← Channels with Peter Kafka
Loading summary
Transcript95 lines
- [00:00]
Peter Kafka
Why send your ducks and hope they land?
- [00:02]
Charlie Warzel
Share a PDF space for microbet. They'll understand A custom intro that's so clear. An audio summary that says what they need to hear.
- [00:11]
Peter Kafka
An AI assistant that makes it all
- [00:13]
Charlie Warzel
click so they'll understand it real good, real quick. Ain't just some files on the screen. It's exactly what your clients need to see. You can do that. Do that. Do that with Acrobat learn more@adobe.com do that with Acrobat if you're tired of endless scrolling to figure out where to eat, same I'm Stephanie Wu, editor in chief of Eater. We've just launched the newish and way better Eater app. It has all the restaurants we love, gives you personalized picks wherever you are, and serves up smarter search results just for you. You can find my list of the best places for martinis and fries in New York City, and save your favorite spots, share lists, follow editors and book right in the app. Download the eater app@eaterapp.com it's free for iOS users. Love don't cost a thing, but weddings sure do. I would say every single person I go to and I'm like, so how much over budget are you right now? And I've never heard someone say they were under budget. Matrimony's rising price tag. That's this week on Explain it to Me. Find new episodes Sundays wherever you get your podcasts.
- [01:41]
Peter Kafka
From the Vox Media Podcast Network. This is Channels with Peter Kafka. That is me. I'm also chief correspondent at Business Insider, and today we are talking about everything. Because we have the Atlantic's great Charlie Wurzel on, and Charlie's a guest you want to hear talk about everything. But to narrow it down a little bit, Charlie works at the Atlantic. He is one of the smartest observers I know about media and tech and their intersection. A perfect channel's Guest in the past, I would describe Charlie as one of the smartest writers I know. And that is still true. But now Charlie is podcasting. Welcome Charlie. So I needed a broader term, and in fact we're going to spend a lot of time talking about podcasting on this one. How it has become increasingly a video product and why it has become that. And how Charlie, well into a distinguished career as a writer and reporter, is now turning himself into a YouTuber too. Also discussed here why meme stocks like GameStop didn't go away, why it matters or doesn't if we live in a world dominated by AI slop, and why Twitter became an entirely new thing but defied predictions from some people that it would disappear. Okay, here we go. Here's me talking the Atlantic's Charlie Wurzo. I'm here with the Atlantic's Charlie Warzel. Welcome back, Charlie.
- [02:59]
Charlie Warzel
Thank you for having me. Appreciate it.
- [03:01]
Peter Kafka
You are a prolific writer and now podcaster at the Atlantic Galaxy Brain is where to go find all Charlie's great stuff. I want to talk to you about stuff you've been writing about and podcasting about. But before we get there, there's news that strikes me as a very Charlie Warzel story, which is GameStop trying to buy ebay. I know you don't cover a lot of straight business news, but this is not a straight business news story. This is a meme stock story. The reason that GameStop exists today at all is because it was the poster boy for the meme stock craze of the pandemic. And even though the stock has cratered from the crazy heights it went to, it's still way above where it was prior to it become a meme stock darling. And they also took their shares, their high priced shares and turned them into cash. So they can plausibly, I don't know if they can actually buy ebay, but they can at least make a plausible argument that they could buy eBay. GameStop, the company has done a little better than people had expected, but it's still not a high flying company. It's not the kind of company you would think could buy ebay long wind up. What is the fact that a meme stock company has at least a plausible chance of buying eBay in 2026. Tell us about the world.
- [04:19]
Charlie Warzel
I have been thinking a lot about the irrationality of the stock market, especially in the, in the Trump era. Especially Trump, Trump too. And you know, I think one of the most interesting phenomena has been the, the war in Iran, the oil shocks, these, the like, all the experts who know anything about global supply and oil and everything basically sounding the alarm bell like doing the kind of January there is a virus in, in, in China thing for the global energy, right? It, it's coming for us. You have no idea how bad this could get and is going to get. And frankly it's going to. Something has to give. And then Donald Trump goes on truth social and is like I'm going to reopen that straight. And then the stock market rallies knowing full well that he hasn't done anything to make that happen. It's just sort of like wish fulfillment, wishful thinking and the irrationality of that also the irrationality a little bit of, you know, the AI investment and worrying about that. And I have this. It feels to me like everyone has sort of internalized this lesson from really, like, Donald Trump's behavior. I feel like. And the first lesson of that was just that shamelessness is a superpower, right? Never apologize, do whatever you want to do. Don't.
- [05:50]
Peter Kafka
There's nothing you can't come back from, Right?
- [05:53]
Charlie Warzel
There is no PR crisis anymore, really. Right. Like, just double down, find the people who are willing to defend you, play to them, and, you know, information war is good business almost. And I feel like another lesson, like sort of a downstream lesson of that is if you build the narrative, the narrative can either become reality or it can be stronger than reality. And that's what I feel like is happening with a lot of things. I haven't really looked a ton into, you know, the GameStop eBay situation, but I feel like, to some degree, you know, if you can trot out the narrative, if you can get people to respond to it, if you can get people to treat it seriously or even in not treating it seriously, you know, do what we're doing here, which is like, turn the situation over, right? Try to analyze, like, what. What chance does GameStop have to actually do this? Can they pull this off? What does it mean? It's making them a really relevant character in this conversation. It's making people talk about the power of the meme, the meme stock, the narrative, getting people bought in. Maybe there's another chance for them to go on a Reddit run where people are like, yeah, we do want this to happen. This is so outrageous. We would like to find a way to propel them towards that. I think in that sense, there is this way in which the narrative is totally eaten reality. And we are watching a lot of people grasp and try to basically take advantage of that moment where nothing is true and everything is possible and there's
- [07:40]
Peter Kafka
reality and there's sort of financial reality, right? Which is you only have to be right today or really in microseconds to profit from you being right. You don't have to be right long term. You don't even have to be right in the physical world. You just have to find someone else to take the other side of that trade, like the Trump company, the spac and all that. The spac, right. So that has collapsed, but it's still worth billions and billions of dollars. And in theory, some people are holding that stock because they like Donald Trump. Other people like it because they like the idea of the pivot they've announced to becoming a nuclear. I always confuse the two. Fusion or fission company. And again, none of it has to be right at all. It just has to be right long enough for you to sell your shares to somebody else. And maybe that is the lesson.
