I've Got Questions with Sinead Bovell - AI Could Disrupt Media… or Reinvent It | The Atlantic CEO, Nicholas Thompson

Episode Date: December 4, 2025

Nicholas Thompson is the CEO of The Atlantic, author of The Running Ground, former editor in chief of Wired Magazine, and one of the most influential voices at the intersection of journalism and technology. He’s leading a legacy media institution through an era defined by information overload, disrupted business models, and the rise of AI systems that could both empower and destabilize the entire industry. In this episode, we get into whether media is a “dying business,” how The Atlantic is growing despite that pressure, and why AI is simultaneously the greatest tool and greatest threat to journalism. He breaks down the economics behind partnering with some AI companies while suing others, and what it will take for news organizations to survive when audiences increasingly turn to chatbots instead of websites. We also discuss Nicholas’s new book, The Running Ground, a moving reflection on running, resilience, and the lessons that shape us far beyond the miles - https://a.co/d/gxLsda3

The pressure on media, competition for time and attention, shortening attention spans, decline in trust. AI is the best opportunity for our industry. It is also unquestionably the biggest threat. AI might obliterate all of the media industry. A significant portion of people now go to a chatbot to understand the world. So tell me about what's happening in foreign policy. That is where you could see a big disruption in journalism. The big AI companies, they're running headlong, they're not licensing, and not because they want media to die.

Here's one thing I know from covering the tech industry. The advice you get from all your investors is you don't need to follow the law. You just need to get big. It takes all of your data, all of your stories, without your permission, and then uses it to train a model that would then compete with your writers. They know their products will be better if there is good information created by professional and journalists. And if you think about what a chat bot doesn't have, it is trust. And what journalism has is trust?

There could soon be an infinite number of fake news publications. And not only that, what if the web goes away? But because you are scared, that does not mean you shouldn't make rational decisions. And it's like, what is journalism? 2.0, there is a world in which AI becomes the best thing to happen to journalism. What impact will artificial intelligence have on journalism? This is one of the biggest disruptions to this industry. Today I'm sitting down with Nicholas Thompson, the CEO of the Atlantic. Now, prior to this, Nicholas was the editor-in-chief of Wired Magazine, one of the most famous

tech magazines in the world. He was also a tech founder, and he is a very influential figure in the world of technology, with 1.7 million people tuning into his history. channels every day to hear his take on the world of technology. So if there was ever a CEO to lead a magazine and to lead journalism through this moment, it's him. I'm Shane Beauvel and this is I've Got Questions. We're in a moment right now where we are quite overwhelmed with information. We spend a lot, way too much time on social media. So it's just content overload. And then now we have this AI slot to deal with and trying to differentiate what is real, what is not real. And then you have this backdrop of declining trust in many institutions.

So are these some of the dynamics that you're referring to when you've described in a few interviews that media is a dying business? Did I literally say media is a dying business? You did. Media is a dying business, but knowledge is the growing industry. So it's good to be at that intersection. So my view is that media is really, really hard. And if you look at sort of job loss across industries over the last 20 years, I think media may be

the most, right? But there's all kinds of amazing opportunities in media, right? It's never been easier to start something new. If you have a, you know, big brand like the Atlantic, there's never been more opportunities to build something new within your institution. So it's an amazing business to be in. To your specific question, yes, completely, totally. Like the pressure on media, we have pressure from, we had an advertising monopoly, right? If you wanted to reach someone in Kingston, New York in the old days. You had advertise in the Kingston newspaper. Now it's buy an out on Facebook, targeted someone in Kingston, right? So we've lost a local advertising monopoly to advertise people who like golf. You would advertise in golf magazine. Now you do that.

So our business model got undercut. And that happened 20 years ago. It took us a while to figure it out. That led to the upward pressure. The other factors are, of course, competition for time attention, shortening attention spans, decline in trust, which is due to some exogenous factors, you know, the most powerful man in the world, the richest man in the world, trying to undercut any institution that is critical of them. Both of the, both Trump and Musk have, you know, rallied a lot of power against an industry because they don't want anybody to challenge their authority or report on what is happening. So you have a lot of pressure and mistakes that the media industry is made. It's not like it's just those two guys. So there's a lot of pressure.

from all the factors that you just mentioned. And then my challenge as CEO of the Atlantic is to figure out a path forward so we can increase trust, right? We can win a more loyal audience. We can get people to pay attention to stories for a long time. And ideally, we can enrich people's lives and help American democracy. And when you mention all those factors, because the Atlantic is growing and the subscribers are increasing.

Yeah. So what are you doing that's different from some of your peers? And I mean, some of it's maybe trade secrets. But the Atlantic is actually growing in this. time. And the trust, I mean, it is really strong with the Atlantic. Yeah, trust is very strong. And our subscriber base is growing rapidly and our business is doing going gangbusters. It's due to a couple factors. It's due to really exceptional work. It's due to focus on the things we're good at and that we want to do. And we're also different from many media organizations.

When a lot of media organizations have felt the pressure, they strayed from what they were best at, or they strayed from kind of what their purpose should be, or they tried to tack too hard in one direction to chase business models? And we, for better or worse, just said, we're going to continue to do what we've been doing since 1857, and then we'll build a business model around it, depending on how the world changes. So instead of changing the core of the institution to serve a business model,

we just kept the core the same and changed the business model. And that worked a lot better. So we still and we still publish a magazine, right? And in fact, it is a monthly magazine. We were called the Atlantic Monthly in 1857. We published 12 issues a year now. And we publish long-form stories. And we put the best information in those stories, sometimes in the beginning, sometimes in the middle, sometimes at the end.

And we then created a business model, mainly a subscription model, did the math really well, and it turned out worked well. And I love that you said you held on to what you were in the business of doing. Because when I work with Come, it needs two, when a new technology or technology, a disruption comes into the mix, it is not now what are we because of this technology. You should, what is your North Star and that should never change. Like who you are as a business and what you're in the business of doing, that stays the same. How you may do it and how it looks and how you're running from it, that changes. But it seems like you've kept that through line with the Atlantic. Yeah, I've tried very hard.

And my whole career, I've kind of learned this lesson at The New Yorker. So my career was fairly linear, which is like I just wrote long stories and I edited long. stories and I didn't worry about the business model. And I did that. And from the time I was 23 years old to the time I was nearly, I guess what had been like 37? And at that age, I was put in charge of the New Yorker's website. Right. And so then that's a real challenge where you have this institution, again, it's a hundred-year-old institution. It's known for its writing. It's known for its quality. It's known for its brand. And you have to produce information at the speed of the internet. So how do you do that? Right. How do you maintain the core? How do you maintain

in the brand, how do you maintain the sense of trust due to the fact-checking? And so we had to set up structures or charts processes that could get information out really quickly, but that was also still true to the point of the brand. And that was like that was the first really complicated management challenge I'd ever faced in my life. But you're also, I mean, not every CEO of a media company has a background like you, right? You were the editor-in-chief of Wired. So you're on the forefront of technology.

