A.I. Beach + Vibes-Based R.T.O. + the ‘Black Mirror’ Quamputer

This week, advertisers swarmed the beaches of southern France for the Cannes Lions advertising festival. Kevin says artificial intelligence is all anyone there can talk about, but admits the conference is making him rethink how quickly generative A. I. will take over the industry — despite the buzz. Then, the New York Times reporter Emma Goldberg on when remote work stopped being the future for tech companies. And finally: What does the newest season of “Black Mirror” tell us about what’s next for TV?

Speaker names marked ? are inferred from speaking time and the show’s known hosts — they are not voice-verified.

• 0:00Speaker 00

  Support for this podcast comes from Dell Technologies. It's time to put AI to work. Is your business ready? The challenge isn't getting started with AI. It's scaling beyond pilots while maintaining control of your data, governance, infrastructure, and costs. Dell AI Factory provides the integrated foundation you need to scale AI securely and efficiently with one trusted partner, Dell, AI on your terms. From idea to impact. Learn more at dell.com/yourwaytoAI. That's dell.com/yourwaytoAI.

• 0:34Casey Newton?

  Well, I was going to ask if you've learned how to say AI in French.

• 0:37Kevin Roose?

  I think it's actually IA. I think it's intelligence artificial.

• 0:43Kevin Roose?

  How's my French accent?

• 0:43Casey Newton?

  Here's the nice thing. If you ever are like writing about AI and you accidentally transpose the letters and someone calls you on it, just say that you were using the French variation.

• 0:53Kevin Roose?

  All sounds so fancy.

• 0:59Kevin Roose?

  I'm Kevin Rusak, columnist for New York Times.

• 1:01Casey Newton?

  I'm Casey Newton from Platformer.

• 1:02Kevin Roose?

  And you're listening to Hard Fork.

• 1:04Casey Newton?

  This week, Kevin goes to France to find out why AI has taken over the world's biggest ad conference. Then, New York Times reporter Emma Goldberg joins us to talk about why tech CEOs are giving up on their remote work dreams. And finally, Black Mirror and the Future of AI in TV.

• 1:35Casey Newton?

  Bonjour, Kevin.

• 1:36Kevin Roose?

  Casey, I am so far away from you right now. I'm in France. I'm in the south of France at the Cannes Lions Advertising Festival.

• 1:42Casey Newton?

  Now, Kevin, you are a technology reporter. What has possibly lured you over to France for an advertising festival?

• 1:51Kevin Roose?

  Well, some fine folks at the New York Times said, you know, we're sending a delegation to this annual advertising bonanza in Cannes, and we would like you to come, talk to people about AI and other topics, including stuff that we talk about in the show. And I've learned a lot of things in my career in journalism. And one thing I've learned is that when someone offers you a trip to the south of France, you don't ask too many questions, you just go.

• 2:21Casey Newton?

  You just get on the plane.

• 2:22Kevin Roose?

  Yes, you get on the plane. So I'm here in Cannes, I've been here for a couple of days now. And it is really wild.

• 2:29Casey Newton?

  Yeah.

• 2:30Kevin Roose?

  Imagine like the best beach you've ever seen or been to.

• 2:35Casey Newton?

  Okay.

• 2:35Kevin Roose?

  And then like, put a business conference down there. So if you just look out at the water in Cannes, you see what's essentially happened is that these like, beaches have been taken over by giant technology companies. There's like Meta Beach, and then there's Google Beach. Like, everyone's got a beach. These companies spend millions of dollars, they fly all these people to Cannes. And it's really amazing. I mean, these are companies, most of them have had, you know, pretty big budget cuts this year. They're laying off people. And I kind of wondered before I got here, like, is it just gonna be like, you know, two tents on the beach with like, someone handing out, you know, mojitos? And no, it's just as big as ever.

• 3:19Kevin Roose?

  Everyone is here, they're spending tons and tons of money. There have been some very funny, like, attempts at cost cutting.

• 3:27Casey Newton?

  Like, like they're handing out drink tickets instead of just open bars?

• 3:32Kevin Roose?

  Well, like, one of the corporate beaches that I will not name, like, the only other year I've been in 2019, there was like a huge buffet, like a huge spread. They had like, you know, people making custom gelato on the beach. And this year, it was just like handing out popsicles. So that is like their version of austerity at Cannes.

• 3:51Casey Newton?

  You hate to see it, it sounds terrible. Are people managing to keep their spirits up despite these nuance darity measures?

• 3:59Kevin Roose?