- [08:29]
Charlie Warzel
It's possible. You see it too, with the Allbirds pivot, right? The sort of selling off the actual shoe company and then pivoting to the thing and the stock totally. It rises off the backs of the idea that they're just going to lease out their stuff to data centers or pivot and take their infrastructure and turn it into whatever. And that felt both outrageous and at the same time, like, I don't know, maybe. Maybe I think something has broken into.
- [08:58]
Peter Kafka
That one does even have a tiny bit of logic to it because there are people taking old real estate and turning it into data centers. I just saw when I was in Minneapolis a few months ago, base. It's called the Sleep Number Headquarters, the Sleep Number building. But it's where they actually make the Sleep Number beds. And they're selling that at a huge, huge markup because apparently it can be a data center. And the suggestion was maybe this will be what happens to lots of downtowns is they'll take their old unused office buildings and turn them into data centers. That's very dystopian. But at least it works in the moment.
- [09:34]
Charlie Warzel
Yeah, we know what to do with malls now. Right?
- [09:37]
Peter Kafka
Let's keep talking about AI. You write about Internet culture. What AI does to the Internet culture is obviously a core concern for you. You've been writing a lot about slop and veracity, and I think they're. I don't know if they're two sides of the same coin, but they're related, right? There's did a robot make this? Did a human make this? Or did a robot make it? And do we care, I guess is the second part of it. And I think you approach all these discussions from the idea like, obviously we need to know if a human made this. Obviously we should know that. And kind of to be a devil's advocate, but kind of because maybe I think it. I wonder if we don't need to know that in lots of cases, there's cases where it obviously matters whether the thing is true or not and. Or if someone is representing themselves and it's actually a robot doing it. You could imagine circumstances where Donald Trump or a fake Donald Trump shows up and it really matters which one is real. But in a lot of cases, I would argue maybe it doesn't matter if the head of sales at X Company had ChatGPT write his LinkedIn post. Do we care?
- [10:47]
Charlie Warzel
Well, I would say that it's dependent on the situation. Right. I had a chat with Max Spiro recently on my podcast who was one of the co founders of Pangram, which is the AI detection software that some people believe is the gold standard. Other people think all of AI detection is kind of bogus. But he is very obviously interested and committed to trying to tease out this question, figure out what's what. He built this browser extension that lets you look through like any Reddit or LinkedIn or X post or substack or whatever and try to identify. And I, you know, it goes through my head in that moment is like, oh, this is going to just trigger this, this witch hunt, right? Where it's just like people are just going through everything.
- [11:33]
Peter Kafka
Oh, that's already started.
- [11:35]
Charlie Warzel
Well, it has, it has, but, but not everyone has access to, you know, the tools and not everyone's thinking about it at this, in this moment.
- [11:41]
Peter Kafka
So I think he's like, I will tell you based on my, my kids high school experience, which is whenever something weird gets published or distributed or someone who is in disfavor, write something thinking of a specific person in this case as a memo. And some people say, well, that must be AI because it doesn't make any sense. And then the kids themselves are putting their homework through AI checkers. Presumably they're not using AI to do the homework, but the AI checker says there's AI in here. The kids then go and redo their work to make it less likely to trigger the AI check. Mostly means taking out EM dashes. But anyway, the point is that in certain parts of the world people are already quick to say that's AI if they don't like something and they're quick to worry that their work will be found as AI.
- [12:27]
Charlie Warzel
Yeah, I think that's, that's obviously totally correct. And anyway, so he's very invested in this and I, you know, I kind of posed a similar question to him of does this really, like at what point does it stop mattering? Because it's just there's so much out there and it's just so difficult to tell in this process. Right. And I think he broke it down a little bit to say it seems like we're probably going to be living in a world where there is a lot of the writing that comes out of that's not like stuff from professional writers, but it's stuff that comes out of Marketing material or again LinkedIn, CEO speak or whatever kind of thought leader stuff, posts on social networks that it's all going to be AI assisted to some degree, that people are going to either write the thing themselves or come up with part of it. It's human generated ideas that then get fed through because it's just quicker and more efficient. And these are sort of low stakes things anyway. And that future seems plausible to me. What he really is against, and the thing that I think is I'm pretty much against is this whole cloth creation of minimal, minimal, I kind of don't even care effort in order to put out this slop. So this is what I would say is like the AI SEO farm scam sort of websites that exist to prop up a certain topic, to do whatever is left of banner ad click arbitrage where it's just like I'm going to flood the zone with the 10 best smoke detectors, blah blah, blah. And it's like AI slop Lorem Ipsum. It drowns out the people who actually go out and try to tell you what is the best smoke detector. Because in a pinch you want a smoke detector that's not going to malfunction or whatever that it seems like a really relevant, we should care about the slop when it is choking out the usefulness of the Internet, the utility of the Internet, which is to connect people with information that is good and that you can vet and anywhere. And the other is this making stuff up out of whole cloth in the laziest way. So I think a LinkedIn thread from a CEO where they're like, yeah, I ran it through ChatGPT or whatever, very who cares? But I think the people who are building like hustle porn, you know, posts that are like how to optimize A, B and C that is just sort of like you put in a prompt to ChatGPT, like how can I get 45 people to sign up for my class plan? You know, I want 85 different tweets with you know, 18 bulleted lists of whatever. And I think like when people are putting in that like that minimal amount of effort to sort of either scam people or hustle people people out of their money or basically try to build a business in an hour type thing, I think that that is really just generally shitty as an experience and a little bit poisonous. And I think that is the type of stuff which then can be obviously transferred over into politics, right? Where it's like how do I build a very. Claude, build me a very effective propaganda campaign. Against AI data centers, blah, blah, blah, blah, blah.
