out our amazing Olympians who are doing a great job speaking from their heart and winning medals from Breezy Johnson to the quad God to the two Corys. Let's go. USA Today. I am joined by my friend Nicholas Thompson. Nick's the CEO of the Atlantic and the former editor in chief of Wired. He's also a runner who set the American record in the 50k for his age group, which I still can't even comprehend, and ran a 2 hour and 29 minute marathon in his 40s. His book, the Running Ground uses his love for running to share the story about his relationship with his father, experience with thyroid cancer, and his own experience as a father to his sons. Nick, since this is lunch with Jamie, have to start with a food question. When you're enjoying a cheat meal in New York, what is it?
- [04:19]
B
When I'm enjoying a cheat meal? Yeah, like just like a comfort meal.
- [04:24]
A
Yeah, Just something that you're just like, no one's looking and you're just like,
- [04:28]
B
I want to go vegetarian pizza.
- [04:30]
A
Vegetarian pizza.
- [04:31]
B
I had that the other day. Yeah, I stopped, parked the car and I was like, oh, this is kind of all I can eat right here. It's fine.
- [04:35]
A
And where are you getting it from?
- [04:37]
B
Wherever. The pizza place around the corner. Oh, you know, I'll go to like a burrito place. Just camp.
- [04:41]
A
Nick, I feel like we need to work harder on our relationship with meals together. So next time I'm in New York, I'm going to take you someplace. Next time we're in New York.
- [04:50]
B
Sounds good to me. Any place you want to take me, I'm down with.
- [04:53]
A
I appreciate it. Well, thank you for being here.
- [04:57]
B
So
- [05:00]
A
let's start with your book, the Running Ground. We have a lot of topics I want to get into, but I love this book so much. I bought five copies to give to friends. I went through those five and I just bought five more. God bless you. So I just, I think it's just great. Will you just for though, if there is somebody listening to this podcast who hasn't read the book, which seems impossible to me, will you just give a high level of kind of what inspired you to write the book and, and what the book Is about.
- [05:32]
B
Yeah, it's about three things. It's the story of my life, running, and as you mentioned, the weird fact that I got much better in my 40s than had been, even though I had trained hard in my 30s. It's the story of my father, who led this remarkable, broken, crazy life, and we connected through running. Simplest way to describe him is people would say, what do your parents do for a living? Say, well, my mom is an art historian in Boston, and my dad runs a male brothel in Bali. And we bonded through running. And then it's the story of different runners who take you to the edge of experience and teach you about pain and this sport in complicated ways. And they're all people I competed with or knew or who intersected with my life. So it's those three stories intertwined.
- [06:14]
A
Not to make it sound more salacious than it is, but I heard you on a podcast talking about the introduction and finding the memoir and thinking it was this sort of very sort of just like emotional, tasteful, like family friendly book. And how quickly you realized that wasn't the case.
- [06:32]
B
Yeah, this isn't in the book, but my dad had an unpublished memoir, and I remember reading it after his death and the beginning, it's. It's dedicated to my children, my sister's children. That's like to my seven, you know, grandkids. And it uses whatever the Filipino word is for grandkids. It's quite lovely. And so I start to read and it's. It's a first, like, page is about Matt, and then you get to page three and it's about the relative penis sizes of people of different races. And I was like, is this really what your grandkids need to know? So my dad's memoir was a trip because he's an astonishing guy and he grows up in relatively tough situation in Oklahoma, busts out, rises up the American meritocracy. Andover, Stanford, Oxford, Rhodes Scholarship, married into the NHTSA family. My maternal grandfather at that point was under Secretary of defense. Skyrocketing up and then plummeting down. Comes out of the closet, burns up his life, runs out of money, becomes a tax fugitive in Southeast Asia, runs this thing. It's kind of a brothel and has all kinds of crazy stuff happen. So an amazing, remarkable guy.
- [07:43]
A
Well, if running doesn't isn't the thing that gets you into the book, maybe being talking about brothels will. So for those of you who haven't read the book, order it, read it, you'll love it.
- [07:52]
B
The book for you, if you want to run a sub 2:30 marathon or know all about Filipino brothels.
- [07:58]
A
There's so many parts of the book I would love to talk about about, but we have too many topics to get through. But there's one thing that I actually sent to a friend that was a quote, that regard relates to your journey with cancer. And it was a conversation you had with Arthur Brooks and you asked him the number one predictor of happiness and satisfaction, to which he answered, get cancer and survive it.
- [08:23]
B
Yeah, that was kind of. It's kind of amazing. It was the. Arthur didn't know. Arthur didn't know me, didn't know my story. When we had that conversation and we knew each other casually, what had happened with me is that I had been diagnosed and gotten thyroid cancer, had multiple surgeries, had a really tough time when I was 30. And one of the mysteries of my life is why things professionally went so poorly until I was 30, and then so well after I was 30. And when I. And my running as well, where I, you know, couldn't run a sub 3 hour marathon in my 20s, when you're supposed to be the strongest, and then in my 30s, was able to go much faster in my 40s, faster still. And when Arthur said that, something clicked. I mean, his theory is that there's this phenomenon of post traumatic growth where, you know, you come close to the edge, you think you're going to die, and you get through it, and then you take life more seriously and your relationships are deeper, you care more about important stuff, you care less about trivia, you become more religious, you think more deeply about things. He said you can't force it, but that it is in many people, if you study sort of life satisfaction, folks who have gotten through something like cancer, and mine was on the relatively benign spectrum of it, end up being able to make choices that make them more fulfilled.
- [09:36]
A
Wow. I think that's such a. It was really eye opening to read that and to think about that. And I think that something great for people to take with them. All right, we're going to jump into the Atlantic, which is such an extraordinary publication, and just thank you for everything you do and thank you for being a steward of this publication that has been around since 1857.
- [09:58]
B
It's true.
- [09:59]
A
Which is just wild to think. How do you think about kind of the core ethos of what the Atlantic is. If sort of no one's ever. If someone's never heard of the Atlantic, how do you describe it and what sort of its guiding principle?