  It's interesting, like, obviously, for a lot of people here, this is just like a boondoggle, right? This is just like the thing that like is the kind of justifiable business expense that like gets you to the French Riviera in the summer. But people really do deals here. It has a high enough concentration of executives from not only the big tech companies, but all the big advertisers, you know, all these companies that spend millions of dollars or billions of dollars a year on advertising are here. And so it is actually like quite a stressful week for people who are in the ad business.

• 4:35Casey Newton?

  Yeah, well, you know, here's where I think that this tale of the festival actually comes into our orbit on Hard Fork, which is that over the past year, many of the companies we talk about here, which are reliant on digital advertising, have been having a lot of trouble, right? The advertising market is down and they all need a new story to sell. And man, AI came along at just the right moment for them to have something new to sell. And so what I want to know from you is what is the story that's being told down there? And what is the conversation about what AI is going to do to all of these folks?

• 5:11Kevin Roose?

  Yeah, well, it is like essentially an AI conference this year. I mean, that is sort of every other panel is about AI and advertising. I mean, it's sort of become a running joke among the people here in Cannes that like, if they have to hear one more speech about AI, they're going to lose their minds. Because it is just everything that everyone is excited about and fearful of.

• 5:32Casey Newton?

  Wait, wait, I just, I love the idea that people are like, if I have to do one more bit of work at this work festival, I'm going to lose my mind. Give me my damn popsicle.

• 5:43Kevin Roose?

  Exactly. So one surprising thing to me is like, you know, everyone's talking about AI, but nobody here seems to know exactly what to do with AI. Like, all these companies, they're talking up their AI strategies and their products and how they're using generative AI in experiments for maybe their creative team. But if you really start drilling down into like what they're actually doing, it's like not cutting edge generative. AI stuff that we talk about in this podcast. Like, most of it is like, stuff that frankly existed a few years ago. It's like, we built this tool to like, optimize some part of the ad buying process.

• 6:21Kevin Roose?

  Or, you know, we're doing, you know, AI classifiers to do sentiment analysis for some other piece of their advertising business. They're talking about these kind of traditional AI techniques that existed five or 10 years ago. Not the kind of like, large language model powered generative AI stuff that we usually talk about on this show.

• 6:40Casey Newton?

  Yeah, I think that's really important to say because we're using AI as a catchall for a bunch of technologies. But it sounds like what you're saying is that it can, we're really talking about the kind of AI that is just doing automation, right? It's just figuring out how to take tasks that were once done by human beings and do them much more quickly via computers.

• 7:00Kevin Roose?

  Totally. And, you know, I did hear a few examples of things that probably weren't possible to do a few years ago. Meta was talking at one point about how soon, if you're an advertiser, you're going to be able to use like, an image generating AI to like, change your company's ads for the holidays without reshooting the photos. Like, you could just say, like, make this look like the holidays or something. Or, you know, a podcast with a celebrity who's big in India and you want to use generative AI to like, make them speak Hindi or something, that kind of thing.

• 7:34Casey Newton?

  I got so excited about that stuff, Kevin, because I just think about the brand disasters that are going to come out of this. You know, just imagine that the junior brand safety person forgets to check the new creative and all of a sudden, you have the most offensive Christmas advertisement that anyone's ever seen, all thanks to the AI. So, that's coming our way and that's just something to look forward to as a consumer.

• 7:54Kevin Roose?

  It's true. So, that's the kind of thing that it seems like a lot of the big ad agencies and advertisers are doing. This kind of like, small around the edges thing. There are some tools that seem like they could potentially be more useful than that. I went to a talk today where Google was sort of showing off some of its generative AI stuff. And they had these experiments where you could take like, the Android logo and, you know, color scheme and basically have one of their generative AI tools like, create a bunch of variations on that logo. So say you want to make like, a dog that looks like the Android logo. You can do that. I don't know why you need to do that.

• 8:34Casey Newton?

  The future is now. Everything is happening.

• 8:38Kevin Roose?

  So everyone's very excited about AI. And then when you start talking about what they're doing with it, actually, it's kind of underwhelming, frankly. And, you know, I gotta say, like, this was a surprise to me. And I guess this maybe means that I'm in the San Francisco tech bubble. Because like, we've been talking on this show about how generative AI is gonna like, totally upend every industry and already has sort of started really affecting the way that like, students and teachers are interacting, or how doctors are communicating with patients, or, you know, any of the other ways that we've talked about generative AI being used. And I kind of expected to show up at this giant advertising conference and like, meet people who are using these tools in some crazy new way, like some new upstart ad

• 9:21Kevin Roose?

  agency that has just replaced all its copywriters with ChatGPT and Midjourney. And it's just like, doing 10 times as many ads as the big guys with many fewer people. But that's not happening. In fact, a lot of the people here sound kind of unimpressed and a little bit over it, which is very surprising to me.