- [15:47]
Peter Kafka
So we're except. But everything you're describing to me sounds like an acceleration of stuff we already had, right? We had Macedonian people making fake posts in the 2016 election, right. And, and Google has been a shitty place to find all kinds of product related stuff for a long time because there's a lot of value in jamming up the search results with, with shitty product stuff. And you would have thought, I would have thought over the years Google would have done something about. Even before we were worried about AI stuff clogging up the feed that Google would take would be more careful about pruning that. And periodically you'll hear they've made some announcement about the stuff and they're cleaning it up, but it's still full of it. And when the AI boom kicked off in 2023, I was talking to Neil Mohan at YouTube about this because it seemed obvious to me that they'd be flooded with crummy. And I don't know what his real answer is, but the one he kept giving me because I asked him about it repeatedly was basically, if people like it, we'll elevate it. And it doesn't really matter if it's AI generated or not. And if they don't, like also doesn't matter if it's AI generated or not. Again, I don't know what he really thinks, but there is some logic to it, right? Which is if people like the crap, we'll give them the crap. That's, that's what we're in the business of doing. And I always, this is kind of a fundamental tension of the Internet, right, Which is do people actually want what they're getting or is, are the platforms incented to give them a certain kind of thing that the platforms like or that solves a short term answer for the platforms and doesn't, doesn't, doesn't solve the long term, very obvious problems.
- [17:42]
Charlie Warzel
I think this is definitely an outgrowth of the platform stepping away from any kind of editorial role, right. Like this is like this has been fully realized since, you know, late 2024, where all these platforms are like, oh, you know, thank God I don't have to moderate.
- [18:00]
Peter Kafka
We never want this to begin with, any of this.
- [18:02]
Charlie Warzel
Yeah, we didn't want to do this to begin with. This is so like, I really think part of, and I've written about this, part of this symbiotic relationship with Donald Trump. There's a lot of reasons, right? But one of them is like the companies themselves don't want to have to go through the hassle of setting these standards and having to deal with the shifting cultural values and norms. They'd much rather just let it rip. It's so much easier for them. It's more lucrative in a lot of senses for them to just let it rip and let people put whatever they want.
- [18:30]
Peter Kafka
And there was ideology with it. I don't mean left, right ideology. I mean, we should not be in the position of telling people what they like. That's the great thing about building these software systems, is the people will tell us what they like.
- [18:43]
Charlie Warzel
Yeah. And also, it just blows up in their face a lot easier. Like, you know, that's how you get dragged in front of Congress is when you start moderating things, you don't get dragged in front of Congress as much. Maybe, you know, in a different administration, who knows? But if you just sort of say, like, we're letting the people decide anyway, I think that, that, you know, that's very much the ethos of the moment. I don't know that that's gonna be, like, the way that we decide we wanna do intern on these platforms. I could conceivably see a tipping point of, like, man, we did let it rip during the sort of period of the AI boom, where everything just got flooded. And now we are dealing with this wild west. Like, we actually do need to clean it up. Right. It's sort of like an industrial revolution to, like, okay, now the skies of London are black and choked with smoke.
- [19:32]
Peter Kafka
It was not an unforeseeable consequence. It was something that. That we could track. But yet we did it anyway.
- [19:40]
Charlie Warzel
Yeah. And so I think you have a little of that. I also just gotta believe that it's fundamentally, I think, and I'm not a platform executive, obviously, but it's fundamentally a little strange to me to build a company with thought and care and wanting to be, you know, YouTube is like the destination for, like, news, information, content. It's. It's like, in some senses, it almost is the Internet now. Right. Like, it is where so many people are going for everything. To build that out, to care about it and then just be like, whatever shit people want to watch, like, we'll feed you. Like, you know, come here, piggy. It's the slot bucket. You know, it's like, that's an odd, like, way to approach a business. You'd think you'd want to say, like, no, actually, what we want to do is we, you know, we don't care as much about ideological content. We're Sort of an equal opportunity provider. But like, we do care about quality. And I think that that's if you pose it to them in terms of quality. I don't know, it's just a very odd thing of if the people like it, we're going to give it to them.
- [20:54]
Peter Kafka
Even if they tell you they like quality, when they're pitching in front of advertisers and they want to tell blue, blue chip advertisers that we're making great stuff here that you can feel comfortable having your ads against. I do think that there is a long seated thing and it's not a Trump era thing. I think it really, really, it goes all the way back to the beginning of the platform. So they said these are inherently better than east coast establishment institutions that tell us what we should think and shoot. Now this is obviously when we thought if we let people, if we left people to their own devices, stuff good stuff would come out. This is the 2008, 2010 era, so it's easier to be comfortable with this. But it was pretty deep seated and I think that has remained, to be
- [21:41]
Charlie Warzel
fair, a lot of good stuff does come out. Like, I think a lot of people are consuming low quality garbage.
- [21:48]
Peter Kafka
Sure.
- [21:49]
Charlie Warzel
And that is, you know, people were consuming low quality garbage before the large language models. My, like hopeful vision is that, that this period of honestly, kind of a little bit of lawlessness in terms of how we govern the Internet, what we're doing, let it rip that era, I do think these things sway or move back and forth. They boomerang back and forth. And I do think that there's a feeling among just normal people of not loving, being surprised by something that you thought was real and it turns out to be fake. A kind of exhaustion in that way. An idea of wanting things that are made for humans by humans or just, you know, I think AI does not poll very well. Like people say they don't like AI content. And I can see a period of time where things, a lot of the platforms decide that it's in their best business interest to actually separate this stuff. Right. I think certain platforms are already saying it's so hard for us to identify slots, but we can identify verifiable humans who are making good stuff and we can put a badge on them. I can see a world in which we do swing a little bit more towards prioritizing that stuff because that is actually what people are going to yearn for. They're going to desire that.
- [23:16]
Peter Kafka
And maybe the middle ground is kind of the market solution. We already have. Which is if you want good stuff, however you define good, made by humans or farm to table restaurants or whatever partisan made goods, you can go get those things. You'll pay a premium for them. You probably have to be in a specific demographic group to even know they exist, afford them, you can go get them and everyone else can get the free shitty shit. And they can kind of know that's what they're getting. And some people won't know and. Or care.