You've started tech companies yourself. You're a very influential figure in the world of technology. Moderately. I would say not so moderately. You have 1.7 million followers on LinkedIn that tune in, myself included every day, to hear what you're going to say on technology. And even separate from the Atlantic, like, what is his take on the moment or in with respect to tech or quantum or whatever? So you make decisions like, let's partner with Open AI.

We can sue some of the tech companies, but let's partner with them. So what is some of the calculus and how you go about steering something like the Atlantic in all of these continuous tech disruptions, many of which we don't know when they're going to end? Yeah. So one of the really interesting things about my career, which wasn't planned but was a real benefit, is that if you cover tech, you learn about tech. And it shouldn't just be tech.

If you cover the thing that is growing, you learn about the thing that is growing. And that thing creates opportunities, right? So if you obsess over something that is shrinking and its importance to the world, you can become the one great expert on it, which can be super valuable, but it narrows your opportunities. If you become an expert on tech, suddenly you learn all kinds of things that are very useful. So I ran the audience team at The New Yorker and it wired. How did I know about the Facebook algorithm? Well, because I had covered the Facebook algorithm in depth, right?

And so you learn these things that then set you up. To the big question, which is how I make tech decisions for the Atlantic, again, I come from it, you know, I have some advantages. One, haven't been a tech reporter, I know a fair amount. Like, I know a little bit about how product decisions are made at these companies, which allows me to think maybe a little further out in the future into how the Atlantic invest inflect upon them. You know, I know the players because I've interviewed them. So I have that advantage. And having come from journalism, you know, I understand what scares journalists, what journalists don't like. Now, that doesn't mean I make the right decisions and I have some disadvantage. I didn't

spend my life like negotiating deals. So I have some advantages and disadvantages. On the open-dai deal specifically, which was a really big moment, you know, my view has always been there should be a fair exchange of value. If an AI company scraped you. And by scrape you mean takes all of your data, all of your stories without your permission, and then uses it to train a model that would then potentially compete with your writers. Exactly. If they do that, there should be some exchange of value. And the exchange of value could be they provide links to your stories in a deliberate and systematic way. The way Google did, right? Google was great, right?

They scraped all our stuff. We fed it all in. And then they provided a bunch of links. Facebook, same thing. There was an exchange of value. I think there should be some with all of these tech companies. And so with Open AI, we made a deal. They send us money.

We also work with them to improve their search products. It's a good deal. With some of the other AI companies, there's no exchange of value. They just took everything and they give us nothing. So with them, we're involved, we're party to one major lawsuit. Who knows, there could be others, right? You know, we're working to stop the scraping as much as possible to give us leverage in those conversations. And we'll figure it out.

It's a long, ongoing process. And it's one in which, you know, I also, here's one thing I know from covering the tech industry. One thing you learn when you cover the tech industry is that there is a belief Actually, I probably learned this more when I was in a startup. When you're in a startup or you're at a small company, the advice you get from all your investors is you don't need to follow the law, right? You just need to get big because if you follow the law, you'll never get big, right? And so what's the point? You'll just go out of business.

If you don't follow law, maybe you'll get big. And if you do get big, you'll have enough money to pay everybody off. So that was essentially the same position taken by all the AIA companies, taken by Anthropic, taken by Open AI, which is like, you know what? Let's pretend that scraping all this data is fair use and whatever. Maybe we'll go out of business in two years, but maybe we'll be worth a trillion dollars and we'll just be able to pay everybody off. And so that's what's happened. So the calculus you made was AI companies are going to, they've scraped this data.

They've already committed whatever we're going to call it, whatever it's going to end up in court. We might as well get a check from it because the damage is done. So at this point in time, especially when the willingness to pay from an AI company is probably higher, you might as well take that check than the ones that we can. And the ones that we can't will sue them and see how that process gets figured out. Because somebody might listen and say, well, how can a company that stands in the pillar of journalism partner with an AI company that's also potentially going to try to take them out of business? But the calculus in terms of the economics makes sense from that angle. Yeah.

I mean, you can, this is one of the debates in my industry. You can yell at them and just say, you know, bad, bad AI. companies, and I understand that, right? But, you know, if there's a rainstorm coming, you can yell at the rainstorm, or you can, like, get an umbrella and go outside and, like, try to figure out how to navigate the rainstorm, right? AI, like, as I said at the beginning, AI is the best opportunity for our industry. No question about it. It is also unquestionably the biggest threat. Like, AI might obliterate all of the media industry. I hope it doesn't. I'm doing my best to navigate it, And I can see why everybody is scared.

But because you are scared, that does not mean you shouldn't make rational decisions and you shouldn't negotiate to get more money. The mistake I made, looking back over the last couple of years, is I overestimated how valuable our data would be to these AIAC companies, right? So when these deals were happening in mostly in, you know, sort of the winter, spring of of 2024, summer of 2024, the value of human-created data was probably certainly higher than it is today because since that time, the A company's have figured out how to make synthetic data, right?

And they have figured out how to pull data from other places. And they kind of need us less. So the price has gone down. So if I could go back, I would probably have negotiated even more deals back then. But it was also a very hard political time because, as you remember, the whole media industry was saying, you know, AI is the devil. How can you possibly even sit with these people? They stole their data.

And my argument was, I don't know, I'd rather get paid than not paid. Yeah, and then deal with point two later. And so you said two things. So the first, AI is the greatest tool, the best tool for your industry, reporting, editorial. So from that standpoint, how do you see it that way for journalism and how is the Atlantic using AI today? Right. So I think AI, you know, particularly. Keeley chatbots. And I can't see five years into the future. I have no idea how we'll be using

a. I'm going to challenge that because I think you kind of, I can make things up. Okay. But so I'll tell you about the way I used AI in my book, right? So I don't control the editorial side of the Atlantic editorial reports to the board, business reports to the board. So they can, they make their own decisions. But the way I used AI in my book, I never used it to write a sentence because A, it's bad writer. B, I think it's unethical. My name. is on the book. You should feel like I wrote every sentence. I read out loud every sentence. There's no AI voice generation. I never used it to write a sentence or a word. I did use it, though, to help with research. So I uploaded my book is about my life running, my father's life. I uploaded all of his

diaries, his emails to me, his memoirs. And I would ask it questions. Like tell me everything that he ever wrote about Roger Hansen, right? And, you know, AI search is much better than just sort of the keyword search or whatever you could do in sort of Microsoft Finder or Apple Finder. So it was much better for that. I then uploaded transcripts of interviews of conversations I've had. There's a woman I interviewed Bobby Gibb over five years. I can't remember everything. And so I uploaded you, I wrote, let's say, 3,000 words about her.

I upload those 3,000 words. Here are the transcripts of all the interviews. Is there anything that I wrote that is inaccurate based on she said? Is there anything she said, any quotes that are better than the ones I used? Are there any themes from my conversations with her that I'm, I should better represent. Right? Super useful.