- [10:14]
B
Well, the guiding principle comes from 1857. Right. It was founded by a group of humanists, Ralph Waldo Emerson, Henry David Thoreau, Oliver Wendell Holmes, Harriet Beecher Stowe, trying to create a publication to debate what America meant as we approached a civil war. And that's what we are right now. We are, as they say in the founding principle, we are a magazine of politics, literature, and the arts of no party or clique. And our idea is to. Our goal is to present the biggest ideas we can about what it means to be this country, what it means to be America, what it means to have debates in America, what it means for American democracy to exist. And we try to hire the smartest people, the best writers. We try to run it as a consortium of writers, a great big writers collective, as opposed to a bunch of people with a staff voice and particular points of view or particular directives that we give them. So we run it as a big collective. We hire people across the political spectrum. We tell them to go out and find the best stories they possibly can. We ask them to write it beautifully, we edit it carefully, we check the facts, we copy edit, and then eventually they return, and they've done great work. And so I started there five years ago, and so my job has been to run the business. And so my specific role inside the Atlantic has been to take those editorial ideas, take those objectives, and then figure out a business model so you can run those objectives in the most principled way. You can write the best possible stories. You can hire people of the highest quality. And then I try to make money, and then all the money that we make, and we've gone from losing a lot of money to making a lot of money. We take all the money we make and we put it back into hiring more people. So we just hire 25 people from the Washington Post, and the hope is that we can increase the amount of great journalism in the world. So that's the goal. Like my. My job is to be smart about the business, do the math well, sell a lot of ads, and then take whatever we get. Doesn't go to our owner because she's quite well off already. It goes to hiring more writers to do more great work.
- [12:17]
A
So just to reiterate something, you're actually increasing profits these days, making more money, and then. And you just hire 25 people from the Washington Post. Can you give us a little bit of the secret sauce? I thought print media was dead. I thought, you know, old. You know, old guard was gone. I thought it's all about new media. You don't have to Share all your secrets. But is there a couple things you can use as examples of how that's happening and why it's happening?
- [12:45]
B
You know, the craziest thing I. The other day, for I don't know what reason, I was listening to a conversation that Jeff Bezos had in 2005 at the Stanford Graduate School of Business, and he was describing his philosophy of testing and iterating. And I was like, that's just kind of what we're doing. It's sort of ironic that, of course, the Post has made a bunch of mistakes and stuff has gone really well over here. But the philosophy that we took in is, you know, a whole lot of people come and read us. Everybody who comes to the Atlantic site, maybe they'll become a subscriber, maybe they won't become a subscriber. If you can set your paywall and set your price and set your offers in such a way that you increase the percentage of people who subscribe and you decrease the people who bounce away. It's a really hard math problem. You have limited information on what you come in, right? People come in and you know their IP address, you know what website they came from, you know what day of the week it is, you know what story they're reading, you know certain things about them. If you can take that information and then identify whether you ask them to subscribe or don't, and you can do that as efficiently as possible, you can run a good business. And so we just started really changing the math and really started changing the rules. And you could see the conversion rate skyrocketing, going up 600% over the last couple of years. And that combined with the amazing journalism. Without the amazing journalism, none of this works. That turned the whole thing around. And you can see our. If you look at the chart of total Atlantic subscriptions, it goes down, down, down. And then we basically make a bunch of mathematical changes. Then it goes. And the question is whether that's replicable, right? And could you take the same people, the same math, the same sort of basic philosophy and apply it at other media institutions that are having trouble? Could you do it at the LA Times? And I don't know the answer to that because I haven't tried it. But there's been. We've done tons and tons of stuff. You know, digitized the archives, we set up an affiliate business model. We set up, like a little IP studio. Most of it hasn't worked. Like, what has worked is we've become relentless about how we build and how we execute our Paywall, our paid marketing strategy, our price testing and our price conversions.
- [14:50]
A
Yeah, because I'd say, or is it fair to say the Atlantic hasn't invested as much in podcasts and video
- [15:00]
B
overall
- [15:01]
A
compared to majority of old media trying to enter the new media world.
- [15:06]
B
I mean, that's a yes and that's an interesting, it's an interesting observation. It's an important one. Like part of what sank the post or what part of what has caused the post troubles is, you know, they put hundreds of millions of dollars into trying to build out this special CMS company which turned out to not really work. And other publications have put tons of money into these quixotic video efforts. And my view was podcasts were an amazing business if you got into them in like 2014. It's very hard to build a successful podcast business if you're ramping up now. So we've been pretty slow and pretty cautious in it and we've been conservative video. It's extremely hard to do videos that are on brand for the Atlantic and that are revenue positive.
- [15:51]
A
Right.
- [15:51]
B
We've now, we're now moving into it and our video numbers are going up, but we're. It's kind of like low budget interview shows like David Fromm, Charlie Wurzel, all the stuff that publications like us chased and including what the Atlantic chase, like what the reason I was hired in 2021 was because the Atlantic had. The finances had fallen apart important. They put all this money into this sort of high end video operation and they'd fired everybody in the spring of 2020 when they laid off a third of the staff. And so we chose very much not to repeat those mistakes. So now we're in a funny position where we haven't invested in podcasts, we haven't invested in video. Small, but we're starting to because we do need to increase our top of the funnel and because we're making lots of money and everything else.
- [16:36]
A
Yeah, it's amazing. I'm sure you'll deliver great podcasts when you invest a little more or video. But whatever you're doing is working. Working for now, working for now.
- [16:46]
B
See what happens once AI like obliterates all text media. But until then we'll be all right.
- [16:50]
A
You're really jumping ahead, aren't you, to our AI conversation. I'm trying not to dominate everything. It's funny when you talk about why the Atlantic was formed and talk about a civil war where you guys have done a start a great series on the 250th anniversary of the country. Do you think back to what those founders of the Atlantic would think about at this moment? Often. And, and what your kind of, what your role as the Atlantic is in this moment that we're in right now because of that?
- [17:26]
B
It's an interesting question. I don't think that I haven't read enough Emerson to really have a deep understanding of, like, what he would think right now. You know, I've read some, I think back in a couple of ways, though. So one, I often think back about just how important it is. Like, you really don't want to be the person who sank the publication that Ralph Walter Emerson built. And as I think through the choices we make, that is a burden and also something to live up to when you're trying to figure out what you're going to do at the next hour of your day. Maybe think through the question of how the Atlantic can survive the next 10 years. But then the second thing to remember is that every now and then when you get panicked or you worry, like, will we be able to, you know, we'll be able to find our way through this Trump administration? Or what will be, what will we do when Trump's, you know, not here or, you know, what will happen after this moment passes? You realize, well, wait, we've been here for, you know, almost 170 years and it keeps working. So the idea that they came up with is a good idea. And as long as we can keep it going, we'll be all right.