- [23:48]
Charlie Warzel
Can I give you one more thing on that that just makes me think. I think a big problem on the Internet, especially a problem that people like you and I who are trying to write, report and comment on all this stuff, is that nobody really knows what everyone else is actually seeing. And there's all sorts of metrics out there that suggest that people are viewing whatever. But view metrics online are historically pretty shoddy. On places like X, it's very difficult to tell what constitutes as a view. And now you can see them on all the videos and all the tweets and the live streams and everything. And I think sometimes when you're analyzing this information landscape and you see 12 million views on a piece of AI generated garbage that just is, quote, tweeted everywhere and then screenshotted and put on Instagram and all these types of viral pieces of content that are mostly rage, bait or slop or whatever, there's this way of saying, like, this is the dominant culture. This is the thing that people are consuming. And the reality is they're swiping by it and they're pausing on it for two seconds or they're, you know, quote tweeting it and then saying like, this sucks. And then forgetting about it completely. And I think there's this idea with a lot of this. This stuff is like, oh, everyone is consuming. They love the slop, they love this. And it's like, what is love this really mean? I think so much of the consumption is so passive that it's tough to assign this value of, like, no, this is like the fans are crying out for this. When it's like, this isn't really actually popular. It's just popular in the sense that, like, it glanced by someone's feed a lot.
- [25:31]
Peter Kafka
We'll be right back with Charlie Wurzel. But first, a word from a sponsor.
- [25:38]
Charlie Warzel
I'm Maria Sharapova and I'm hosting a new podcast called Pretty Tough. Every week I'm sitting down with trailblazing women at the top of their game to discuss ambition, work ethic, and the ups and downs that come on the path to achieving greatness. We'll dive into their stories and get valuable insights from top executives, actors, entrepreneurs, and other individuals who have inspired me so much in my own journey. Follow Pretty Tough wherever you get your podcasts. I'm Mitch Firth, two time Indigo Champion, championship mvp, and forward for the US Women's National Team. Before I went pro, I graduated from Harvard with a degree in psychology, which comes in handy more than you think. Any athlete pursuing greatness knows there's a certain mentality you have to have. What people don't know is what that costs. In my podcast, Confessions of an Elite Athlete, I sit down with the best athletes in the world and explore the psychology, mindset, and unseen battles on the path to greatness. So take a seat and learn from the Confessions of an elite athlete on YouTube or wherever you get your podcasts.
- [26:55]
Peter Kafka
So we are 250 years into this American experiment, and I'd say it's going okay. I give us like a C. There
- [27:02]
Charlie Warzel
is no perfect past, but there is also no exclusively negative past, because humans are gonna human. That's what we do. I think the story of America is the struggle of people who have not been included in the promise of America to expand those principles to include more people.
- [27:23]
Peter Kafka
What's gonna determine the next 250 years of America? And how do we write a new social contract that can give us the democracy we deserve?
- [27:31]
Charlie Warzel
Okay, so I'm just gonna be a jerk here because I'm a historian. So we have to have a prologue explaining, you know, we the people.
- [27:39]
Peter Kafka
Okay. You know, I do still remember it from Schoolhouse Rock. We the people anointed the former perfect union, established justice. What is it? Ensure domestic tranquility.
- [27:50]
Charlie Warzel
So you're talking about a foundational document. So I'm building a document that will protect American democracy.
- [27:57]
Peter Kafka
That's this week on America Acts Relief. And we're back. You are now making a video podcast. I am the last podcaster not to be on video yet. And one of the things you always hear about, the reason we should do video podcasts is because people love to consume video podcasts and they. The numbers bear it out. Look at YouTube. Look at these huge numbers. I'm sure there are people who are watching this stuff, but I don't, I don't. I haven't done a full survey of literally everyone I know, but I feel like the people in my life, they're not podcast video consumers. We think there's a real audience, but how big is that audience for people who want to watch people talk on YouTube.
- [28:44]
Charlie Warzel
Well, I think there's a lot of people. Again, I'm going to go somewhat anecdotally here, but there are lots of folks who are younger, and by younger I mean like millennial and down, who are putting opening up. You know, they have a pretty finely tuned YouTube algorithm for them. They've got the people that they know they like, and they're opening up the little like podcast tab on YouTube and they're letting it play in the background of their day. The first episode we did of our video podcast was with Hank Green, the sort of famous YouTuber. And I had an acquaintance of mine who I was just talking to, and they were like, oh, dude, yesterday I was just sitting there working, and all of a sudden I heard your voice. And I wasn't really even listening to the podcast that was in my feed, but I just clicked over to the YouTube tab and there you were talking to him. And it's. You know, it got recommended to him because he watches every single Hank Green video in the background right of while he's just doing his job, working, doing whatever. So I think there's a lot of ambient watching.
- [29:49]
Peter Kafka
I've heard that suggestion brought up. At the same time, YouTube will say, Listen, more than half our consumption now is coming from televisions, suggesting that people are sitting on the couch and choosing to watch this stuff. And again, I can imagine you choosing to watch lots of YouTube videos on a big screen, but watching you and me talk, I can sort of imagine it, but it doesn't really sync up. And I see a ton of video podcast content, but it is all clips in my TikTok feed or my Instagram feed, and I see a ton of it. I can tell you what Megyn Kelly was talking about recently and six months ago, but it never compels me to go watch the actual podcast, which I think is also, from what I can tell, statistically true that people aren't. The idea of you're discovering a podcast on TikTok or Instagram in clip form and moving to the main thing does not happen that much. They seem to be different vectors.
- [30:43]
Charlie Warzel
Yeah. I spoke to Ed Elson, who's a podcaster himself. He wrote recently about this idea that the clips are actually the real product here, not the podcasts, especially when it comes to streamers. Live streamers.
- [31:00]
Peter Kafka
Hasan biker.
- [31:01]
Charlie Warzel
Yeah, the Face Smashing guy. Clavicular. But one that I thought was really interesting is the podcast or the live stream show, tvpn, which is the technology show. I think you had them on your podcast Once before that got bought by OpenAI for an undisclosed but, you know, pretty large sum. He did like an analysis of just, you know, 10 of the TVPN videos and clips and saw that the, the live views of the show on X T VPN, it's about like averaging about 7,000 views, whereas the clips were averaging like well over 250,000 views. Right. That the clips are the product that are the thing. They're making the show in some senses for the people who do want to watch it, but knowing that it's a small core audience that they can engage with, they can throw the ads up, they can do that. But the real game there is to have the clips identify these best moments, advertise around the clips, make money off of that. And basically this idea that you are playing for that casual audience, that glancing audience, because that's how all this works. And I do think that there's a lot of truth to, to that. I don't love it at all. But I think that there's a way in which these really longer form podcasts or video things, what they're ultimately doing is waiting to find the moments that are the best to strike and then those moments can go viral. And yes, you can recruit some portion of that to new audience. A very miniscule part, but a big part of it is you're trying to become popular in the clip realm.