I was trying to make sure the book was really tight, right? And really risk. I would upload sort of every thousand word section. Is there anything in this section that is extraneous? Is there a word? Is there a sentence? Is there an idea that is repetitive, right? I would upload the whole manuscript.

Is there anything in the manuscript where I repeat the same idea? Because you can't, you piece together 75,000 words and you've moved things around 100 times, right? So it was a very good editor thought partner. I would ask it to help me understand complicated processes. I would ask it to, like, read the whole manuscript, find any contradictions, find anything where Nick says one thing that is then contradicted at another point. It's like an infinitely fast editor, which was so helpful.

You don't ask it to fact check or make stuff up. It has all these limitations, but once you learn how the models work, back to the Atlantic. So we use it on the business side. We're trying to figure out how to use it in coding, of course, right? Go and clean up old Jira tickets, right? We're working on that project. right now. We use it for a bunch of other internal processes, like figuring out how to be most

efficient in the way we use Facebook's marketing tools. On the editorial side, they have a similar rule. We have a similar rule, not to use it ever for writing, but we're going to try to design. We have a suite of tools that everybody in the journalism side can use. It's not fully optimized. My goal would be in a year to have a suite of tools that will help the reporters find stories, prove stories. You know, I think of like finding stories is really the key. like finding a great story, it's like a needle in a haystack, right? Like an AI is kind of a magnet. And so we can build a big magnet for the editorial team.

I think it'll be incredibly useful. But, you know, like a lot of organizations, the feeling about AI is very different on the business side from the edit side. And it's very different in management than in staff, right? Because management is always like, AI, let's go, right? It'll make us more efficient. And staff is always like, wait, this is going to take our job. So navigating that balance is tricky.

Yeah, I mean, it's a general purpose technology, so it's here. So our question becomes how do you build on top of it? And I think that there's the mindset, like, how do we just resist it? But that's probably not going to work out in the long game. So it's how do you take this tool and turn it into a scalpel for the work that you do? And it seems like that's exactly what you do. And I do it on my substacks too. I'll upload my substack and I'll be like, be this person that I know doesn't probably like my takes

or has a different view of how the works and find out where, like, where do I have a gap and anticipate where they would call that out. But you also stated... But we stated it for a second? Because I think that's so great. And if I were still... When I was on the editorial side at Wired, right, that's the Entry of Wired,

I would tell people, okay, you're writing a story that covers politics. Read this story out loud and imagine that the person sitting across your view completely disagrees with all your political assumptions, right? And generally, like, it's, you know, the staff is generally left wing, but imagine you're reading this to the smartest right-wing friend,

you know. And that was you have to do that kind of in your own head. Now you can do that with AI, right? Like, hey, model, this is my story. Give me feedback as though you are the smartest political critic of what I've written. So if you're David Frum put in your story and imagine that, you know, you're like I'm a Maldani supporter, right? And if you're on the other side of the political aisle, imagine it's like David from, right? And like, it's going to be a helpful exercise.

So I think what you do is great. And it kind of reminds me when you look at some of the historical. archives of when the computer first came about. And there were a lot of critics that now writers should be impeccable because there was infinite edits. Before with the typewriter, you had to either start again or you went to really know what you're going to say. And now you have this infinite edit machine and there was all that controversy. And now with AI, it's like, it should actually probably raise the bar on what people bring to the table because you have no excuse to miss that gap or not see that perspective or to contradict yourself. I see AI as a scalpel if you can use

it that way. Totally. I mean, you can also, you can become too reliant on this. I was just seeing my friend Maria Konniko is a wonderful writer. She writes all her first drafts out by hand because it's a forcing mechanism to stay inside your head and not rely on these tools. You need to find the balance between using the tools and not dulling the scalples of your own mind. And I think we're figuring that out in real time, like literally building this plane, this cognitive plane and the next to flying it and hoping that we still have the brains on the way out. But you also said that AI could obliterate the business model and the industry. And so my hypothesis is this what you're extrapolating to. So some of the

trends show today that a significant portion of people now go to a chatbot or an AI assistant to understand the world. So tell me about what's happening with the war in Ukraine. Tell me about what's happening in foreign policy. And that is coming through a chatbot. So are you also envisioning that is a trend that increases and more and more people are going to use chatbots as the portal through which they interact with the digital world. And that is where you could see a big disruption in journalism. Yeah. So there are maybe there are three big things that could really disrupt our business model.

So one would be that. Right? People don't, you know, they no longer go to the Atlantic.com, right? They don't even go to the Atlantic app and they don't even go to the Atlantic's blue sky account, right? They're just going to their chat box. Tell me what's going on. And so to exist in that world, we would have to figure out how to monetize. our information. Do we license it to certain chatbots and not to others? That's a hard,

hard game to credit. Well, how do we have an advertising model? Do we buy an ad from Chanel and then create branded content, which we then license? Like, that's, who knows how that's going to work? So that's problem one. Problem two is what if the web goes away, right? So right now, people go to the Atlantic. You type in a search term, you know, explain the Ukrainian peace deal, right? And we have an article on it, so you read our story on it. Or, or Gemini just answers it, or Chatsypiti just answers it. And right, we don't get any credit. If you go to ProRata, we get credit, but most of the search bots, we don't, we don't get any credit.

We don't get any money. So that's the second thing. And not only that, there could soon be an infinite number of fake news publications that are just scraping, putting stuff together. So, and then the third is that advertising just completely breaks because, you know, advertisements are created on the fly. They're all AI generated. You know, we know exactly what Nick likes at this particular moment.

We know that Nick's a little tired because he took the red eye. And so we're going to serve them this particular ad at this particular moment with this particular music. And it would be very hard for media get in that game. Right? Like the trend I talked about at the beginning where our advertising model was sort of wiped away by Facebook and Google, there is still an advertising business

that works for us and in fact it is growing, but it's usually a sort of top of the funnel brand advertisement for elite brands that want to reach our readers. It's possible that AI-based ads sort of totally change that market again. So you have those three trends that could turn it upside down. So we have to figure out how to make a business model that is resilient to all of that. But to me, there's a way through because, of course, I mean, the chatbot trend, it's, I don't know how much of your business right now is referrals through Chajibouti or comes through ChachabuT traffic. Zero.

Well, zero point two. Okay. I meant that. But more and more people, especially younger audiences, that is how they are engaging with the world. It's like this filter through AI. Yeah. And if you think about what a chatbot and AI doesn't have, it is trust. And what journalism has is trust.

So there's a world in which journalism transcends this disruption. I mean, and I'm just thinking, let's say there's a world, it's 2030, and I'm asking chatypte for the state of the world. And I know that the answer is going to be powered by the Atlantic. I'm okay with that. Right. Right. So if I've subscribed to the Atlantic in some model, some form or fashion, and that's how it pulls through, because don't, Don't you also have to meet the moment?