- [18:40]
A
Well, you guys are becoming like the Dodgers of publishing. I mean, the fact you just added David Brooks, Anne Applebaum, Hannah Arendt from earlier days, back when it started ta Nehisi Coates today, it's really amazing.
- [19:00]
B
Can I pause you there for a suit? It's super interesting, right? Because the challenge. And actually on my wall, you can see all those stickies. That's our strategic plan, right? And one of the things that I'm trying to do, one of the, one of them says Dodgers versus Devil race, right? And we're very good at actually the Yankees versus Devil race. We're very good at hiring the right star players, right? Like, we're very good at getting Garrett Cole, right? Every David Brooks is going to work, right? Going to be great, right? Everybody's gonna like David Brooks. We know who drives subscriptions. We have the data. The interesting challenge is whether you can do what the devil rays do, which is identify the right 30 year olds and then build them up and they become an apple bombs. And we're good at that. But we're you know, we need to get better at that. So it's funny that you mentioned that we're like the Dodgers. The Dodgers are interesting because they do both, right? They hire, they, you know, they, they sign Shohei, they trade for Mookie. But they also have folks coming up their farm system that are great. I would like us to be like the Dodgers.
- [20:01]
A
Well, Dodgers are the best in baseball, so. And you're the best publication. So you're doing. You're kind of close. Before we move off the Atlantic, obviously, one of the biggest moments. It's got to be semi Atlantic history, obviously. Although I'm not going back to 1857 was Signal Gate. I happened to be in New Orleans when Book Fest was about to happen, when Jeffrey Goldberg got added to the Signal chat. A How big a moment was that from Atlantic? I mean, just from a subscriber standpoint, but also, I mean, there's very few moments as far as a journalist career. I mean, you experienced that a little bit with the Boston Marathon bombing in some ways, but as far as, like, having to be on the, you know, write a story. But that's one of the biggest moments I gotta imagine you'll experience. And how do you respond to that as a CEO?
- [20:58]
B
It definitely was one of the biggest moments. You know, it was. I was a little removed because I was the CEO, not the author. Right. And so Jeff, like, Jeff was much more in it. And the decisions, the important decisions were Jeffs and he, you know, the very, the key decisions of, like, do you stay in the chat or do you think it's fishing? Okay. Do you leave the chat or do you stay in? And possibly once, once they start sharing secrets, well, then you're potentially in violation of the Espionage Act. So how do you weigh those risks? Right. The public's right to know. How do you exit the chat? What do you do? Okay, how do you confirm it? What do you. How do you tell the government what you're doing? How do you prepare the staff for what's going to come? How do you respond when the government starts lying about you? Right. So you take all of those different things. And I think Jeff made all the right choices all the way along, including the decision to be quite cautious to call the White House to say we're about to publish, to prepare us in the eventuality that if the White House, the White House could have responded to that initial phone call by raiding our offices and arresting people. People. Right. So we have to be prepared for that. Weirdly, the White House responded by saying, oh, yeah, that's true. No problem. And then a couple hours later, lying about it. Usually they do it the other way around. They lie and then they confirm. They don't confirm and then lie. But in any case, once they then started lying, the decision to publish a second story, which was the actual screenshots and transcripts. So he made all those right decisions in full credit to him, to our lawyers, and to everybody else. How much was it worth? It was a. It was a. It was a. It was a big deal. It was great. Tons of people subscribed. And we're actually in a really interesting moment right now where, you know, we're about a year on. Our subscriptions are year term. And so my job has been to model out, like, what are projections for the year's Economics. If people, you know, resubscribe at the rate people traditionally do, we're looking great. If they resubscribe at 15 points, worse, we're in trouble. And so what can we do to get them to resubscribe at our standard rate? Can I explain my theory about how I got in the chat? I don't even think he agrees with this, but it works. And you guys can tell me whether you like it. So my theory, there's a government theory, there's that Elon stuff. None of it made any sense. And my theory is that Mike Waltz met Goldberg at a party, which we know Jeff had his business cards. I can't remember meeting Waltz, but we must have exchanged cards. And so Waltz then puts Goldberg in his phone and labels him JG because he doesn't want his staff to know that he's talking to a reporter. But he wants to be able to talk to a reporter. Right? He wants to have influence. He wants to be a player, but he knows that, like, his phone is handled by his staff, so he puts him in his jg. Then it's time to start the signal chat. And he wants to add Jamison Greer because it's trade representative and he wants the straights of Hormuz, so he adds Goldberg. And then because the way signal works, you see the person's name if they're in your address book, but Goldberg's not in hes address book. And so nobody notices it's Goldberg. They just see a 202 number and. And trust results. That's my theory. I think it works.
- [23:45]
A
I like it. I'll trust your theory. Yeah, why not? Yeah, why not? Yeah, exactly. Okay, we are going to jump into the most interesting in tech, that's which is what you've Been doing for years now.
- [24:00]
B
Eight years.
- [24:00]
A
Eight years. Wow. Which is primarily on LinkedIn. Although we just talked. You're now becoming a TikTok star.
- [24:07]
B
I'm not a TikTok star. I'm a TikTok micro micro micro influencer. Yeah.
- [24:11]
A
But for those who aren't following the most interesting tech, can you kind of quickly say what that is?
- [24:18]
B
Yeah. So every day, you know, for these eight years, I've done a short video, you know, like, it's usually less than three minutes. And it's about something that interested me during the day from tech. Right. So, you know, the last two days, it's, you know the story in the Wall Street Journal about why the AI in the Run app doesn't work, which I think shows a kind of an interesting imbalance between the sophistication of the data you can get on running and the complexity of how we think running works. And then a big Google paper on how, you know, what we've learned from studying the way AI solves really complicated problems. And so I kind of just read a bunch and I find something that interests me and then I do a little straight to camera. You know, it'll often be I'm just walking down the street, or if I'm running home in front of some graffiti, I'm traveling. It'll be in like a public park. And wherever I am, it's kind of a fun game among the commenters trying to figure out where I am. And it's worked really well on LinkedIn. It people just like. I don't know what they say is. They like that. It seems unrehearsed, which it is. It's one take. They like that. It's just what's on my mind. And you know, they like. It's not breaking news. Sometimes it is. Like, if there's a big TikTok deal, I'll talk about that. But usually it's just, here's an interesting thing. It's not the most important thing in tech. It's just, here's an interesting thing.