- [32:42]
Peter Kafka
I mean, look, I started watching Chernobyl, the HBO series about Chernobyl, not an upper great show because I was seeing clips of it on TikTok, so it can convert you. The remaining late night talk shows are all built basically for Internet distribution and clips, which is why I think SNL has had a bit of a renaissance, is that it's so clip friendly you can consume it all in two minutes. What's your experience like going from, from serious guy writing to serious guy making podcasts? How does it change what you talk about, how you think about it?
- [33:20]
Charlie Warzel
Okay, the real reason I wanted to do this and do it as a video podcast was not really just because that's where everything was moving. Even though that's, you know, it's helpful. The reason I wanted to do it was because I write about these platforms forever and I have always found it was sort of a lesson. I was at, I was at BuzzFeed for like seven years. And what I always really loved about working there and the ethos of the place was like, we are covering this thing while living it, right? We are writing about the rise of these platforms and virality and the way that it's shaping culture and politics and all this stuff. But we also have to live this. We have to play around with these platforms. We have to find ways to recruit audience, see what works, and that is going to generally inform the work that we do. Writing about this Stu, is especially true, covering technology there. And there's a little bit of that in this, right? YouTube Is this incredible engine for content, but also has, as we've seen now, so many downstream effects in the way our politics work, the way that just broadly culture works. And so I wanted. It's a platform that I don't. I spend time consuming, but I never spend any time as a creator. I don't know what the incentives actually are. I don't know how it works to grow. I had no idea, if I put this stuff up about things that I find interesting, whether it's going to do really well or do really poorly. Because I come with a writer's sensibility. And I think that I've found that it's a completely different animal, which is not super surprising. But I think there are all these ways that YouTube works as a creator that I have found really fascinating. The Atlantic has a very large YouTube channel, right? You know, the Atlantic was early to YouTube. So there's this huge base of people, and a lot of them are coming for a very specific type of thing, right? A very specific kind of conversation about democracy and politics in this way. Again, this is just anecdotal from my experience. I think they're coming for people who are a little more established, a little bit more distinguished in their careers than, like, me, the guy with a beard, who's just, like, yammering about, you know,
- [35:40]
Peter Kafka
clips, let alone a Mr. Beast, right?
- [35:43]
Charlie Warzel
And so there's this way in which I feel like we have this group, and yet I have to sort of endear myself to that audience of our subscribers. And if I don't successfully do that at the outset of a video, right, if the video doesn't immediately sort of grab that group of people, it's going to be less likely to get thrown out into the broader pool of, like, people who don't subscribe to the channel. Channel, right. So there's this way in which I can see that other people who are creators may be getting, like, audience captured very, very early, right? It's like you develop a reputation, and this is similar to a lot of other places, but you develop a reputation and people subscribe to you and they want a very specific product. And if you don't give them that product, YouTube sees and identifies that and says, okay, that's not, that's not their best work. So we'll send it out to the rest of the world. But not in the same way that we would. If you are lighting your subscriber base on fire and they're like, yes, this is amazing. We love this. YouTube's like, okay, let's get a bigger audience to this and see how it goes.
- [36:49]
Peter Kafka
In your mind, you're making content for the Atlantic's core audience first, and that's the only way you have a hope of getting to a larger audience. Right. So. But if that's the case. So in my mind, the things that work best on YouTube are very unatlantic. Like things. Right. They're. I mean, if you just look at the most popular things on YouTube, they're nothing like what you, you consume from the Atlantic down to the aesthetics. Right. We were talking before we started recording about the YouTube face, which is what you see in all the thumbnails of all the top. You, yeah, it looks like you're in distress, you've been shocked, or maybe you were on a toilet or something. Like something is going wrong. And those are the faces and they're very unflattering faces. And all of the top YouTube people choose those faces to represent themselves. You don't. On your YouTube thumbnails. I was looking at that. I'm assuming that is some aesthetic choice for you, but maybe it's also a choice that my core Atlantic reader viewer doesn't want that isn't going to respond to that. Again, long winded way of saying, is there a world where what works for an Atlantic consumer is so radically different than what works for the rest of YouTube that you're going to constantly sort of be bumping into to that limit?
- [38:04]
Charlie Warzel
Well, I think for me personally in this experiment, I would say first off that like my North Star, in making this show with our great team is just doing stuff that we think is interesting and that is quality and ideally, you know, the sweet spot. The things that have worked the best for us have always been. And it's obviously no, no surprise. It's stuff that someone who's very versed in a piece of technology or a piece of Internet culture can find something new and interesting to latch onto, but also someone who's coming to this cold can find, you know, so something that is like, can work on a number of different audience levels. I don't think that the, the content that is popular on YouTube necessarily is at odds always with what a Legacy media organization like, say, the Atlantic, with the standards and the practices that we have, I don't think it's at odds always, because I also think the largest growth sector for YouTube over the last year or last bit, I should say, is boomers, is older people are coming into this. A lot of people who are, you know, maybe now just being retired, there is a bit of like a cable news replacement situation happening where like msnbc, fox, cnn, they were on, you know, the TV kind of going along all day in the background. Now it's autoplay videos from these people on YouTube. You know, like, I have found that, you know, I get a lot of feedback and that's definitely like a big component of people who are watching. And those people are Atlantic subscribers, End Time subscribers. And, you know, obviously there's a whole slew of this on the other side of the ideological spectrum too. So that crew is not coming from Mr. Beast style thumbnails. Right.
- [40:10]
Peter Kafka
And YouTube is big enough to contain that and serve that up to them.
- [40:14]
Charlie Warzel
Yeah, but in terms of the thumbnail thing, I mean, this is part of the experiment, right? Like, we have changed the thumbnails slightly. We've gone from, I think, like a black and white stylized to another thing. These subtle things really do change it. And I think to go out a little beyond just my experience, somebody who I've talked to some about all of this, about how do you break in to try to do journalism on YouTube in a way that is both native to the format, but also taking some of the best from it. No one, I think, out there is doing it better than Pablo Torre, who, who is a, you know, is doing investigative stories, but also doing, you know, by his own admission, like, more lower brow, just sort of like, you know, culture, pop culture stuff, deep, deep dives
- [41:04]
Peter Kafka
into Bill Belichick and his young girlfriend.