• 9:40Casey Newton?

  Well, but you're getting to my real question here, which is like, is this the last Cannes Lions Ad Festival, right? Because in the ad business, you really only have to do two things. You gotta create the ads and then you have to sell them. I mean, I guess you have to measure their performance, too. But increasingly, these are tasks that computers can do. So are you hearing any amount of anxiety among people wandering around the beach with their popsicles about whether their job is still gonna be needed in five years?

• 10:06Kevin Roose?

  You know, I sort of expected to hear more of that. And there was, like, people say it as a joke, oh, you know, we'll all be replaced five years from now. But I think they don't really feel that worried about it. And it's got me thinking about, it's got me thinking about elevators. Do you know the story of push-button elevators?

• 10:27Casey Newton?

  I don't.

• 10:28Kevin Roose?

  So elevators, you know, the first elevators were operated by humans. There was like, literally a person called an elevator operator who sat in there and made the elevator go up and down. And then in the 1880s, this thing called the push-button elevator.

• 10:41Casey Newton?

  Wait, I have a question. I have a question.

• 10:43Kevin Roose?

  Yes.

• 10:43Casey Newton?

  What, what, the human operator in the pre-push-button elevator?

• 10:48Casey Newton?

  Were they not pushing a button? Were they like pulling a rope on a pulley? Or what were they doing?

• 10:53Kevin Roose?

  At first, they were pulling ropes on a pulley. And then it was like a lever system.

• 10:57Casey Newton?

  Okay.

• 10:58Kevin Roose?

  But yeah, it was a mechanical process and you had to have a trained elevator operator to like, sit there and move the elevator up and down. And then in the 1880s, someone invented the push button elevator where you as the person riding the elevator could actually work it yourself. And everyone at the time was freaked out because they're like, all these elevator operators are obsolete now. We don't need them anymore. They're all going to lose their jobs. What are they going to do? It was sort of like an automation panic. And then what happened was nothing. People didn't use the push button elevators. So push button elevators basically didn't become popular for quite a while after they were invented.

• 11:35Casey Newton?

  Wow.

• 11:36Kevin Roose?

  And one of the reasons was that the initial sort of push button elevators, they weren't all that reliable. They didn't have like the kind of like sophisticated like routing software that, you know, allowed them to like service huge skyscrapers. But people just also didn't trust them. They didn't think they were safe and so they didn't want to get in one. You really wanted like a human there to operate your elevator, even though they weren't technically necessary anymore. And so even though this technology existed to automate this thing, moving an elevator, it really didn't happen for a long time after it was technically possible. And, you know, what does this have to do with advertising and AI?

• 12:17Kevin Roose?

  Well, I think that AI is capable of doing a lot of work that is currently done by humans in the advertising industry, as with lots of other industries. But I think it's a very different question of when people will actually feel comfortable using these tools, when they will, when it will actually switch over. And I think after being in Cannes and talking to all these ad executives about how they're afraid of AI, like I think it could actually be a little while before the ad agency goes fully.

• 12:45Casey Newton?

  That makes sense. It does remind me of something I don't like though, which is when they show ads in elevators now. You've noticed this, right? It's like, I'm going up three floors. Can I have a moment's peace? I'm sure that idea was probably hatched in Cannes.

• 12:57Kevin Roose?

  Well, aren't you looking at your phone anyway?

• 12:59Casey Newton?

  Yes, I mean, but only be I would happily stare at the doors, but um, now I have to look away because they're trying to get me to buy something.

• 13:07Kevin Roose?

  Well, well, if any elevator companies would like to show clips of Hard Fork in elevators to a captive audience, uh, we're interested and our sponsorship team will be in touch.

• 13:16Casey Newton?

  Remember when Cheddar signed a major deal to get all of their content in gas stations across America?

• 13:20Kevin Roose?

  On gas pumps. I love.

• 13:23Casey Newton?

  I just know that deal was cut at Cannes, too.

• 13:27Kevin Roose?

  I did get a pitch this week at Cannes for a company that wants to put uh, smart advertising inside driverless cars so that like you could be in a driverless car and it would just say like, are you sure you don't want a Whopper right now? You sure you there's a Burger King? You sure you don't want us to veer off and go to Burger King?

• 13:43Casey Newton?

  Yeah.

• 13:44Kevin Roose?

  Like that kind of thing. So that is coming to a taxi near you.

• 13:46Casey Newton?

  If you want to be able to leave the car, you're gonna need to write down this coupon code. Okay? And then we then and only then will we release the locks.