I mean, there's going to be people like my dad who will always have the Atlantic magazine and with coffee Saturday morning. I love your dad. Yeah, without a doubt, unequivocally, you're a huge fan. But then there's also people that are going to have an AI-first device that doesn't even have a screen, and they're going to chat to some chatbot. And that is going to be a different way to interact with media. Yep.

And we know that that's coming. So, like, what are, do you have ideas on how that could look? Yeah. I mean, in some ways, it gets back to the licensing questions. So the first challenge we have to find is to make sure that whatever that device is, you know, whether you're wearing it around your neck, you're wearing it on your hand, you're speaking into it, cannot get our stuff for free, right? And we know they're going to try to get our stuff for free.

And so we have to lock it down, right? We have to make sure that we're not syndicating it to someone who's created a backdoor. We have to make sure that we are blocking their bots, right? And they will use headless scrapers. They're very creative at getting around all of our blocks, right? So we've just partnered with Cloudflare, which is very good at blocking these guys. So we'll see who wins that battle. So we have to have control of our data.

Once we have control of our data, we have to then figure out who we license it to, who we sell it to. And I think you're right. We will find, you know, we will sell it to some devices. We will not sell to others. And my hope would be that the future you've imagined

is the right one. So two people will come out on wearable devices that you put around your neck that tell you about the world. And we will license our data to one of them. And that will give them a marketing advantage. And so therefore will be valuable to them. And so then they will pay us. maybe they will even pay us a referral fee.

The other thing that I hope happens, I mentioned this company ProRata, so ProRata has built a kind of reverse AI search where anytime there's an answer, it figures out who contributed. This is 17% from Wikipedia, 3% from this answer on Reddit, 6% from the Atlantic, 9% from the National Review. And then you can take the revenue generated from that answer, take 50%, give to the company, 50%, you share out 6% to the Atlantic, 9% to the National Review. reviewer, right? And if ProRata's business model gets adapted, either through licensing deals or the courts, buy the big AI companies, media is in great shape. I'll say one more thing. The big AI companies, they're running headlong, they're not licensing, and they're mostly

not licensing, not because they want media to die. Like Grock and Elon want media to die, right? They want to be the sole. Elon wants to be able to control everybody's information because he knows that's right and nobody else knows what's right. Like he wants to kill off all sorts of information. The other tech companies aren't like that, right? Like Google Anthropic, OpenAI, they may have not been fair in these licensing negotiations, but they know their products will be better if there is good information created by professional journalist and fact check by professional fact checkers. So they have an incentive to create a world in which we don't die. And so they have an incentive to partner us. They also have an incentive to run like headlong towards AGI and if we're going to

slow them down with court cases, well, that's a problem. But there is a force kind of pushing them in the direction of doing what's good and just. A hundred percent. I mean, there probably isn't in the world where Open AI wants to have a journalism division. But journalists would want to work for the Atlantic. So there is this kind of potential match. It just might be that there's a division of the Atlantic that is unrecognizable in how it creates content, right? Because if it's more oral and you have a journalist that is kind of creating stories that I get to talk to and talk with. And if that's where younger users are moving, then that's really interesting. Well, so this is, okay, so this is the thing that if I could figure out one problem, it would be

this, right? So I know a lot of our readers, right? I know your dad is going to keep reading the Atlantic, right? And it's going to keep reading the Print Atlantic and, you know, maybe your dad's younger brother reads the app, right? And like, sort of, I know how a lot of people, and they're going to keep doing that for a while. So we need to make our app as good as possible, or print magazine as good as possible. The web's not going to be completely obliterated, you know. But what I don't know is I don't know in five, 10 years how people will be consuming our kinds of content. And as soon as I have a strong hypothesis on that, I can build for it. Right now I just sort of build generally.

Let's do the best we can. Let's build the best tools we can. Let's be as efficient as possible. Let's get the most fastest moving product team we can. As soon as I have that hypothesis figured out, that's the moment I'm really looking for. And I have, you know, a bunch of stickies on my wall with bets and ideas. But I don't have strong conviction on that. Maybe the Atlantic will be consumed in like short one minute serial videos, right?

Could be, right? Maybe it will be consumed just as part of like a walled garden of a small number of publications, like some kind of future state Apple news. Maybe it will all be audio. Who knows what it's going to be, right? And so there are all these different. And maybe it'll be different in different countries. Anyway, once I figure that out, then I can start building the roadmap for the next 10 years.

And this is where I see journalism winning, right? because it's not just media that gets disrupted. It is the creator economy, and I talk about this a lot. Everybody in how they make content is going to change because of this technology. So there's a world in which traditional journalism leapfrogs over all of the different ways that we try to consume content now. And it is all of those things. So if you're the person that needs the one-minute summary with your AI agent, the Atlantic or traditional company is there. But if you're also the long-form reader, you're also there.

So there's a world in which journalism actually wins. in this next chapter. It just, it has to be, and I like how you're very open to aspects of the business may be unrecognizable, and that is how you transcend the technological change. You don't just hold on to the medium, which would be writing only forever.

Yeah. It is writing and. It's definitely going to be writing and. You know, and I've been through a couple of smaller versions of this, like the creation of the podcast economy was super interesting, and if you look back, if I go back and I look back, say, 10 years ago,

right? So 10 years ago, it would be 2015. I was at the New Yorker back then. it was probably right about when we were figuring out that audio was going to be big. And we made a, we didn't make exactly the right bet, right? We bet on the New Yorker radio hour, so we created a podcast and a show, but we probably should have just created like the daily, right? And, you know, so there was, it was like moderately successful.

It was also a moment where everybody was like, hey, it's all going to go towards short form video, Facebook videos. That turned out to be the wrong bet, right? Because the business model just couldn't sustain it, right? There are a few places that ended up with YouTube channels, but like traditional media companies by and large didn't succeed with that. Individuals did, right? There are like a bunch of people actually worked for Vox and created their own shows. But like the economics of traditional media and sort of 10 minute video, three minute video, it didn't work, right?

And so the New Yorker sat that one out, which was clearly the right choice. So you don't exactly know. You can see where things are going, but putting your chips on the right number is hard. And I guess it's challenging, too, if you're the Atlantic and you're led by Nicholas Thompson, if there's anybody who's going to survive and kind of ride this wave, it's you with your background. Maybe, or I'll make all the wrong bets because I'm like super excited about something that goes under. But yes, I'd like to think that's correct. Thank you for your confidence.

I will stop undercutting your praise now. But if you were a small, regional, local, independent company, I mean, this could either be amazing like a leveler, because now you have this team of systems that can also help. Yeah. Or, I mean, in that world where we're describing where, say, AI agents do become a portal for some people to get their information about the world, how do you even filter through that? So this is kind of fun, right? So if I was a small regional paper, right, I would be in decline. And when you're in decline, A, that's bad.