- [41:06]
Charlie Warzel
Right. Stuff like that. But if you look at his thumbnails, I actually, I was talking to somebody about this the other day. They're like a perfect hybrid of like, mine and Mr. Beast's, right? Like, it's not. Not scammy. It's not, you know, it's not like, hey, fellow kids, you know, like, trying to be someone that he isn't. But there is also usually, like, something going on. People are making animated faces. There is sort of a, you know, a thumbnail that is, you know, enticing in that way. And I think that that's, you know, we will eventually probably settle on, on a way to merge those forms and find the thing that works well is
- [41:53]
Peter Kafka
making Video changing your thinking and or reporting. Are you more likely to pursue something because you think, oh, that would be a good video one day or this topic I'm really interested in. I cannot imagine how we have an interesting conversation about it on YouTube, so maybe we won't do it. Or even if you don't reach those conclusions, are they in the back of your head?
- [42:15]
Charlie Warzel
I think something that's hard, that has bummed me out is guest selection in the sense of if you have somebody who is a stone cold expert, really great, knows everything, is not somebody who like, who presents. Like sometimes you'll have experts who present as like kind of anxious. They don't want to be in front of the camera. They don't like, they don't like that they're not, not excellent Communicators in a YouTube and video centric way. There's a way in which that can, can hurt. And you don't get that in audio. In video, you put them in front
- [42:51]
Peter Kafka
of a camera and they freeze up. And this has always been an issue. You know, I did do podcasts and conferences. What kind of person is good on stage? Right. There's a lot of people who are interesting and smart and accomplished, but they might not be dynamic on stage or they might have a heavy accent, which makes it difficult to listen to for our audience. And even though you try to counterweight against that, that affects who you program for. And I'm sure it's even more so for video.
- [43:20]
Charlie Warzel
Yeah. But broadly speaking, I do think, because people are, you know, you don't know how exactly they're consuming this, whether it is like third screen. Right. Like it's just like on. In the background. Like, I don't think think a ton about. We have a producer for my show, this woman, Renee Klar, who really understands and consumes a lot of YouTube herself. And she's so good at accenting these interviews with charts and clips and other elements that I think really do add to the overall experience and make them kind of sing as video. But also I am not, not overly at this point as I'm trying to like, if I'm doing it a straight interview, I'm not overly worried about, like, you know, okay, is this, you know, is this subject really great for it? And I find that, you know, some of our wonkier videos that are, you know, aren't necessarily like just me and Anil Dash talking about, you know, AI and the. In the current AI moment. You know, it was a really great performing video for us. It's just Two dudes talking about AI. But it found a really cool, good portion of the Atlantics audience and then also just like, you know, went wide and found a good portion of people there. And I think it's, that's just because it's a, it was a good conversation where, you know, we, we got someplace new.
- [44:49]
Peter Kafka
I'll be right back. But first, a word from a sponsor. Hey, I'm Matt Buchel, comedian, writer, and floating head you may or may not have seen on your fyp. And I'm starting a brand new podcast. Wait, don't swipe away. It's called that sounds like a little lot. I'm going to start by breaking down whatever insanity is happening in the world and then I'll sit down with a comedian or actor or writer or honestly, anyone who responds to my DMs. This is not the place to get the news, but it is a place to feel a little bit better about it. You can watch on YouTube or listen wherever you get your podcasts. That sounds like a lot. Part of the Vox Media Podcast network,
- [45:27]
Charlie Warzel
Elon Musk spent most of this week sitting in a courtroom litigating some of the most important moments in the early history of, of the AI revolution. He didn't do a great job. And the ways in which he didn't do a great job may come back to haunt Elon Musk in a pretty big way. This week on the Vergecast, we're talking about what's going on in Musk vs OpenAI and how it might affect the rest of the tech industry. Plus the most exciting laptop we've seen in a while and maybe the most exciting game controller we've seen in a while. All that on the Vergecast. Wherever you get podcasts.
- [46:09]
Peter Kafka
And we're back. Let's keep talking about you, Charlie Wurzel. As you mentioned, you were at buzzfeed, then you went to New York Times, then you really had a great job, I thought, at the Times, doing really cool opinion stuff, cool reporting projects, tons of prestige. And you said, I'm going to go do substack. And I don't, I don't think you were on Substack more than a year before you went to the Atlantic. Maybe it was a year. You tell me it was like nine months.
- [46:32]
Charlie Warzel
Yeah.
- [46:32]
Peter Kafka
Okay, there you go. I often on this podcast talk to solo proprietors and substackers and people who are increasingly successful on their own. And then the other half of my podcast is talking to people who run media companies and often. And they want to hire those People or somehow participate in that and they are kind of stuck. They can't figure out how they could work with an Emily Sundberg, for instance. Like, you know, what could they pay her more than what she's already making? What could they offer her that she already doesn't have? So you've now been on both sides of that fence. How do you think about it from a solo proprietor's perspective? And how would you, if you were running a legacy media organization, was trying to work with these folks, how would you think about that?