• 13:56Kevin Roose?

  So I think being here and hearing all of this chatter and buzz about AI and then comparing it to like the actual use of AI in a lot of these ad agencies business is like really making me rethink my own views about how quickly AI is going to transform not just advertising, but lots of other industries. I mean, you know, I spent a lot of time in the San Francisco sort of tech AI bubble. And you know, a lot of people in San Francisco will tell you that like every field, law, medicine, finance, journalism, you know, advertising, like they're all going to be disrupted by AI. And I think that is true, but what it ignores, I think, is that it often takes a while for technology to like

• 14:40Kevin Roose?

  diffuse itself throughout an industry, right? You you have to give people time to kind of adjust their workflows and adopt new processes and like actually figure out how to use the damn stuff. So like, I actually think that like ChatGPT came out seven months ago. Mid Journey, Dolly two, those came out a few months before that. Like these tools have been in the world. And the fact that they have not.

• 15:01Kevin Roose?

  Not already upended the advertising business makes me think that maybe this is just going to be a slower transition than I thought.

• 15:07Casey Newton?

  Yeah, so that sounds right to me, although I think partially for a reason that you haven't mentioned. I go back to when GPT four was released and Sam Altman said something to the effect of it gets much less impressive the more you use it. And that actually has been my experience with some of these generative AI tools over the past couple of months where at first you're like, wow, I can't believe this thing is spitting out so much text at me and it's so coherent. And then after you do it for a while, you realize that it really is just sort of plagiarizing blog posts on the internet and rearranging the words. It just doesn't seem as impressive as it once did. And so I think if you're an advertiser and you see this dream of, well, eventually I'm just going to be able to write like, give me a beautiful ad for Coca-Cola for Christmas time and the thing does it for you.

• 15:52Casey Newton?

  You can understand why that would be very interesting and why you'd spend all your time on the beach at Cannes talking about it. But practically, we're just a long way away from there. So I think that just by looking at the state of the art and seeing how much it is lacking, you can adjust your timelines into the future in terms of when all those jobs are going to get automated away.

• 16:12Kevin Roose?

  Right. And I think there's another sort of important piece of this too, which is that humans have to decide to use this stuff, right? They have to be able to trust the technology, that it's reliable enough, that it's not going to cause some gigantic error that costs them their job and embarrasses them in front of their clients. Like you, you, you just have to like factor in the fact that these, there are these humans who have been doing this job, who are not going to give that up easily. And who have a lot of pride in their work and who want to come to Cannes every year and celebrate all the great ads they made over the past year. And like the culture of an industry matters a lot more than I think I gave it credit for.

• 16:53Kevin Roose?

  So I really think like being here has made me think that it might be a while before we start to see these tools really show up in some industries.

• 17:02Casey Newton?

  You know, that's exactly what I would say if I wanted an annual beach vacation in the south of France. So well done, Kevin. Expense approved.

• 17:13Kevin Roose?

  Well, someone's got to keep an eye on the robots and I volunteer frankly.

• 17:16Casey Newton?

  Thank you for your service.

• 17:22Casey Newton?

  When we come back, New York Times reporter Emma Goldberg joins us to tell us why tech has changed its tune about remote work.

• 17:45Speaker 00

  Support for this podcast comes from Dell Technologies. It's time to put AI to work. Is your business ready? The challenge isn't getting started with AI. It's scaling beyond pilots while maintaining control of your data, governance, infrastructure, and costs. Dell AI Factory provides the integrated foundation you need to scale AI securely and efficiently with one trusted partner, Dell, AI on your terms. From idea to impact. Learn more at dell.com/yourwaytoAI. That's dell.com/yourwaytoAI.

• 18:19Speaker 03

  AI spend is growing fast, but most businesses still can't see exactly where that money is going. Ramp gives businesses visibility and control across AI providers with real-time guardrails by team and model, helping ensure spend goes to the highest ROI areas. Stop guessing where your AI spend is going. Get started with Ramp. And if you're new to Ramp, you can get $250 off at ramp.com/hardfork. Go to ramp.com/hardfork. Ramp.com/hardfork.

• 18:51Speaker 04

  Ripple has been powering modern finance for a decade. It's not our first trip around the blockchain, and it won't be our last. We're leading the charge to bridge the gap between existing financial systems and the future of finance. You can have it all: legacy, stability, and digital innovation. Systems that work with your other systems. Because with the right partner, the question isn't crypto or traditional finance. The question is simply, why choose? A new standard of modern finance. It's happening with Ripple. For informational purposes only. Availability of products can vary and may be provided by different Ripple entities.