It's not fun. But it also allows you to take bigger risks, right? The Atlantic is in descendants. So you don't really want to risk what's working. If you're in decline, I would be looking at me saying, you know what? We're going to move the whole newsroom to, you know, one-minute video storytelling. And then, like, you try to build a product and maybe you alienate the audience you have, but maybe you build the future of media and then maybe you're acquired by, you know, someone for a zillion dollars.

Or maybe your business model works, right? It's like a little bit like, you know, if it's third and two, you're kind of conservative. It's not as much fun. If it's third and 17, right, you're going for the Hail Mary down the line. So we're down, you know, off the side. So it would be kind of fun in some ways to be a, a place like that. So if you are a small media company right now or an independent journalist, you're saying get in startup mode, like this is actually a good, could be a great time for you

to present to try. Absolutely. Absolutely. Try something, try something new. And in part, you know, as the Atlantic, my job would be to be looking at all of these, you know, startups and figuring out, oh, wow, they've kind of figured it out. Okay, wow, they figured it out. So yeah, there's a, you can take more risks. And twice in my career, I have seen an opportunity. and seen a market need and built a startup. Both times they became quite valuable and then less valuable, but I don't know whether we had execution failures or whether we had sort of lack of product market fit

or who knows why, but I love the process of trying to see where there's a gap and then trying to fill it with a small company. Yeah, there is a world in which AI becomes the best thing to happen to journalism. Oh, completely, completely, right? And it will certainly, there's no question that, you know, 2035 will look back and be like, oh my God, right? This incredible thing started out of, I don't know, the San Francisco standard, right? Whoa, or this thing like Concord Monitor built this amazing, or the Philadelphia Inquirer. Philadelphia, what's the big newspaper in Philadelphia?

Philadelphia Inquirer built this incredible new thing, right? Who knows? And, you know, and then there'll also be things that like in 2015, my God, it was like BuzzFeed. It really is right. Just syndicate your content everywhere and rely on. the social algorithm. That didn't work, right? It wasn't sustainable. And I think what also really inspires me is if you look at the history of journalism and why it grew to the extent that it did, the printing press. It created so much chaos in the world that it started with business people,

had to pay people to get to the bottom of the truth. What are the facts in this moment? And that is really what propelled the industry. So to me, we're in another one of those printing press moments. And it's like, what is journalism 2.0? Because if the printing press is what, you know, launched it, does this technology mean for it? And it could actually be the thing that expands what we consider journalism. I hope so. I hope it does that. And I also hope that as it does that, we and all of our peers are able to stay true to our core. Like back to what I said at the beginning. Like the Atlantic, you know, we were founded with a very specific set of principles, in fact, with a very specific mission statement by, you know, people we look up to. When you're founded by like Ralph Waldo Emerson,

you kind of care more about it than when you're founded by some Schmo. And so we, we'll try to figure out how to navigate this crazy new terrain while holding true to these principles that, you know, we're a writers collective, we're of no party or clique, you know, we're literature, politics and the arts, right? That we are a publication, that we are using whatever form exists to try to create debates and public understanding that help America function. Because you really know who you are. Yeah. But maybe not all companies do. And we're also seeing, I mean, in times of disruption and change, power consolidates to survive. Yeah. And we're seeing a lot of those dynamics. dynamics in media. Actually, I'll tell you another story. So this is an interesting one which relates to that. So the first company I started, which was in 2009, was called The Adivist. And we started because, you know, two friends of mine, and I can still remember the day,

we're like, you know, there should be a multimedia magazine. Like there's a new way of telling stories, right? Like, you should be able to switch between listening to a story and reading a story and seeing a video. And, like, there's got to be a way to make stories that is different from the way they're being made today. And so we started it, and the iPad came out, which was magical, because, oh, now there's a device where you can look at it. And then we started, like, we did a bunch of other things. So we created this magazine. It was very cool.

We had some very good stories. In order to publish it, we needed to make a new CMS. So we made a new CMS. And then suddenly we had a product, which we then licensed, right, which was then, you know, at its peak worth many tens of millions of dollars, right? We had a, like, a great new CMS. And then we were kind of out-competed by the market and driven down. The interesting thing is that the ad of us still exists, right? and is still true to its initial ideal,

which is like great multimedia, amazing, long-form narrative storytelling. And so it still publishes every month. The stories are great. It's still out there. Got acquired by WordPress and, you know, put a nice ending. And so if you were to give advice to somebody who's aspiring to be a journalist or step into the field, because we had an economist on Ajai Agri-R-Wall, and he was saying how, you know, AI is going to change power in organizations.

Yeah. Because the skills that gave you dominance before AI might not be the same skills that give you dominance. After AI. He's very right. And so when you look at a skill like writing, AI, it's very bland. And if I see any more, it's not X, but it's also Y logic. I just don't think I can handle that at all because of AI. But there is a world in which the premium on great writing maybe goes down a little bit.

And so the competition in journalism, it shifts a bit upwards towards can you even spot the specific story? Like maybe you're no longer the best writer in the game or you're not an amazing writer. But you have an amazing perspective on the world, this nuanced take, and you specialize in economics, and you know how to spot the story and get the facts and then you have the sources, but you don't really write, but then you have AI. And in this future world we're describing if there is a world where somebody is chatting to chat watch, that person also has an interesting role. So how do you think about who could even be a journalist in the future? Because it might look different.

Yeah. I think if I were, I mean, if my, let's say my eldest son, he's 17, he's going to go to college next year, Let's say he wanted to be a journalist, but I don't think he does. On the writing question, you know, you're right. Like the quality of bad writing will get better with AI. I would probably argue that you should try to have a distinct style, right? Because AI will, I mean, people can change the temperature ratings. If you're a very good prompt engineer, you can make it like write in a creative way.

But it'll probably be pretty bland. And so if you can develop a sort of a specific creative style that will probably give you an edge. But you're right, that maybe that's not where the edge is. I would tell him that the most important skill would be very fluent in these models and to understand how they work, right? And to be able to know, like, when do they hallucinate? Why do they hallucinate? Okay. How can they help me get through things?

When do they not help me get through things? And to become a person, if I were hiring and I were hiring like a young journalist, let's say the Washington monthly where I was hired for my first job, I'd be looking for someone who knows a ton about AI to bring into the organization. There's all this theory that, like, young people. People are in the worst position, right? If you look right, the Eric Bernalson data, which is quite convincing and quite well done, pretty convincing and quite well done. I actually think it's the best time to be a 23-year-old, right?

Because you're native to these tools. You'll have really studied these tools that every organization is going to want people who know these tools. Also, organizations are going to flap, right? Traditional hierarchies are going to go away because the line between engineering and product and product and design, it's going to blur because the product designers are going to know some product, the product people are going to know some design, right? The engineers will be able to do basic wireframes, right? So AI is going to kind of start to blur job types and job categories.

