“if you got to use ChatGPT to write, you probably have bigger functional problems that you need to deal with.” Jim VandeHei · Prof G Markets

let me give you a super specific example because I think this will give you a good lens into what I think will happen over the next three to five years. Take local news. Local news is an almost impossible business pre-AI. Why did all those reporters you talk about lose their jobs? Because the economics of local news went to crap. Well, we'll go back in time. Local news used to be: I have reporters in a building I'm paying for, printing words onto paper that's produced by a printing press I either own or lease. I'm then taking that paper, paying people to roll that paper into a ball, paying to put a rubber band around it. Then I'm paying a boy or a girl to put it in the back of their basket and throw it at your house. All of that to get the product is just the word, the word on that paper. All of that money to get that word to you. Then along comes the internet, takes away the newspaper, so you take away some of the cost. Still the expectation of like, oh, I'm working for a big company. I'm sitting in a big office building. You still have a lot of cost. COVID comes along. Now people are used to not working from home. Okay, now we don't necessarily have to have an office. We don't have to have a printing press. We don't have to have a paper. You still have all of the things that it takes to assemble news. Like when you be underneath the reporter is all kinds of activity, right? We've got to be able to produce that content. That content has to go through an editing process. Then someone has to match that to a visual. Maybe that visual has to be created. Then it has to be put on a website. Then someone's taking that and they're putting it into an Instagram format, a LinkedIn format. What if you could use artificial intelligence to do everything other than the reporting? So a reporter is building the content. They're doing their job. They're doing their reporting. But underneath that, you've created an AI that helps you do the copy editing. It's reading it. It's knowing that it needs a visual or a data visualization. It creates it for you. It puts all of that together. You then hit the button and it gets distributed to your website and you use artificial intelligence to break that into an Instagram version, a LinkedIn version, a Facebook version, a whatever comes next version. Now you've got mass distribution for your reporters at a very light cost. That's what we're doing with Axios Locals. We're now in 43 cities. We're building what we call the super system. We've been working with OpenAI and others to build out all of the stuff underneath it. I can tell you that each passing week, the stuff we're able to build is kind of blow your mind stuff. And the model that I just described will happen. A big part of it will be launched by the end of this year. But I think over the coming years, other people will see this and they'll be able to do variations of that. So even the work that you guys do, you have this, I don't know how you guys view your kind of podcast and content kingdom, but you have all of this. You still have a team of people that are spending a lot of time to try to get more derivative value out of that content. All of that is going to be done by machines. You're going to be able to do that if you haven't already done this. You might be able to get a bigger audience, might be able to find more ad slots for advertisers, and you might not need as many people to do that. And so I think that that is like how it's going to work. Where I don't think AI is useful or very useful for 99.9% of people is like, it's not going to write your story. It's not going to replace human journalists. It's excellent at research. It can be helpful with some of the writing. It's definitely good at challenging and helping you see some blind spots potentially in a story. But what I have found, and I would put Axios in the, not just media companies. I said we're probably in the upper 2% to 5% of non-Silicon Valley companies in terms of using AI. So I think we're lab rats. But I've watched journalists, like most of them find it at most modestly useful right now. I personally find it exceptionally useful, but I use it a lot and I've put a lot of deep thought into it over the last couple of years. But most of my reporters don't. They're like, yeah, it's okay. It helps a little bit with research. And that's fine because I want the reporters reporting. I want them talking to people. I want them breaking news. I want them applying analytical judgment that a machine can't apply, often drawn from understanding the people, understanding the historical context, understanding the piece of news as it sits in the broader context of today.

You mentioned that you think that AI doesn't really have a role in writing and doing the writing itself. Maybe it has a little bit of a role in terms of research. I mean, right now we're at a place where AI is pretty powerful at conducting in-depth research, collecting the research, putting it together, also helping with writing. But we also know that a lot of the time the AI is wrong. It says things very confidently that it seems accurate because of the way that it's presented, but sometimes it's factually incorrect. And then you say, hey, I think you're wrong. And then it suddenly reverses course and tells you a different thing. Are you concerned about that at all when it comes to just the way that we consume information, the idea that AI might be telling us things and we're just going along for the ride, assuming that it's correct? And in that vein, would you ever think about putting some policies in place around the usage of AI when it comes to your reporting?