• 19:26Kevin Roose?

  So Casey, I am working remotely this week from a little apartment in France. But next week I will return to the office. And you know who else is returning to the office, Casey?

• 19:38Casey Newton?

  Who's that?

• 19:39Kevin Roose?

  Workers at large American tech companies.

• 19:43Casey Newton?

  That's right, because it's not just people returning from their beach boondoggles. It seems like across the nation, managers are demanding that the people they once told, "Hey, stay at home, just hop on Zoom and Slack," those same folks are now being told, "Hey, you gotta come in."

• 19:57Kevin Roose?

  Absolutely, and this is something that has really surprised me because I feel like for the last couple of years, we've been hearing about this kind of return to office battle where, you know, workers, some of them want to go back to the office, many of them don't, but bosses are trying to get them back into the office. And it just feels like a topic that we should really spend some time talking about, not only because it is sort of a story that impacts a lot of workers in tech. These are a lot of the companies making these policies and changing these policies are tech companies, but because it really gets to this question of, well, how much does technology enable remote work and what might be lost in that process? So this week I wanted to bring in my colleague, Emma Goldberg. She covers sort of the future of work for The New York Times.

• 20:41Kevin Roose?

  And she just wrote a great article called "Return to Office Enters the Desperation Phase", which is all about the many strategies that company executives are using to try to entice workers back to the office. A very interesting and funny story. And I really wanted to bring her in, not just to talk about this sort of overall work from home, remote work dilemma, but really how the tech industry has changed its tune, because it has been quite dramatic.

• 21:08Casey Newton?

  It has, and my hope is at the end of this conversation, your bosses, Kevin, will have some really good strategies for how to bring you back from France and get you to do your job again.

• 21:18Kevin Roose?

  I'm working. Look, it's not easy over here.

• 21:21Casey Newton?

  Yeah, someone told me your croissant was stale this morning, so I'm sorry to hear that.

• 21:28Kevin Roose?

  You know, we make sacrifices for our jobs. All right, Emma Goldberg, welcome to Hard Fork.

• 21:34Speaker 05

  Hi, Kevin. Hi, Casey.

• 21:36Casey Newton?

  Well, so Emma, I'm really excited you're here today because for a while now, I've had this question on my mind, which is, when did remote work stop being the future? You know, you rewind back to 2020 and I was on the phone with CEOs at Meta and other places, and they were telling me, look, this pandemic is a portal that we're stepping through. And on the other side, we're just not going to go to the offices as much anymore. And we need to essentially invert the systems that we had before. Instead of having offsites, we're going to have onsites, right? Because most of the time, people are going to be far away. And then little by little, over the past couple of years, we've seen announcement after announcement from these companies saying, you know what? We actually do want you back here two days a week. And then six months later, it's like, you know what?

• 22:21Casey Newton?

  We think we actually need you back here three days a week. And we just kind of see that thing continue to trickle on. So you've been doing some reporting on this subject lately, and we wanted to ask you what you are learning. So talk to us a little bit about where you dived in and what you found.

• 22:38Speaker 05

  Yeah, I mean, I think in watching the kind of return to office saga play out, to me, it's almost felt like that game of chicken. I don't know if you guys ever played that as a kid.

• 22:50Casey Newton?

  Yeah.

• 22:50Speaker 05

  But it's like, I feel like it's been a little bit of CEOs saying like, okay, you can come back to the office now. And workers kind of looking at them and saying like, are you going to make me? And there's this weird standoff. And you're seeing companies take so many different, you know, tacks to that. For some people, it's been this like perk game where they've been like, well, will you come back for a Lizzo concert? Or, you know, like being serenaded by a jazz pianist in your lobby? Or pizza, like who wants pizza? And then you're seeing other companies take a more strong arm tactic. Like, Google has now started to communicate that whether you go to the office could be taken into account in your performance review. So I think we're seeing a little bit of like a good cop, bad cop routine here

• 23:34Speaker 05

  with companies kind of taking a gamble to see what approach works out best.

• 23:38Kevin Roose?

  Yeah. Yeah, you recently spent some time reporting in the Bay Area where we live. And what I found remarkable about tech companies sort of return to office about phase is that these are some of the companies who make the tools that made it possible for people to work from home in the first place, right? Companies like Google, companies like Slack, which is now owned by Salesforce. These are the companies that sort of allowed for the work from home boom during the pandemic. And these are the companies that are presumably very well set up themselves to work remotely. So what happened? What shifted in the minds of the executives at these big tech companies that made them say, we actually need workers back in the office?