- [25:33]
A
It's great. I love it. It's really helpful. I buried the lead a little bit. And for those who don't know much about you, and we skipped over the book in the sense of how critical running is part of your life. And I found this out at the time we had drinks together, which was 5pm in New York, and Nick had run from his office to our drinks meeting. He was about to run from our drinks meeting to go pick up one of his kids. And I think your child was actually going to be allowed to not run home, but just to just that we'll go back in time for a second. You basically run everywhere, no matter what city you're in. Within somewhat reason. You did tell me you have run to JFK before for a flight. So I don't actually know what you don't run to. You know, the Oscars maybe you wouldn't run to in your tuxedo, but I'm not even sure that's the case. But just for those who don't know, that's basically true. Correct.
- [26:30]
B
I love running and it's very hard to fit it in, you know, it's very hard to get it to like work in your life. Right. If you have a busy job, you have a lot of meetings and so one of the ways to be efficient is you run places. And I do shower. Maybe I didn't shower before those drinks, maybe I didn't shower when I picked up my kid, but I showered this morning when I ran to the office. So I've just tried to figure out. And I like, I'll run with stuff in my backpack, you know, and I'll have clothes in the backpack, I'll switch. It's a time saving mechanism in my mind.
- [26:58]
A
I want to be just like you, Nick. So safe to say the most interesting tech for the last year and maybe two years. I would venture, if you did the math, 75% of those stories relate to AI if not 99%. Since everything in our life indirectly or directly relates to AI. So for somebody, and this sounds, maybe sounds sort of impossible, but as somebody who's one of the most knowledgeable people on AI today, what do you, if someone says to you, I haven't used AI, I don't know much about AI. Where do you sort of, where's the starting point for someone in AI today?
- [27:48]
B
Yeah, if, if I have time, I sit down with them and I say, you know, think about like what is the most boring thing you have to do today? Right? And then we kind of work through like here's how AI can help you solve it. Right? Okay, now what is the most interesting thing you have to do today? Now here's how AI can help you with that. And I sort of work through different problems with them. And I'm often trying, like even at the Atlantic where people are very hesitant about it, trying to sort of model, well, this is how you use it, right? So I'm, you know, I'm working on a story, writing a story, and I've set up like six different agents. I've given them all different personalities. They're all kind of doing research. They're all kind of working like 24, seven research assistants who can answer any question with infinite cheeriness. It's an incredible tool. Now, you have to be careful. Like, you have to use it so that you learn its limitations. You have to learn when it hallucinates. You have to learn not to sort of offload your core intelligence to it. If you're, you know, at the Atlantic, readers hate it. So you don't want to have any public facing AI to be honest, but, like, you want it to. I'm in a funny situation where, like, we want AI to make the organization operate as efficiently as possible, as ethically as possible. Like, there's so many things writing code with AI So much quicker, right? You want to use it on the back end, but nobody wants to deal with it on the front end. So we're trying to set that up at the Atlantic. But back to your question. I think that there are a lot of people who either are scared to use it, who don't want to use it because they don't like what it's going to do to society, which is real, or who've kind of used it in the wrong way and haven't gotten. Gotten much out of it. And I just feel like if you use it in the right way, you put time in it, you learn how to write the proper prompts and use the proper tools. Like, there's no question it can improve everybody's life in some way. Not that it will improve the world, but it can improve everybody's life.
- [29:40]
A
So it's really interesting you say that. I just got back from Sundance, had a great experience, as I always do, and saw a lot of great films. The one that stood out to me, that is a must watch from the festival was called. It was called the AI Doc, or How I Became an Apocalyptimist, which was by Daniel Rohrer and Charlie Tyrrell, which is being released by Focused in March. Daniel Rohr did the Navalny doc. It's. It's a. It's a brilliant doc.
- [30:12]
B
It's.
- [30:12]
A
It's just so well crafted and put together. And I don't. I don't really want to talk too much about it because the narrative and the way he structures is so brilliant. But just from a simple standpoint, he basically sets the film up with. There's a group of people who believe AI is going to end the world. And these are very smart people who have very strong point of view and have very good arguments and strong arguments why it's all over. Then he has another group of people equally as smart, equally as interesting, equally as accomplished, who explains why AI is going to create a utopia and going to solve every problem there is. Not climate change, hunger, I mean, it's just, there's nothing AI is not going to solve. And he brilliantly crafts the film. So you are, you are in it with both, you know, again, not to ruin the film. You are at the end of the apocalypse section, you are, oh my goodness, I have to get out and hug my kids. And then after the optimist section, you are, it's all like, you know, champagne and roses and life is too good and you don't know what to do. And not surprisingly, you know, he weaves a great story that, you know, winds up some varying degree of one side or the other for the most people. Where do you stand today and where do you, as much as you can, like anybody, what are some kind of benchmarks you see, looking 5, 10, 20 years into the future at this moment, about AI?