- [47:21]
Charlie Warzel
The way that I see it, or the way that I've experienced it over the last five years, or actually longer than that, I guess now is that as a person who is making the content, right, who is doing the journalism, whatever, I am perpetually struck by the feeling that the grass is always greener, like wherever you are, right. Like I'm always having FOMO about something. I watch certain people who are independent creators responding to news moments in this real multi platform way. So quickly. Just, I've got some thoughts. Boom, here they are. Perhaps they're disorganized. Who cares? The best stuff will find its audience. What matters is being in this part of the conversation as quick as humanly possible. And I think that that's really helpful, right? And that the business of, of quality control, which is also really important that institutions like the Atlantic or the Times or wherever, that means it's harder to be as responsive in terms of how to work with, how there can be more of a fusion of this. I don't know. The economics are so weird for creators because I've thought about this myself. The thing that I don't, I didn't personally like about substack. I actually really loved the freedom. I loved the one to one with an audience. I loved all the different sort of like the YouTube stuff, like learning from every single thing that you do because you're the only one who has, you have the God mode visibility to how your stuff performs. I love that. But I did not like just being alone. I love bouncing ideas off people. I love collaboration, I love writing stories with other people. I think that that is how I tend to work best. And there was a point in the substack universe where I was growing my thing and it was like, well, could I invite another writer who's great into this thing and could we do this together? And it's like, well, obviously you'd want somebody who does something sort of similar, right? I don't want to just get somebody who writes about know ballet or Something like, I want some cohesion, which means that we have similar audiences already, probably. So you're probably. If you add them into your media company, you split everything 50, 50, so you're taking a 50% pay cut. And then you're like, so what is the recruitment on that? Well, if, you know, if I bring in somebody who's doing tech journalism stuff like, chances are it's not one to one, but the ability to grow your audience isn't quite as much. Now, you could make the case that that means we'll double how good the work is, the value for a subscription will go up, and therefore we should be able to do it. I think, though, that it's a bummer because if I could have found someone, I may have stayed on, but the economics did. Did not work for me at all in that. And I love the collaboration that you get. You had Joe Weisenthal from Odd Lots on, and you kind of talked a little bit about this whole, Joe, why
- [50:42]
Peter Kafka
don't you leave Bloomberg and go off on your own?
- [50:44]
Charlie Warzel
Yeah. And I honestly think what Bloomberg has done with Odd Lots is a really excellent example of how it's a good template, because that is a show that if you just, if you're describing it to somebody and they have no context, you might be like, who's the audience for that? Right. It might not totally make sense. And then when you look at the way that they are treated inside the Bloomberg ecosystem, it's just like, do whatever it is that you want to do. Right. Write your newsletter as many times a week as you want to do. If it's a crisis, put out seven podcast episodes in eight days or just two a week. Like, you know, it doesn't matter what days, it doesn't matter what. And it's very much like we have these people who have genuine interests who really care about this stuff, who are, like, lit on fire in the. In these moments, and we're going to empower them to just do their thing. And I think that that's what's really hard. Inside establishment institutions, there's just natural bureaucracy. Right?
- [51:53]
Peter Kafka
Natural bureaucracy. And even beyond the money part, because let's assume that Bloomberg is giving Joe and Tracy a gazillion dollars, which they deserve. Inevitably, if you have Joe and Tracy off in one part of the organization doing basically whatever they want, you then have a bunch of other people who don't have that freedom. And they're like, well, why can't I do that? And how come so and so. And that is a tension again, I Used to work with Walt Mossberg and Kara Swisher. And that was a thing at the Journal. And then one of the reasons. And even when they split off from the Journal, there was still a lot of animosity within the greater Dow Jones universe about why those people got those special privileges. So there's a. I can imagine lots of executives not wanting to deal with that type of.
- [52:41]
Charlie Warzel
Yeah, no, I think that that's true. But also I think the way that if I were to advise media executives on this, I would say that there are so many different types of people who do broadly what we call journalistic work. Right. There are people who are so great in the moment as contextualizing and analyzing things on the fly, writing the day two story or podcasting the day two story. On day one, they just sort of see the matrix. They are great at coming up with the frameworks for that type of thing. Those people usually often can't translate to. I publish one investigative story every 18 months and I get somebody freed from prison or someone sent to prison or whatever happens because I am so good at working the phones and I see the long thing and I. I can do that work. And that has, you know, you wouldn't tell that person, I need you to blog every day, right?
- [53:38]
Peter Kafka
No, you're a marathon runner. Run a marathon. We have a sprinter over here to sprint.
- [53:42]
Charlie Warzel
Exactly. I think there's a reticence among, like media executives to see the type of person that is this multi platform creator type individual who plays by their own rules a little bit more, lets it rip, but also in a way that's not like discrediting. I'm not talking about like a conspiracy substacker or somebody who works within the. The concepts that we would call respectable journalism. That person is totally different than those other types. They may be a little closer to the day two story on day one person, but they come to the Internet with a kind of creative, extremely generative lot of that bats vibe. They're very experimental. I know that they're established folks, but I think like Joe and Tracy and the odd lots, people embody that kind of thing. Right. And you have to empower people like that who are just like, I know how to find that perfect blend between what people want right now and what is good and my own sensibility and
- [54:48]
Peter Kafka
what I'm interested in. And I'm also comfortable being wrong. This thing may look bad in two weeks because our presumptions may have changed.
- [54:55]
Charlie Warzel
And I think you have to look at them as like a different desk if you want to work with them and you have to say like, ye, unfortunately there are going to be people at the organization who are like, why do they have all the freedom and I don't? But there's also, where I work, there are people who are excellent long form magazine writers who really do only write a couple times a year. And to me that's totally understandable because they do a thing that I don't do in the same way that I'm not trained at doing or don't have a lot of experience doing. And I think it's totally normal for there to be these different types of positions and things like that. I think what's just really hard, especially in this moment where you know, you've got lots of litigious people out in the world and it's tough, I think, for media executives to let people have that kind of freedom and control. But the whole point of this is that you build this audience by doing something that is really authentic feeling, that is exploring your curiosities in the way you want to do it, how you want to do it, when and if you try to take that person and then put them in this box, we have so much evidence from the past that it doesn't work that well.
- [56:10]
Peter Kafka
These are very wise words, Charlie. And if I was a better podcaster, I'd end there. But I do want to ask you one last question about Twitter. I think you no longer have an account there. As we know, when Musk bought Twitter in 2022, lots of folks said, this is the end of Twitter, it's going to blow up. It has stuck around, it remains a big platform, but it also is radically right. Lots of people like you have left entirely. People like me spend less time there. And I can't think of another online institution that has gone through a change like that. We're used to things, you know, cresting and then falling away. You know, the MySpace, Friendster, whatever. I don't remember anything transforming like this. Can you think of another example and does it tell us anything about either the specifics of Twitter or the Internet that remains a big thing this many years into the Musk ownership?
- [57:05]
Charlie Warzel
I think that Twitter is one of one. I really do. I think it just, it captured something. It captured a group of people. And I think frankly, one of the reasons that it still exists in the way that it does is because there's a number of people who didn't want to give it up, who felt, felt totally like, yeah, I'm not leaving. I built the audience. But it's also just. It's part of my daily workflow. It's part of what I do. It's a thing that I check whether it's an addiction, a compulsion, or just they think it's a good place for them to be.