And so hierarchies are going to flatten. And as that happens, it creates opportunities. I know. I like to say that jobs might go away and job titles, but work isn't. No. And we see this flattening of the org chart. And I agree with you that if you're a young person, right now you're living through the disruption.

So it is a bit challenging because businesses are figuring out, like, who do we even need to hire and why? But there is a future where young people actually leapfrog over everybody that's been in the workforce because they know how to use AI in that field. And so if you can just hang on or get really good at these tools, there's a world in which you step in and you're actually at a different level because the metrics of success change for what it means to be good in that industry. I also, I would advise that I may even be more than that, which is I know that any company or any person, or I'm convinced that any company, company or any person who doesn't use AI is going to lose. Maybe you're so good that you without AI can beat the next person with AI, but probably not. Like, probably if you don't use AI, you'll be beaten by someone who's equally good who does use AI. That's on an individual level. You know, AI won't take your job. Someone uses AI better than you will. And I think the question for organizations

is whether that's true of organizations too. Will it be two companies compete? Coke and Pepsi compete, right? Pepsi learns how to use AI better than Coke. Pepsi beats Coke. Or will it be that like some AI native company beats both Coke and Pepsi, right? Because they figured out the right way to make soda in a much more efficient way. And, you know, in an industry where there's like, you know, factories and water contracts and all that, like, it's much harder. But in a industry that's just like your mind, you can imagine AI native startups like disrupting it massive, just as happened with the internet and mobile. And so if you're a young person, look at the industries that are out there and say, you know what, I can go A&A.

native and disrupt this one, and I'm going to do it. And so that's an even better opportunity. Yeah, like the people, the companies that are leading pre-general purpose technology, it's usually an entirely different crop of companies after the disruption of that technology. Yeah. And so now I want to move to the world, Nicholas Thompson. Let's do it. And start with you as almost this independent force as well.

You create a lot of content. Thank you. You are a very influential figure in technology. I tune in every day to your videos. Thank you. Almost 2 million subscribers, almost 500,000 on your newsletter. So the choice to build and maintain such a strong personal brand, what value is it bringing you? It's a little weird, right? It's a little surprising. I'm probably the only media CEO. I mean, many of them have very strong personal brands, but I'm the only one of relatively few who's like an active content creator. And so it started when I was at Wired and I kind of wanted both.

I mean, really was like trying to experiment to understand what was happening. I was like, well, I'll make a daily video. If I make a daily video, I'll learn how daily videos work, right? And I'll also have like a forcing mechanism that will force me to understand what's happening in tech. And then there's also like a broadcast mechanism, which is, I don't know, instead of like sending an email out to the staff saying I'm really interested in like this paper, I'll just make a video. And I also send an email out of the staff, but like put it out in the world. And so I started doing it and then it hit. And then I kept doing it.

And then there was a moment where I became CEO of the Atlantic. And a bunch of people were like, okay, Nick, time to stop, right? And you're like, now you're like you're wearing a suit. You're a CEO of the Atlantic. Like, why would you do this? It's going to interfere with the editorial. And I was like, I don't know. Like, I might have actually stopped for a week when I took over making these daily videos.

And I was like, you know what? I kind of like doing them. And I think they're good for the brand because obviously like, you know, as CEO, I sort of represent the Atlantic. and these videos are good and people like them. And then it's turned out that it's really good marketing for us because lots of advertisers see them and we've built this whole kind of business around it

where I do these videos, people watch them, communicate, I then host dinners, right? And we have like a whole ecosystem of dinners and private things that I host that are kind of based on the brand I created through the videos. And so all that money flows back to the Atlantic, which is great for us. And so, and we'll figure out different ways of expanding it. You know, we've turned it into a podcast, the most interesting thing in AI, which had a good sponsorship, wonderful partnership with PWC.

So it's turned out to be a very good economic decision. But the thing that's been best about it is that it's allowed me, it's forced me to keep my finger on the pulse of social media audience engagement. Like, why don't these videos work on TikTok so well? Okay, well, maybe I should figure out TikTok a little bit better, right? And you kind of just get a sense of what's going on in a way that you wouldn't otherwise. So it's, remember I told you earlier that some of my, some of the reason my career has gone well is because I had like reported on tech and because I had reported on tech and how I'd be a business person.

This in some ways is like a little, it's not reporting per se, but it's understanding and understanding how social media works and how the attention economy works and how distribution works helps me as CEO. Understanding the moment that we're in and the mediums. and the tools that we have allows you to strengthen how you apply that strategy to the business.

And I think we're starting to see other business leaders get it. Like you see venture capitalists starting a podcast or a scientist starting to go in a podcast tour and having that visibility and being present

because I mean, that is where the market is, that is where eyeballs is and it also forces you to continue to learn and stay on the pulse. And you have to, like, you have to,

you know, I'm very deliberate. So yesterday, I decided I had read this, I read this paper and it was about basically how AI bots, if they're competing in a game theory game against humans, dumb themselves down. If they're competing against other AI bots, they like immediately go to the Nash equilibrium in the most efficient spot, right? They assume that humans are dumb and bots are smart, right? And you have to declare that you're human or just know it's because we're just darn smart?

Well, you feed in and you say to the bot, you're competing against humans. And so then the strategy they choose is like, oh, basically the strategy you would choose against. Yep. Like if you're playing chess against the grand master group, you choose a different strategy. you're playing chess against an imbecile, you take more risk because you know they're not going to notice them, right? And so basically the chatbots were playing like at a dumber level when competing with humans, which is really interesting. What does that say about us? It doesn't say anything good, right?

But so I was then like, okay, I'm going to film this, but I just do one take. I put it up, I do it. And I do that both to sort of, because it shows authenticity, but also because you don't want your staff to be like, Nick spent 30 minutes like filming this video. No, Nick spent six minutes filming this video. You know, when I wrote my book, I actually didn't tell anybody that I was writing the books. I didn't want anybody to worry that I was distracted. And I wrote it all, like 95% of the book was written before 6.30 in the morning, right?

I just would get up early and write it every day. At what time? Sometimes even at six, right? I would just work on it for 30 minutes. Sometimes I'd work on it for 45 minutes. Sometimes I'd work on it for an hour. And you do that for five years. And the book gets done.

But it means it's never interfering with the Atlantic work, right? And I never touched it. I didn't make, you know, he didn't write a word while. in the Atlantic offices, right? And so what you can't do, if you have a job like mine, is you can't have your direct reports and your direct direct reports and your skip levels and all that think that you're distracted.

And that you're optimizing for yourself. Yeah. And if you do that, then you lose the trust of staff and lights out. Yeah, and you can tell that the way you manage your time between being an elite runner and then creating content and then running one of the most, the famous and legacy organizations,

in the U.S. So bringing into your book, you recently published a book, The Running Ground, this book is extraordinary. And I know running is the through line. Yeah. But it is, it's so much more than that, right? It's a book about coming of age, the complexity as a family, making career bets, setting ambitious goals, navigating life. I mean, I've passed this book to so many people. And it actually changed how I'm thinking about my own career, family goal setting. Yeah. So it's an extraordinary book. One of the things that stood out to me, so you run sometimes over maybe 100 miles a week, maybe more sometimes? Never more than 100 miles.