- [31:41]
B
So I, I just think AI is a multiplier of intelligence. And intelligence is neither good nor bad. And if it's, you know, harnessed in the wrong way, you get really bad outcomes. If it's harnessed in the right way, you get good outcomes. And so if like both sides of that movie, which I haven't seen, like you, you are unquestionably going to get effective terrorist attacks, you're going to get malware attacks, you're going to get people crashing markets, you're going to get, you know, hyper intelligent AI that makes mistakes, you're going to get misaligned AI that makes mistakes. And we'll warp through that, you know, the way we work through hurricanes, storms, nuclear meltdowns. And then you're going to have AI that does incredible things, right? Well, unquestionably, I think in a few years lead to meaningful scientific breakthroughs and maybe it will figure out carbon sequestration and allow us, or make, maybe we'll figure out a new composition, a new material you can use in solar panels, right? And you can make solar panels four times as efficient. Now we still have to construct them. We still need the space for it. It's not like we instantly get the energy benefits, but it can be a multiplier for intelligence at different steps in a process. It's definitely, I mean, actually that example is important because it shows why it's not going to be Utopia or death. Because even if there's a specific task of making solar panels more efficient. And it figures out the math. You still need humans at a whole bunch of steps to, like, order the panels, install them, get the land rights, like, clean them right when there's bird on them, all that stuff. And so I think it's a. It will make just. It will make life go faster, and it'll make life go faster in ways that are. Some ways are bad and some ways are good. And the most important question is how much faster does it make life go? Because if it makes economic change go three times as fast, two times as fast, maybe we can't handle that as a society. And maybe there's so much disruption and maybe there's political unrest, and then maybe bad things happen there. If it speeds things up 25%, well, we can handle that. And so that's maybe the biggest question. As for the benchmarks, I mean, my view is that it'll kind of. The closer an industry is to the training of the AI model, the sooner it'll get disrupted. And so engineering is first. And we're starting to see that. And it's pretty interesting, right? Like, you're seeing a lot of things. You're seeing some places where they're not hiring any new developers. We just saw Meta today was requiring all their developers to spend, I think, 60% of their time using AI. What we're seeing at the Atlantic is it's kind of shifting. Like, it's become much easier to write code, which means we have much more code to review, but it's not that great at reviewing code. And so we need to hire more QA teams. Right. So it's sort of shifting the composition of the engineering team. But the closer you get to the training of the data, the sooner your industry gets disrupted. So engineers are gonna get disrupted now next year, maybe it's us in media or it's consulting. It's people in written words and, like, as you go further and further out until you get to, like, barbers, plumbers, and dancers, you just have a longer timeframe where you're disrupted.
- [34:36]
A
Do you feel cautiously optimistic about how AI is going to help you help your industry, help, you know, the average person?
- [34:48]
B
I feel cautiously optimistic about the average person, and I feel generally pessimistic about my industry. I think that it's an incredible. In my industry is pretty interesting. So it's an incredible tool for writing, reporting. Right. And I used it and editing, it's just great. Right. Like, my. My book is like, it's an old generation of AI it's like six months ago that I finished it, but I would ask it when the remember the final drafts. It's a very complicated structure. Please read this all incredibly carefully and identify any moments where there's a jump in chronology that isn't appropriate. Do I mention Covid and then afterwards mention a race that I ran in 2019? The kind of subtle thing that you may not notice as you've moved things around a complicated manuscript, but that is even subconsciously jarring for a reader. And super helpful. Please identify any places where I've repeated a three word sequence in the book. Right. You know, incredibly hard thing to do. Right. But you don't want that in the book because you don't want unless there's intentional echoes, of which there are a couple, but you don't want an unintentional echo or something where you've like repeated a description of someone you haven't noticed it. So super helpful. But the overall effect on our business, right. A is going to obliterate top of the funnel search. Most people come to the Atlantic by searching and going to Google and clicking on a link. And like Google's not going to send those people to the Atlantic anymore. It's just going to answer that. Right. I've been doing lots of studies on this and lots of work and it's not as bad as I thought a year ago, but it's not looking great. And it's already, that's like, look at the memo that Matt Murray put out out the Washington Post. He's like, we lost 50% of our search traffic the last three years. And so AI eat search. You know, soon people will no longer go to websites. They'll just send out agents to summarize the news. So then how do websites make money? How do you sell ads? How do you sell subscriptions? Integrate with the agents. And then three, it'll become very easy as intelligence gets multiplied for people to build like really competitive websites at really low cost, which might be good for the information ecosystem, but it's probably bad for people at the very top of it, like the Atlantic, where somebody can build something right now that's 30% as good at the Atlantic, but maybe in three years they can build something that's 90% as good at the Atlantic in five minutes. And so how do you compete in a world like that where you have $100 million in costs? And so in general, it's probably going to be net negative for my industry, but I'm not sure. And my job is to try to make it as net positive for the Atlantic as possible.
- [37:19]
A
And you were one of the earlier publications to do a deal with OpenAI. Well, I hope it's a net positive because we need more things to support and help the Atlantic. Who, if you had to say who's kind of which company is kind of winning the AI arms race today? Let's take Nvidia out of it for a second. I think more of the sort of meta Google OpenAI and who are you? Not as a stock advice here, but if you were betting on one of these companies right now to sort of be the 800 pound gorilla, who do you think it becomes?
- [38:01]
B
I think I would probably bet on Google. They have so many structural advantages. So a Gemini according to the benchmarks, surpassed OpenAI and Claude. They seem to have solved their problems. You know they had, they had two giant AI teams that were a little competitive and confused. They seem to have solved that problem and they just have all the data in the world like the data that they can access, that Claude can't access, that OpenAI can't access. And then they have the advantage of having monopoly slash dominant positions in YouTube, in Chrome, you know, duopoly position in Android devices like the way they can search in Gmail, the way they can use those other products to sell, promote Google Docs, use their AI just gives them a huge leg up. Now probably the most impressive product right now is Claude. And then the one that of course has the largest market share and may have lock in is OpenAI than Microsoft. The others are all struggling in different ways. It's also possible maybe the best outcome for the world is one where it becomes commoditized. The models are all equal and then there's an open source model that we can use for free that we can modify and have much more control over. And that's almost, in my view, almost certainly better for humanity than a market dominated by these frontier models that accumulate all the wealth. But so far it hasn't headed in that direction.
- [39:27]
A
I agree, it seems like that is the best thing for all of us. So I want to go through just a couple of as quick as possible, kind of get through your take or thoughts on these AI topics. So maybe 30 seconds or less I pretend like we have a shot clock here. But some thoughts on AI in schools and curriculum.
- [39:54]
B
Teachers need to absolutely be teaching kids how to use AI. It's the greatest tutor ever. And there's a massive risk of kids becoming too dependent and not learning how to learn because they offload the learning process to AI. And it's like the hardest thing for schools, and I think they're getting it wrong for the most part.
- [40:15]
A
Do you have any sort of direction for parents who are listening to this of things that they should be doing to think about AI in schools and, you know, how to. That's one of the things. Actually, back to this documentary. I'll sort of give away the finale for one second, which really says, just don't be a bystander. You know, so if your school says, hey, we're starting to use AI in this way, just raise your hand and say, well, why'd you choose that? And how are we using it? And whether it's work, it's personal life, just ask questions, push back, think about things, and that alone can make a big impact. So when it comes to AI in schools, is there something you would suggest people be thinking about or doing? As a parent?