- [57:44]
Peter Kafka
And we should note that when you and I are talking about the demise of Twitter, we're talking about a version of Twitter that we like, but for other people. They never participated in that Twitter. They were into sports Twitter or talking about tech Twitter. And now AI Twitter is a bear big thing.
- [57:59]
Charlie Warzel
And so, I mean, this is actually very interesting because I think it was like, last week, honestly, I reposted a couple of things. Like, I have spent a long time not tweeting on the platform. Not.
- [58:13]
Peter Kafka
You were dormant but not deactivated.
- [58:15]
Charlie Warzel
Yeah, I mean, I have to use it in some sense to, like, monitor that conversation for, you know, my work, but didn't post anything. I've had a number of conversations with people, people in the AI tech world, especially people who are a little bit younger, who are like, I have no idea who you are. And, like, they will go back through my, like, when I get in touch with them for a story, like a source, and they're like, wow, like, there's like, this stuff's interesting. This is really cool. You have no visibility in my world. Like, I don't read mainstream news stuff unless I see it on. On Twitter, X, whatever. And so, like, that's cool. But this is the first I'm hearing of you guy who's been doing this for 17 years. And that's totally fine. I'm not offended by that or anything. But I've had enough of those interactions with those people who are so heavily in the. Especially, like, the AI discourse and stuff on X that I'm like, all right, I think I probably need to just, like, have some kind of, like, post the Atlantic version of my story presence or maybe, you know, there's a lot of conversations that happen in the, you know, the AI discourse that I'm like, well, I kind of have a thought about that, and maybe just like, injecting myself in there is going to be at least helpful. So, like, people. We're in the business of getting people to read or think about our stuff.
- [59:43]
Peter Kafka
What I've been telling myself why I continue to use Twitter is there's an audience there that I want to reach and be in front of of. And I weigh that against my distaste for the person who owns it and a lot of the behavior that's there. And I'm like, I don't love that my activity is somehow supporting this, and I do it anyway.
- [60:02]
Charlie Warzel
Right. And. And I. And so, like, I. I think I'm in that tortured camp. But I think in terms of, like, why has Twitter not died despite becoming a completely different place? I. I think that a one reason for it is this. The way that the AI conversation happens on X, like, it's not the only reason, but it is in the same way that politics from, like 2009 to, like 2022 were just, like, so powered. Every single piece of, like, the news cycle happened first on Twitter, it was powered by that. Or like, we're just responding to tweets left and right. It' like this conversation is happening and it's not. It's not a mass of humanity. But in that same way that Twitter was for politics, scale matters less than who is there and how people identify that as being part. An important part of the culture. And so I think that as the AI industry becomes so important to everyone's life, because it's just, you know, it's changing, it's rewiring parts of our society, our entire economy is dependent on it. All this stuff, you sort of have to pay a little bit of attention to that conversation. Whether you like it or not, pay
- [61:25]
Peter Kafka
attention, but be in the conversation because you could be lurking through that whole thing if you wanted to, but also
- [61:31]
Charlie Warzel
because it is so, like, it's a great match of form and function. Right? Because the AI development process is happening at Twitter speeds, right? Like every couple of days, someone comes out with a new foundational model or update to the model. Like, one day, everyone's talking about Deep Seq. Then everyone's talking about cloud code. Now everyone's talking about Codex. It's like it's moving so quickly and frenetically that X is a really helpful place for that. But also what X does is it accelerates that conversation. And what I have noticed is if you are not monitoring it constantly, you get lost immediately. And so it. Again, I don't know that any of this is good or right or I don't think it's the way that it should be. Yeah, but it. It kind of rewards this, like, extremely online sicko behavior because it's the only way that you can participate and not, you know, show your ass.
- [62:28]
Peter Kafka
I remember having you and Oliver Darcy in studio to talk about Twitter and politics years ago. And I said, oh, I just came back from vacation and I turned Twitter off for the week or whatever it was, and you gu Looked at me like I was an alien. Like, how could you possibly be doing your job even on vacation? How could you even, like, log off for a week? You'd miss everything.
- [62:50]
Charlie Warzel
And I don't think that that's necessarily good. Right. Like, we saw what that did to politics. I think I've seen, like, I joke a lot. You know, the phrase AI psychosis is about, like, regular people having problematic relationships with chatbots thinking they're real or whatever. I think that there's, like, another brand of AI psychos that's happening on X with all the people in this conversation. Not that they believe that the technology is useful when it's not. I think that conversation should probably be put to bed. Clearly, AI is useful for some people. It is transformative for some people. It is transforming lots of culture and the economy at the moment. But I think that the way in which the conversation happens, you have people talking about, know, like, having their open claw remind them to drink water and, you know, tapping into the cameras on their. In their home to say, hey, you haven't been to the kitchen lately. Like, there's people who are, like, building this, like, sci fi, dystopian weirdness future for themselves, and there's other people who are, like, hunting and pecking onto, you know, chatgpt about what's the best smoke detector. Right. Like, there's this huge gulf. And I think that Silicon Valley and the AI conversation on X is accelerating such that it's isolating those people a little bit from all the other people who are interacting with their products as normal folks. And I think it's ultimately, as we saw from politics, it's not good. It's not good to be so high on your own supply.
- [64:28]
Peter Kafka
Yeah. This seems like an argument for everyone to get off Twitter, but they're not going to.
- [64:31]
Charlie Warzel
And so I'll continue to have my tortured relationship with the platform.
- [64:35]
Peter Kafka
All right. I guess you can find Charlie Wurzel on Twitter.
- [64:38]
Charlie Warzel
Oh, God.
- [64:40]
Peter Kafka
But the best way to consume him is to go to the Atlantic, go to find his. Find his YouTube videos. He's excellent. Thank you for coming.
- [64:47]
Charlie Warzel
Thanks for having me.
- [64:49]
Peter Kafka
Thanks again to Charlie Warzel. Love having him on the show. Thanks to Charlotte Silver, who produces and edits this show. Thanks to our sponsors who bring it to you for free. Free. Thanks to you guys for listening. See you soon.