• 24:21Speaker 05

  The Salesforce journey is one of the most fascinating to me because I felt like they were really gung ho about remote work earlier in the pandemic. And they put out this memo in 2021 telling a lot of their workers that they would be able to be either permanently remote or at least permanently hybrid. And they had this line in it where they said, the nine to five workday is dead. And it felt like this just a big old embrace of like, sure, you can just like chat with your coworkers via emoji all day from wherever you want to be. And now you're starting to see that get walked back a little bit in various ways, whether it was Marc Benioff, the CEO of Salesforce, making comments essentially implying that he felt like some of the lower level employees at Salesforce, especially those who had joined

• 25:06Speaker 05

  during the pandemic, were just not being as productive as those who were in the office. So there's, there's all sorts of kind of subtle messaging starting to occur suggesting that maybe remote work isn't quite as effective as people would have hoped it would be. And certainly I saw that from Salesforce recently with the announcement that they're going to donate $10 per employee per day that they come into the office over a 10 day period to a charity of their choice. And when I asked them about that, they they kind of said, oh, you know, well, we're always looking for moments for doing well and doing good. And it's like, sure, but I could think of a lot of ways to do well and do good that don't involve, you know, strongly messaging to people that they should be in the office.

• 25:49Speaker 05

  And when I walked around the Salesforce building in New York recently, it didn't really feel like the nine to five workday was dead. Like, it was just a Monday afternoon and there were people all around at their desks. They have these like little animal mascots all around the building that feel a little bit like rah rah, the office is fun, you know, like Cody the bear is near the coders. So I felt like it's very much a 2019 version of the office. And of course, their answer to that would be they haven't abandoned remote work and still I think it's 27% of their employees are remote. But it does feel like there's a strong encouragement for people to get back to the office. I think to your larger question, Casey, it feels a little bit like remote work was this experiment and the results are starting to roll in.

• 26:34Casey Newton?

  Right. What are some of the theories, Emma, that that you're hearing among CEOs about why remote work didn't work out as well as they thought it would? What did they sort of assume that turned out not to be the case?

• 26:47Speaker 05

  Yeah, I think that there's a couple of ways that someone who runs a business might answer that. Realistically, I think we have to look at real estate as a factor. Because when I think about what companies have gone permanently remote, Yelp was a major example, and their lease just happened to be coming up during the pandemic. So they were able to cut their lease. And now they're they're leasing space from Salesforce in one of their buildings. But that's a big factor because oftentimes a company signs a lease that's like eight to 10 years long. So if you happen to have a lease that comes up during the pandemic and then you go permanently remote, that feels like a signal. But I think if you if you ask people on a more philosophical level, they might kind of wax about, you know, the mentorship that the office allows for, the relationships, the culture, maybe the productivity.

• 27:32Speaker 05

  I think it feels a little bit squishy to employees, but to CEOs, some of that does feel really important.

• 27:37Casey Newton?

  Yeah, I mean, the thing that always seemed persuasive to me among skeptics of remote work was the idea that if you were a junior employee and you were entering a massive organization and your only real interface with your colleagues was going to be on Zoom, you probably were going to have a hard time rising up the ladder at your office, right? You're just not going to have those same kind of shared in-person moments that seem essential to building a career. So that part of me kind of understands why CEOs aren't seeing their younger employees integrating as well as they hoped.

• 28:10Speaker 05

  I think that that is a really compelling case. I think the problem is that it's hard, like, we can't pretend like the old model of working worked so perfectly for everyone either. Like, that was just always this kind of devil's bargain that parents, for example, had to make where they always kind of felt like they had to be in two places at once: at their desk, in the office at 4:00 PM in the afternoon, and also maybe at school pickup, or maybe trying to get some time with their kid before their kid goes to sleep. And now, you know, there was this really interesting study from Microsoft that showed people have just completely remade the shape of their day. So instead of like two peaks of productivity before and after lunch, now there's three peaks, where you're working in the morning, you're working in the afternoon, then maybe you're taking a break, and there's

• 28:54Speaker 05

  a third peak of productivity at night. So presumably, for a lot of people, after their kids go to sleep. So I think it's true that maybe remote work isn't working so well for everyone. Also, maybe in-person work was really hard for a lot of people.

• 29:07Kevin Roose?

  I actually have 10 peaks of productivity in my day, but that's just because I'm on the grindset, baby.

• 29:12Casey Newton?

  You're on a beach in France.

• 29:16Speaker 05

  Yes, he can.

• 29:17Kevin Roose?