- [40:58]
B
As a parent, I think you should be encouraging your kids to use it and then explaining, never use it to write a paper. And as a school, you should be moving to oral exams, and you should be moving away from anything where the students. You should assume kids are going to use it all the time, and you should encourage them to use it. And then you should evaluate them in other ways to make sure they're genuinely learning. You should teach them all these AI modes have learn mode, which is amazing. And it's such a good tool. And I showed my kids how to use it, and my oldest son is using it in math class. He's like, I would not have been able to get through math class without this. It is an incredible tutor. But if you use it to just solve your math homework, then you're cooked.
- [41:35]
A
Guys, a dumb question. Literally, you can just say to Gemini or SATA, GPT or Claude, whatever, say, turn on Lauren Mode. Or. Is that something you actually. A selection you make in the.
- [41:46]
B
There's a little selection at the bottom, like you click on the plus button and then it's like, you know, study and learn. Got it.
- [41:53]
A
That's great. I love that. It's interesting. One of my daughter's teachers said. I know, I know. When a kid uses AI, it's just. It's very obvious. And I would much rather them not. And I'm. I will grade. Give them a better grade for a slightly worse paper if I know that they didn't use AI.
- [42:09]
B
Yeah.
- [42:09]
A
Which I kind of loved. You talked about it a little bit. I mean, AI in health care obviously seems like one of the sort of greatest aspects of. But again, if you sort of give it. What's a soundbite about AI and healthcare?
- [42:22]
B
Well, so healthcare is gonna come a little more slowly, right? What it's amazing for right now is explaining ailments to people, right? Like father in law has prostate cancer, he gets back a blood test. You have to try to make sense of it, both for your own sanity and to help him understand the choices. Like it is way better at explaining even than doctors and it's particularly good at explaining like relatives and friends. So I find that so useful. And you know, as we're actually improving medicine because it's so regulated, because the data is not centralized, because of hipaa, it's going to come more slowly. It'll improve a lot of research and a lot of medical research, but it'll come kind of, you know, I think that'll be a while until we start really feeling the benefits.
- [43:09]
A
Am I absolutely crazy? The fact that I upload pretty much every one of my medical reports to. It was GPT, now it's now it's Gemini. And ask it to analyze things for me, give me feedback. Here's what my doctor said. I'm one of those people where I believe anything that's in your computer is available to anybody. There's nothing that's private anymore. It's just, it's not possible. So the fact that I'm uploading it obviously is making it a little easier. But is that absolutely insane or do you think there's some merit in that?
- [43:44]
B
I wouldn't do it. I mean I would, I would change your privacy settings and you should click all like un. Click. If they ever ask you, can I train on this? Will you help make Claude better? Absolutely. No. Never. Right? Like change every setting to make it as hard as possible for them to train on you. And then when you're done and you don't need it, like delete it. Use the enterprise account if you can, but they're probably still going to train on it to some degree. But I don't want like my personal medical records inside the open a mall where it can get hacked and then used against me in some weird way or appear in who knows. But like that stuff, the data, the data trails that you leave behind, as we all have learned from Twitter over the last 20 years, can be weaponized in weird ways. And I would, I think you get great value from it. So you should do it. Just change all your settings so that it's. You've minimized the risk.
- [44:39]
A
All right, all right, I'm going to change My settings. What's government getting wrong? What's some things that should instantly happen. What are they missing? What do they need to do with AI?
- [44:52]
B
Yeah, they should create copyright law so that the AI companies have to pay all the creators in Hollywood and in media who they stole from. I mean these companies came and took our data in the middle of the night, violated our terms of service, violated copyright law and gave no compensation. And the government actively is moving in the wrong direction on this. If they're going to pass any legislation, it'll be one that guarantees that they don't have to comply with copyright law as it existed. So they should absolutely do that. They should probably work on antitrust and set it up so that you can't own both the model and the chips. Figure out a way that you don't end up with total vertical integration where Google or another company owns the whole stack and control and dominate the whole market. And then it should promote open source alternatives so that there's competition.
- [45:38]
A
Yeah, it's interesting you say they can't own the model and the chips. I mean it's the same thing entertainment industry is dealing with. We had these FinCEN rules years ago that didn't allow the networks and distributors to own the content. And it's the same thing we're dealing with. I hadn't thought about that. Okay, so just in regards to Hollywood for a second. Obviously in content creation, we're dealing with a situation where the proliferation of fake AI generated content is just exploding at a rate that we could never imagine. Do you know you. Are we at a place. I heard someone say something I thought was kind of interesting, which was it's not about labeling things as AI anymore. It's about labeling things as real because you can't really keep up. And at this point everything is sort of AI. Do you think that's a good idea? Do you think it doesn't matter? Just because at this point everything is going to have AI like it's a.
- [46:40]
B
It.
- [46:40]
A
It's a. It's challenging.
- [46:42]
B
It's challenging, it's messy. Because what exactly like is my book AI? No, I didn't write a word with AI, but I actually used AI to think through it. Right. And I used AI to help edit it. And I was very careful not to have it write a word in part because I thought it was unethical and my name is on the COVID and also potentially lose copyright. I do think that there is a very important battle over whether you can water make content and show that it's real. And show that it's authentic, whether you can labor it as real and whether you can identify it as AI. And that will buy us time because we're about to move into an information ecosystem where we have no idea what's real and what's human and who's a clone. Like I've. I do this daily video, most interesting thing, tech. I have a friend, maybe kind of a competitor who also does it right. He's now created. He does it once a week. He films it the other four days a week. He uses an AI avatar and posts it on TikTok. And you can't tell, right? And so we're entering this crazy world where we're just not going to know what's real or not. So everything we can do to buy time before it's total shit is great.
- [47:45]
A
The media landscape overall has become just very complicated, as you know. I mean, you mentioned in the sense of what's fake, what's real. Obviously there's now proliferation of fake news stories which I assume you could have never imagined even a decade, decade ago. I always say it's hard to kind of have an operating, fully functional democracy when you can't get the facts out to the people. And unbiased facts. Now, someone pushed back, oh, Kara Swish actually pushed back at me and said, who says the facts were unbiased when it was three networks in the 60s just giving their viewpoint. But I think, at least to me, it feels like facts led the news race early on. Today that's not the case. Whether it's newsletters, whether it's cable news, you know, I think something like PBS NewsHour I think is pretty, pretty fair. But that's making it to very small audience. Do you think we are just scratching the surface of this sort of this hyper polarized news media landscape? Do you think there's any way that we get back to being a society that has a couple of news sources that people for the most part, or believe there's some truth to it?