Never really more than 90. But a lot, yes, more than most people. So hours at a time too. Yes, that's true. And you don't bring a phone, you don't, or you don't listen to music. Right. You don't listen to a podcast. It's just you and your mind.

Yeah. What is the decision behind that to just be completely silent? So it's not 100% that I'm completely silent. Like when I commute to work and I don't care, I'm listening to podcasts, right? But yes, when I go out and I go and I run, like in the countryside or on a mountain, or I'm going for a five-hour run. A couple reasons. So one, when I run like that, I'm trying to clear my head, and I'm trying to meditate. Like, my job is so stressful and I'm on this to do list and I'm like moving things between the two lists.

I'm trying to be efficient. And like, you need a moment to let your mind wander, right? Like the stream needs to like just flow around and pick up sticks and, you know, so you need that. I'm trying to increase my awareness. Like a lot of times I'm like literally meditating while I run, right? Like I'm just listening for birds, right? Or I'm like counting my breaths or I'm counting my footsteps, right? I'm like literally doing a meditative practice.

And then the other reason is that to improve it running and you get faster and you get stronger, you have to increase the size of your heart, the efficiency of your blood flow, the muscles of your legs, the efficiency that you move through space. And a lot of that is neurological. And if you are listening to music, you are interrupting the learning process, right? And so what you're trying to do is trying to, like, teach your body how to move quickly through space over long distances. And you're listening to, like, a beat. It gets in the way.

And so I actually think that it disrupts your pacing, it disrupts your thinking, and it disrupts the learning. It's so fascinating that you say that, too, because I remember Jonathan Haidt talks about how these interstitial moments and we have nothing. in our mind. That's when the ideas come. That's when your brain can process things. And now we live in a world where we are so over-optimized. And you can't just cook dinner. You have to also cook dinner and learn about strength training. And you can't just go on the run. You have to also understand, you know, what is the latest in foreign affairs? But for you, it's also your brain, to improve as an athlete or whatever it is that you're doing, you also need to give your brain the silence to process and to grow in that activity. Totally. Which is really fascinating.

Totally. You also demonstrate in the book this extraordinary ability to hear your inner voice and to separate the critic that is trying to keep you small and out of fear put this upper limit. So it's afraid of failure, afraid of getting hurt. And then the voice that tells you, we shouldn't do this, we can't do this, and you know what to move past that. Right. Well, sometimes I do. Sometimes I don't.

I mean, the amazing thing in running is, like, I've been running for at a competitive level, for 35 years. Like I still have yet to do a workout where I haven't had these two voices having a conversation. And, you know, every time you run a workout, there's part of you that's like, this hurts, you know, probably just stop, right? You know, you said you're going to do four by a mile, but three by a mile is enough, you know, you'll get better. And then the other voice that is like, no, you know, you've learned in good habits. And if you do three by a mile today, actually, not only will that make this workout less effective, but it'll make all future workouts it's less effective because you'll have strengthened the habit of quitting. And you'll have, like,

weakened the habit of not quitting. And in fact, I often like moments where every moment is a learning experience in life and everything you do creates a habit, right? And so I like moments where everything is telling me to quit because I know that if I don't, I'll actually strengthen the not quitting habit, right? And so then I'll like kind of push it. The next workout, the voice told me to quit will come in like a little bit later. And so I sort of enjoy those. those moments, unless, of course, I quit, in which case that's a bad moment. The other thing you learn in running is that, you know, pain is not a pure physical thing. It's not just hydrogen ions building up in your blood or, you know, some people say lactic acid.

It's not just neuromuscular failure. It's just your mind sometimes just getting scared, right? And your mind will send pain signals wherever in the body it feels like your weakest or most susceptible. And so, you know, you have to use your mind to control your mind, which is a very complicated process. But you learn that and running and you get a little better at managing pain. And can you do apply that ability to start to the discernment? Okay, I can actually push past this fear to other areas of your life. Like if you're taking a job or trying to decide, is this next step for me?

I know you said that you'll set ambitious goals and then you don't think twice about it. You don't waste time being like, can I even do this? Yeah. But when you see that or you hear that inner critic, because we all have it, you can't take this leave. This is a little bit too ambitious. I mean, how do you discern it outside of the track? You know, I was talking to my kids, one of my kids about that where, you know, he plays on this great soccer team, right? And he plays on, we were talking about this last night in the car.

And, but he plays on one of the best soccer teams in New York, maybe the best soccer team in New York, but he plays on the B team, right? And he's one of the best players in the B team, but he plays on the B team. And we're talking about college and he's a sophomore. He's like, well, yeah, I couldn't, you know, am I going to be able to play at that level? Like, will I be able to go pro? And I'm like, I don't know. Maybe you will. Maybe you won't.

But if you don't think you will, you definitely won't. So, you know, sometimes just like letting go. And I have no idea. And at some point, trying to GoPro will be the wrong decision for him, right? But at 15, it's not, you know, like dream. And if you dream, maybe it does happen. Like lots of people, you know, improve at a certain trajectory. The, you know, most likely rational trajectory would be probably not.

But this thing happens in running, you know, and this is part of the book where I ran marathons. at 243, 243, 244, 242, whatever, for all of my 30s. You know, I couldn't get through that. I trained really hard. And I thought it was impossible to get faster. And then, lo and behold, at age 44, I ran at 229, which is ridiculous, right? It's a whole different, you know, it's an order of magnitude. I don't know, you go from, like, 300th place to, like, be much higher, right?

And suddenly you go from, like, being the best person on your block to being, like, one of the best people in the world. And how did that happen? Like, I was somehow blocked in my 30s. and thinking I could go faster. And so I do try to apply it also in work, right? Like, could the Atlantic get to 2 million subscriptions? You look at current trajectories, you know, we've gone from 900,000 to 1.4,

and I've been there, right? But going out to 2, if you look at journey, you model it out and political. I don't know, maybe you just set that goal you can get there. Now, maybe it will end up being a bad goal to set. But maybe it will end up being, like, setting the goal and just trying to do it, maybe we'll get there. And you say that you trust in the value of the value of, the compound interest of steady work. Yes.

Oh, I like that. You picked that. That's good. That's like, that's, I just, I don't know, there's so many moments in my life where if I go back to like what I got wrong in my 20s when my career was sort of not really on track and like versus in my 30s and 40s where it's been on a much better track, it's kind of that belief, which is if you do the right thing and you do it as well as you can and you do it with sort of trust and respect and dignity and

you do it every day. like good things happen, right? I remember I was saying that to my kids. I was just talking to journalism students at Stanford where I went to college and like, you know, what can I do in my career? Like, get to know the people you work with, edit them well, trust the process.