  Touche, touche. So, Emma, I want to talk about these sort of factors in the remote work equation. So one of them is productivity. And I think a lot of people who are invested on one side of this debate or the other will say that, you know, actually, remote work is Good or bad for productivity? And on the good side, there's a study that came out by a professor at Stanford called Nicholas Bloom, who studied remote and hybrid work for years before the pandemic. And one of his big takeaways was that workers who worked remotely tended to be more productive, in part because they're not commuting, but also because of the things you mentioned, like sort of adding additional productivity to your day.

• 30:00Kevin Roose?

  And that seemed like a compelling case if you were a CEO during the pandemic to say, well, let's just let our people work remotely because they'll be more productive. So if there's evidence that workers are more productive from home, why are some bosses changing their tune? Why are they now saying, we want you back in the office?

• 30:19Speaker 05

  I think the problem is there's kind of evidence on all sides where it comes to remote work. It's not really a satisfying answer, but I don't think we actually know yet where quote unquote productivity is highest, or like what kind of work is really getting done. For example, if you look at some of these productivity studies about remote work, some of that has to do with like call center employees. So maybe it's about like literal time you're putting in. And when you're working at home, you're saving time on the commute. On the other hand, there's a big study that showed that when you're working at home, you might get less feedback on your work. So they followed engineers at a company and they found that when you're working at home, people got less feedback on their code. So they literally just got like less comments on lines of code written.

• 31:01Speaker 05

  So maybe you might be churning out more work, but not getting as much feedback on it. So that work might be, you know, lower quality. There was another study that showed that people actually form fewer weak ties when they're working at home. So that means like, maybe you're not expanding your network as much. Maybe you're not having as many impromptu mythic water cooler conversations. So I think it's a little bit like jury's still out. I think it's possible that maybe like people just churn out more work when they're working at home. Maybe they cut out their commute time and, and they put that time back into, back into their job. But I also think that there are questions to be asked about the quality of that work.

• 31:39Kevin Roose?

  So Emma, you mentioned Salesforce and its $10 sort of charity bonus if you come into the office. What are some other perks you've heard about that companies, especially tech companies, are offering their workers to try to get them back into the office at least a couple of days a week?

• 31:56Speaker 05

  I actually think there's very little that I haven't heard about in terms of getting back to the office perks. I mean, Google did a Lizzo concert, which was pretty cool. I was like, can I go for reporting, which I did not. But a lot of companies did free meals. I went to a home buying company called Orchard recently that was doing a game night and happy hour. And it was really fun because they had this like bar cart that said it's 5pm somewhere on it with Jimmy Buffett, but they like started rolling it out like, like before 5pm. It was like solidly in the fours. So they were just in the office for the bar cart, which was fun. I have seen companies do like swag, you know, free t-shirts, mousepads, like pretty much

• 32:40Speaker 05

  everything you can think of, companies have tried.

• 32:43Casey Newton?

  So this stuff sounds undeniably fun. I would love it if, you know, work provided me with a Lizzo concert to kind of get my motivation up, you know. But at the same time, these things are just one offs, right? They don't have any lingering benefit. And if the reason that you didn't like coming into the office was that your commute is long, or you want to be home to pick up the kids, none of that really addresses that. So does this feel like the right solution to the issue?

• 33:09Speaker 05

  I mean, there's a weird reality in which sometimes covering return to office debates can feel like covering a sort of like fraught parent teenager relationship. And I like, I literally feel like I'm like studying parenting a little bit and figuring out like, our company, like, is it working when they just like throw candy at people? Or do they need to like kind of bring down the hammer? I think what works in terms of the perks is when they service this kind of like social anchor. Because the biggest problem, what people really hate is being called into the office so that they can sit all day long on their laptop and like zoom with colleagues who aren't in the office that day. So I think what works about the perks is that they get people all in on the same day at the same

• 33:50Speaker 05

  time. That's like one of Nick Blument at Stanford, the hybrid work researcher, that's his big thing is he thinks like, make the office worth it for people. Because there's like a meme going around of like, you know, CEOs saying get people back to the office. And then the office is like someone sitting in a privacy booth, like that's soundproofed so they can replicate working from home. So that doesn't work. I think if you can say like, get back to the office and there will be a Lizzo concert where by the way, all of your colleagues will be, then that works really well.

• 34:20Kevin Roose?