- [49:16]
B
It's getting worse and worse. It got worse. It's gotten worse for a couple reasons. One, of course, social media pushing us into filter bubbles. Two, AI will make our filter bubbles much worse. Where like the problem with the filter bubble is you can select, I'm only going to follow these information sources now. You actually be able to create the information, right? Like, I believe that, you know, the officers did nothing wrong in Minnesota. Like, please give me my arguments, right? And so the way to counter it would be to somehow, you know, increase trust in Unbiased publications, which is hard. The best thing that could happen would be if, like, an inverted Elon Musk were to take over Twitter and change the algorithm. So instead of, like, pushing people to extremes, it pushes people to understanding other. I actually tried to construct this a couple years ago with a friend of mine who had been the VP of engineering of Twitter. Like, could you construct an algorithm so that, you know, if there's a post and people of differing political opinions agree on it, then you promote it? Right. Like, if a bunch of people who tend to agree say they like it, well, then it doesn't get as much value if it. Like, if somebody flags it, saying, this changed my mind, just changed my view. Well, that's an even better algorithmic signal. And, like, can you figure out ways to structure conversations so people can talk to each other again? Right. Like, get people to start talking about the Dodgers and then, you know, move to the thing they disagree on and, you know, we couldn't get market product, market fit. We worked on it for about a year and then we sold the tech. But if somebody who's more successful at it could do that, maybe you do change the trajectory. But the trajectory on right now is terrible.
- [50:45]
A
Yeah. You seem in our conversation, someone who's pretty. Wants to find the truth and is. Is not sort of, and comes to their day without a strong bias. And I'm curious in what you. Where you're getting your news media these days. And you can't say the Atlantic.
- [51:08]
B
I mean, a lot of it is. I mean, I'm mostly in the tech information ecosystem, and so I have a. Like a set of bookmarks, a bunch of stuff. And I. Some of it is, you know, serious publications. Wall Street Journal, Wired. Right. Some of it is aggregators, like hacker news, some of it is newsletters, some of it is Twitter lists. I have one amazing information source, which is because my daily videos. I have a hyper fan who's so smart on AI, who's like a hardcore MAGA guy in Ohio, but who's like, really into the series, and he sends me interesting AI papers every day. And so I'm pulling in from a bunch of weird sources on politics. I'm less. I'm less engaged in it, so I don't have as good an answer.
- [51:50]
A
Okay. Yeah. I mean, no one seems to have a great answer. I mean, I think ultimately I do my best to try and, you know, get information from different sources, but it's. It's hard. It's really hard. It's really challenging, and it's Hard.
- [52:04]
B
You can live in, like, totally different worlds. You switch from threads to Twitter, and it's like you just live in a different. You're in a different universe. It can snap your head around.
- [52:12]
A
Yeah, it's interesting. You know, it appeared that Twitter was kind of. I don't want to say kind of going away or becoming less and less mainstream, but it's actually interesting. I noticed the Olympics is highlighting Twitter posts, and I was thinking they could easily just list threads posts today. I mean, I don't need to do Twitter posts. I'm sure there's enough of them. So I'm curious what the thought process goes into doing that. What's your take on the, you know, the sort of substack podcast generation now of these individual voices? Do you think that's overall a net positive? Do you think it's a negative? Do you like. Do you like seeing it? Do you. What's your. What's your take?
- [52:58]
B
You know, there's this really interesting thing that happened. I was wondering, like, why you look at the podcast world, it's pretty to the right. And you look at the YouTube world, it's even further to the right. Why is that? And my working hypothesis is that, like, we in the liberal media and all these established organizations, for a period of time, less so now than five years ago, made it so that anybody who centered right. A Republican wouldn't want. Have anything to do to work with us and, like, would, in fact, have been chased out of an organization. Like, you could have been a moderate Republican working at the New York Times, and you'd have been, like, chased out of the doors eight years ago. And so all those people didn't go into media. Right. And there's nothing. There's very. I mean, great people who are interested in media and public information come from all sorts of political perspectives and all sorts of geographies. And so a lot of them went into this new thing and they went into podcasting and they went into audio and they went into video. And that's partly why there's this ecosystem and there's, you know, it's a bigger story than that. But that, to me, is one of the more interesting media phenomenon to the question of whether I like it. Yeah, it's great. I mean, there's, like, people in substack who are amazing. The economics sometimes don't make sense to me. Why, you know, like, Derek Thompson left the Atlantic for substack. It costs as much to read Derek Thompson on substack as it does to read the entire Atlantic sometimes. The economics seem peculiar, but Substack's built a great business, bless their hearts and I hope nobody else leaves to go right for them.
- [54:24]
A
Well, I don't know anybody would ever leave the Atlantic. I'm going to go to one of our questions from one of our members in the audience, Mark Feuerstein. Mark, can you ask your question?
- [54:33]
C
Hello Nicholas, thank you so much. This has been fascinating, really fascinating.
- [54:38]
B
Thank you.
- [54:39]
C
I'm going to ask a totally self serving question in the hopes that it's useful to someone else besides me, but my daughter's a sophomore in college, she's majoring in neuroscience, she's doing an internship in Silicon Valley for a month. She's going to be at an advertising company, maybe in the summer. And I loved your disruption pyramid. And if I don't want her to have the job of cleaning bird shit off solar panels, what might you advise to a graduating college senior today, especially with those, you know, filter parameters that I just gave you possibly for the areas that show the most promise in industry today.
- [55:26]
B
Now? Yeah, so my general hypothesis is that we don't know how much stuff is going to change and so but we know it's going to change rapidly. And so the most important skill is flexibility. And anybody who wants or is like dead set on doing thing A, you know, be prepared that also have to do B or C. And so how you learn flexibility, curiosity and the ability to shift from one career to another, super important. I think about that with myself, like what exactly will I do next? How will it work? Right, so that's one and then two, you know, because we don't know, it kind of is a little freeing, right? If you, can you focus, if you, if your first decision point is like focus on what I most love and I'm most passionate about because I actually don't know, like my generation, you focus on what you're passionate about, you become a dancer, focus on what you know will work out best, you become a lawyer. This generation, like would you rather be a dancer or a lawyer? Well actually with AI you'd probably be a dancer, right. And so, you know, in some ways I would say follow your passion and be flexible.