I think this probably ties into your story of starting several companies. And I think something that's kind of interesting is you transitioned from reporter to guy who manages reporters into someone who actually started a business. And now you're more of a founder, owner, operator. Tell us a little bit about that transition from going from writing stuff to then starting companies.

It was wild, right? Like we just had this idea to start politico. We did it. Like it wasn't a lot of thought. I wasn't like, okay, one day I'm going to have to be a founder. One day I'm going to be an entrepreneur. One day I'm going to be a CEO. Say, I'm a journalist, man. Like, let's go do this. It sounds fun and I think we can do it. Well, once you do it, you realize, oh, this is like a real living, breathing company. It has needs. It needs management. It needs. Leadership. And so I had to learn to become a leader. And I had to learn about all hiring, firing, all these things. It's not like you can't read about it in a book and do it. You just have to do it. And I had never fancied myself. I was never an editor. I never managed anybody. I've said many times, like until I had to help run Politico, the only management I ever did was the night shift at Little Caesars when I was in high school. Right. And that's like, I don't know if that gets you ready to run a media company. And, but you learn, you put those reporting skills to work. You pay attention. I think I have the humility to know what I'm not good at and to pick up the phone and call people who are a lot smarter than me. A lot of people have been very generous over the years in helping me understand the things that I don't understand. I'm a relatively quick learner. So I was able to pick things up. I'm a very hard worker. I've always worked a ton of hours. And so I do the work that it takes. And it's kind of intoxicating. And to be honest, maybe a little addictive. Like I love it. Like I love starting companies. I love running companies. I love stories. I get just as excited when we break a story today as I did at Brilliant News or at Roll Call. And that's invigorating. And so, you know, now, yes, I guess we're, it's our second, we also have XSHQ, so maybe third company, and, you know, but I'm still running it. I still write all the time. I don't feel that different than I did 20 years ago, other than I know a lot more and I probably have more capabilities than I did 20 years ago. But I tell any young person, I would tell any old person, if you can be part of a startup, if you can take an idea, especially an idea that people say is crazy, and you can will it into existence, it is one of the greatest things you can do that's legal and that you get paid for. It's crazy. You learn so much about yourself and you learn so much about humanity. I know more about people, the good, the bad, and the ugly, because of starting companies and watching people in the good and the gory and the glory. Like you, you just see things and you become almost like a student. You're an anthropologist. You're like watching like this weird species of a human and like, how do they behave? What motivates them? Why do weird people do weird stuff all the time? I mean, that's the funny thing about being a CEO. By the time you hear about something happening with people, it's the weirdest shit you could possibly imagine. It's just wild. The stuff that like, wait, like someone did what? And it gets so weird that like by the time you're 55 and you're running your second company, almost nothing surprises me. I'm like, oh, okay, that happened. But it's a neat thing. It's a neat thing to do. And I think, by the way, I think that's going to be another advantage of AI. And I'm not an AI optimist. I'm kind of an AI realist. There's definitely days where I'm quite dark about what AI could do to society. So I just want to talk about some of the good parts, which are, I do think it's going to is and will be a lot easier to start a company. I think it is and will be a lot easier to be a one or two or three person company. And not necessarily you're trying to create a new chip to sell to NVIDIA. It'll be easier for my brother to run his construction company in Oshkosh. It'll be easier for our two plumbers to turn that into a franchise or maybe get rid of the stuff that most of the most of these guys don't want to do anyways. They hate doing bids. They hate doing the paperwork. It's just going to be easier. I think that could be really cool. I think that's a really neat thing. And I think given that it's kind of in the blood of Americans to start stuff, I think that that could be a very positive dimension to this. And so you guys talk to founders all the time. I'm sure you pick up a similar pattern of like, it's just, it's a, it's a neat thing to do. And, and I think a lot of people think you can't like, there's like, you've got to be special to do it. Like, I don't know. If you can work hard, you have an idea and you have passion around it. There's no IQ test. There's no, you went to a certain university. You didn't have to work at Goldman Sachs or do two years at McKinsey. None of that stuff matters. None of it. None. Your GPA, who gives a shit, right? Like it's like, do you have a good idea? Can you will it into existence? Do you have the grit? That's the stuff that matters.

You mentioned how you've kind of seen all the facets of humanity, really, from running a company. Because I think running a company, it's like you're managing people. You're just dealing with people. I think that's sort of what you explained there. What have you learned about people from managing companies? What about humans has surprised you, disappointed you, delighted you from running these companies and from managing all of these people?