Like, if you do good work consistently, like you may not be able to see the good thing that will happen next, but it is much more likely that it will come. But if you get really stressed out that it's not coming or you have to have a giant breakthrough or, you know, then it will be able to be able to. never come. It's a little bit like promoting this book, right? So this book is selling well,

it's doing well, like people really like it. Thank you for the wonderful things you've said. It's not a New York Times bestseller, right? Like, it's not going to sell a million copies but maybe it will actually. Like maybe if you continue to go out and you talk to people and like maybe it will like reach the status here. So all you do is you just like, I wrote a book, incredibly proud of the book. Best thing I've ever written. I'm like very happy. I love the responses I'm getting. You just go out, promote it. You give it to people. You talk to people. You talk to a room here, you talk to a room there, who knows what happens. Yeah, that line really stuck out to me, because I think in this moment, too, we compare ourselves to so many people, and it seems like

the only version of life you see from other people is when they're at 100. No, everyone's only posting when they've made the thing. And if you, if you aren't where you think you need to be at a certain point of some fake timeline you've set, you maybe quit and try something else. Yeah. But you're saying it's just, it can look really steady, and you might not see the progress in real time, but it compounds over time. Okay, this is going to sound really strange and like apples and oranges or like apples and jet airplanes. But like one of the lines that influenced me the most, and I read it 20 years ago, my grandfather was an arms controller. And he spent his whole career, the entire Cold War, working on arms control.

And I remember reading something he wrote as either an interview or a quote he gave. And he was like, if you look at the future and you look at the odds, and he was a math guy. He's like, the odds that we will have nuclear war are 100%. Right? And if so, why keep going, right? It seems so clear. You know, like, what are the odds if there's an accident? What if there's, like, an accident?

It's a set off. Like, we have the ability to destroy the planet. The Soviets have the ability to destroy the planet. And then he said, and if you look at it that way, it's paralyzing. But is it possible that we're going to get through tomorrow and not have a nuclear war? It is possible, right? And is it possible that tomorrow we can do things that actually make it less likely

there's a nuclear war the day after. Well, we can do that too, right? Well, then if we can do it tomorrow, can we do it the day after tomorrow? Well, that seems possible to it. If you can do it the day after tomorrow, can you do it the day after that, right? And it was this amazing thought experiment where it's like, if you go out to a thousand, you know, of course it's not going to work. But if you look at one and two and three and four, actually you can imagine getting out to a thousand. And he took that philosophy. And a lot of people would argue that my grandfather, Paul Nitzha, increased the odds that nuclear war would happen. He made all the wrong decisions.

People marched against him. But, like, he inflected on American strategy. We had a strategy that changed every day the Cold War. And we didn't have a nuclear war, right? And we're now 75 years since the creation of the hydrogen bomb when it looked like everything was going to spin out of control. So, I don't know, pretty interesting idea. The calculus changes each day or the likelihood, the odds change. each day that you continue to work in the direction that you want something to go.

Yeah, right? So, like, writing this book, right? So, again, completing a, like, personal memoir about running not the same as saving the world from nuclear obliteration. Different scales, right? Yeah. But, like, it looks impossible. How can you really write a book if you're writing 30 minutes a day and try to run the Atlantic? I don't know. But can I do it today? Yeah. Can I get it a little closer? Yeah? Can I, like, set up some systems so maybe it gets close to the next day? Could do that, Right? And then, like, amazingly, oh, lo and behold, the book is done. And like, lo and behold, I feel pretty good about it. My final question, so you run these extraordinary marathons and your time improves and how you prepare for these marathons. I mean, it's what you eat. It's the months before you get a coach. You figure out the right shoes, what to wear. You look at the weather.

When you think about this future we're stepping into with AI, it is like running a marathon and we don't know how long the race is going to be. Yeah, it's true. We don't know what the terrain is going to be like. If it's going to rain, if it's going to be hot, what shoes we should wear, who's going to be running beside us. Yeah. So how would you prepare for a marathon that you don't know how long it's going to be and what that track is going to look like? That's so interesting, right? There's this race that they used to run at Carnegie Mellon where, you know, normally in a track race, like if you're running it, the way you prepare for a two-lap race is very different from the way you prepare for a four-lap race or one lap race, right?

And so all the runners would go to the track and they would roll the die in one through six. Okay, so that's the first half of the numbers you have to run. And then when the first runner completes that, let's say they roll a three. So the first runner gets through three laps, they roll the die again, one through six, and that's many more laps. So you might end up running a two-lap race. You might end up running a 12-lap race, and you don't know until you're halfway through. The way you prepare, and probably with AI, it's like you're rolling the dice like every four feet, right?

The way you prepare is you try to understand the rules as best as possible. They try to understand what AI is doing, where it's going, what's happening, right? You try to learn the tools as much as possible. You try to create some flexibility so that if, you know, if the race turns out to be 50 miles, you're able to do it. And if the race turns out to be one mile, you're still fairly optimized for it. And then you, like, you don't paralyze yourself with stress, right? When the dye is rolled and it's a six and you're prepared for a three, you just go and run six, you know?

That's all you can do. I'd be willing to just adapt along the way. You try to adapt as best you can. And, like, you know, it's also possible that, like, AI will, like, point the track downwards. I mean, who knows, right? So, like, the variables could change so much. And to try to build an organization that's flexible, you try to build trust within the organization, you try to build habits within the organization that are good.

And you try to, you know, move quickly in whatever direction you're going to have to move. And I think that applies to the individual, too. You try to learn how to adapt. You try to build the skills. You try to build trust within yourself. Yep. And then you're just willing to pivot as the story changes. So do you.

Yeah. You have to like, there's this weird balance between like, you have to stay true to whatever the core ideals are, but also be able to pivot, right? And those two things can be, you know, can be contradictory, but you have to like, what matters, what's most important, what's true to you as an individual or as an organization? And then from that point, how do you become most flexible? Where can people find your book? Where can they follow your work? Where can people find you? So there's a copy of my book right there. But if you're not in the studio. And this is the good copy. You should see my actual copy. It's like highlights. It's the pages are crumpled. So yes, there's a copy here. The book is called The Running Ground. You can buy it online. The best thing you could do is to walk into your local bookstore and say, hello, do you have a copy of The Running Ground by Nicholas Thompson. I've heard it's really good. And if they do, they'll give you one. And if they don't, they'll start stocking it, which would be amazing because that's kind of how the book will sell.

You can find me all over social media at NX Thompson on X, on Instagram, on LinkedIn, where I have my daily videos. You can sign up for the newsletter on LinkedIn or Beehive. And I'm all over the place. Nicholas, it has been a pleasure. Oh, so much functionate. much for having me here. Thanks so much for tuning in to this episode of I've Got Questions. If you've got questions, we would love to hear them. Submit them on our website, IGQ with shnatebebevel.com. And we'll see you next week.