- [56:29]
A
You said something really interesting, which was for the kids who are kind of graduating school now or in the workforce, it's a very challenging time. But you actually are optimistic for your 17 year old. And I wonder if you could kind of based on last conversation, just sort of lean into that a little bit and explain that.
- [56:50]
B
Yeah, so there's this, like, there's this the general consensus, and this is probably something where I differ the most from my to other folks who write and think about AI, there's a general consensus that young people are screwed. Right. And there's the Eric Bernholfsson paper that ran in August, everybody talked about and there's been lots of follow up and it kind of suggests that young people are having a harder time getting hired. And if you look at companies that are close to AI, sort of in sort of circle one, if you think of a series of expanding circles around AI, the closer you are to AI, the less likely your industry is to hire young people. And my view is maybe that's true. Probably it's true they look pretty good, but that's because of the way companies are structured right now. And in fact if you can use AI and you've grown up at a school and you've learned how to use it better than everybody else and you can write better prompts and you can think through stuff, that is a massive, massive advantage. And so start something new or go somewhere where that's appreciated like maybe it's not appreciated at PwC or Delight great companies, but maybe they don't want to hire you. But then go start a competitor and do it differently, right? And or go work for a competitor. And like what is going to happen over the next few years is that companies that recognize there's a market like I don't run the editorial side of the Atlantic. If I did, I would be looking really hard at hiring 23 year olds who are amazing at AI and hyper efficient and bringing them in to do that. You know, and I would love that with engineering, you know, folks who are hiring folks onto our engineering team and just approved a request to do just this. So I think there's really good opportunity for young people.
- [58:19]
A
Great. I love that you mentioned one thing in how you use AI in running but as somebody who mentioned I want to use it for my health and I like to lead people in this conversation things kind of ways, action items, things that can better themselves. Are there certain ways that are just no brainers that we should be using AI when it comes to health and wellness.
- [58:45]
B
I use it to help me like figure out like how I should work out during a certain day or what I should eat. And I think that's, that's kind of useful. Right? I'll like take. I have a big grid of all my data from my Garmin, from my HRV to my resting heart rate to and in my enterprise privacy protected account I upload my Grid. And I'm like, this is the workout I'm supposed to do. Please give me guidance on what to eat and how to do it. And like, that's kind of nerdy for somebody who trains as hard as I do. But, but I think you could, like, if you want to, you have a goal. If you want to be able to walk, I don't know, six miles, right, and not get tired in your six mile walk, have it come up with a plan, it'll do a pretty great job.
- [59:21]
A
Yeah, that's great. I'm trying to use it more. I think it really can be really helpful. How do you see AI playing a bigger role or what role do you see it playing in our elections currently and kind of going forward?
- [59:39]
B
I think it's dangerous, right, because its ability to persuade is quite scary and the ability to create, manipulate information is quite scary. And so the notion, the notion that, like Trump now controls TikTok and be able to control the information ecosystem and like, you know, maybe it'll, I don't know, maybe he'll be able to use it in a way that manipulates people. Like if somebody had infinite power, if Elon really wanted to use AI, like, look at who owns the big information. Ecosys system. Platforms like Threads is owned by Meta, right? Twitter is owned by Elon. Tick Tock is now effectively controlled by Trump's allies, right? This is where we get all our information. If they wanted to skew the elections and they rigged the algorithm to do so, that's super dangerous. And you can imagine that happening, right? And you can certainly see that happening on Tic Tac, on Twitter, maybe less so on Threads. But the relationship between Meta and Trump administration is kind of a sketchy one. So, you know, I wish Blue sky, right? Unfederated, federated, totally open, you know, you can see what's going on. Very hard to manipulate the algorithm. I wish that was the central town square for information, but it's not. So I worry a lot about information manipulation. I worry a lot about deep fakes. I worry about surprises right before the election. I worry about folks who are better at AI and have lots of money being able to manipulate and win elections that they probably deserve to lose.
- [61:07]
A
All right, to finish up first, if you have bought Nick's book, buy another copy. Give it to a friend.
- [61:13]
B
Thank you.
- [61:14]
A
Sign up for his newsletter on Beehive. We didn't get into that, but my newsletters on Beehive. Your newsletters on Beehive. We are not substackers, although everybody probably Says to you. I love your substack, Nick. Yeah. But neither of us are on substack, so sign up for the Atlantic and give somebody a gift. And they also have a great, like, family premium membership now, too, which is pretty awesome.
- [61:35]
B
Here's 200 bucks. Sign your whole family. If you get a great.
- [61:37]
A
Sign your whole family. Sign your whole family up. And I'm going to give you one minute to finish with the three most interesting things in tech.
- [61:46]
B
But you have a.
- [61:46]
A
You have a guardrail. You cannot mention something. You cannot mention AI. You cannot mention anything related to AI.
- [61:54]
B
So the three biggest things in tech not related to AI right now. Rare earth minerals, how they affect geopolitics, where we're actually going to get them, where we refine them, new energy sources. So will we actually be able to. You figure out fusion and then the extent to which quantum computing will be real in the next five years or will continue to be kind of phony?
- [62:23]
A
Okay, I love that. I think I am going to have to ask for some Cliff Notes on some of your. Some. On some of the things, though. So I appreciate. I'll say that for everybody who's sort of like, pretending to, like, feel like listening and understanding everything Nick's saying. Nick, you are the best. You're brilliant. I'm glad that we've become friends. I'm glad that you're the CEO of the Atlantic. And I can't wait for you to teach me how to run longer than a mile.
- [62:48]
B
Great. I can't take you to some better, better comfort food restaurants in New York City.
- [62:52]
A
We both need help in both of those categories. Thanks, Nick.
- [62:57]
B
All right. Thanks, Jamie. Thanks, everybody. Cheers.
- [62:59]
A
Thanks for tuning in to this week's episode of Lunch with Jamie. As always, be sure to subscribe to my newsletter@jamieslist.com for my thoughts on all things food, pop culture, politics, and more. And remember to join these online conversations and ask my guests questions in real time time. Sign up to get a paid subscriber. You can listen on Apple podcasts, Spotify or Audible and be sure to leave a review. Thanks and see you next time.