  Yeah, I mean, I don't want to sound like a total narc here or like I'm sucking up to the bosses who want to get everyone back to the office. But you know, Casey and I, we do tend to record the podcast together most weeks in person. And that's like an important thing that we've identified because there is something, at least for this kind of podcast, better about being in person. It just works better. And so I'm not totally dismissive of the research or just the theories out there that there is something special about being in person. I mean, Steve Jobs famously had an aversion to remote work at Apple. He believed that like, Apple employees did better work when they were sort of bumping

• 35:02Kevin Roose?

  into each other all the time at the cafeteria. He thought that the creativity comes from these sort of spontaneous collisions of people where you bump into something and you ask what they're doing and then you're sort of able to kind of work on something together. And I don't think that's totally ridiculous. I do find myself when I'm in the office around other people having better ideas and getting random inspiration from just collisions that I have with people in the office, not literal collisions, but these metaphorical collisions. So what does the research say about that, about the effects of remote work on creativity rather than productivity?

• 35:39Speaker 05

  Yeah, I think that that is one of the challenging things here is that, you know, we're still a little bit early into this remote work experiment for there to be a lot of like hard evidence around the kind of squishy concept of relationships. And I think that's one of the things that's like rankling workers a little bit. And I think Amazon is an interesting case study of that. Because Amazon recently asked corporate employees to start coming back to the office three days a week. And people have been mad. I wrote a story the other week about a big walkout that happened at the headquarters where hundreds of corporate employees walked out of the office for an hour at lunchtime. And one of the reasons they were doing that was over the mandatory RTO. And workers there that I've spoken with have said that one of the reasons they're so frustrated about

• 36:23Speaker 05

  being called back to the office is because they don't feel like they've been presented with evidence that being back in the office is that helpful for relationships and for culture. And a number of them have said to me, We would never be allowed to like build a product for Amazon if we didn't have evidence that it would work. Like, our company is so evidence-centric, and yet with return to office, we're just being asked to trust our bosses that things will just magically be better.

• 36:48Kevin Roose?

  Right. It's like vibe-based policy.

• 36:51Speaker 05

  Exactly. Exactly. They're like, "Just allow us to just vibe together and just, like, you know, believe." And they're like, "That is, if we just asked you to vibe with our product, I don't think that would go over so well." So I think that has been one of the challenges here. I think it's, you know, the cultural aspects of in-person work are something that we can feel more than we can prove right now. That being said, I have seen a study that said, you know, staring at your computer all day curbs creative thinking. And I think that is something that also jives intuitively, what a lot of us feel. But I think it's also going to be something that, like, the more this kind of wears on, the more research we're going to get.

• 37:33Casey Newton?

  But let's talk a bit about the technology piece of this. You know, so Kevin was bringing up earlier that a lot of the companies who were promoting remote work the hardest were also companies making technology that makes remote work possible. You know, I talked to Mark Zuckerberg amid the pandemic about how Meta was going to be the most forward-leaning remote work company of its size. And they also have recently sort of started to walk that back and move to a more hybrid situation. My curiosity there is, to what extent is that an artifact of the technology not being ready? That we don't have immersive enough video chat? The metaverse isn't really ready to hold a conference call? Is it really a

• 38:18Casey Newton?

  technology question? Or do you think in your conversations, this really just comes more down to questions of human behavior and psychology?

• 38:27Speaker 05

  I think the challenge is the way the technology kind of structurally limits conversation. So a lot of people use the phrase like the meeting after the meeting, which is like if you're in a meeting where some people are together in person and some people are dialing in, the minute those laptops close, like the people who are in person start just kind of chatting and like, maybe that's when the assignment gets made or the billion dollar idea, you know, pops into someone's head. I think there's just something a little limiting about staring at someone on a laptop because it makes you feel like, okay, let's just get right into whatever, you know, task needs to happen so that then we can get back to our real lives. I do think that there is probably going to end up being something a little bit irreplaceable about, you know, the substance of in-person interactions.

• 39:12Speaker 05

  Again, I would just say that that does need to be balanced with like, what's irreplaceable about what people get back in their lives when their lives are more flexible. There's also been, you know, really important studies highlighting the fact that workers of color have been some of the most vocal about preferring remote work because they feel like it allows them to escape certain microaggressions and kind of office politics and clickiness. So I think if we say that there's something kind of irreplaceable about being together in person, we also have to acknowledge that maybe there's something irreplaceable about what you gain from flexibility.

• 39:47Casey Newton?

  Yeah, it's interesting.

• 39:49Kevin Roose?

  Hmm. Can I try out a theory on you guys that I don't have high confidence in, but I do have somewhat moderate confidence in, which is that I think that AI is going to kill remote work. And the reason that I think that is because so much of what is happening right now in the corporate world, as these AI tools get better at doing the kind of mundane corporate, you know, rote work that people do, is that people are getting anxious about their own replaceability, right? And, and what can I do that can't be done by an AI? And so much of that, so much of these sort of uniquely human skills are things that are much, much easier to do in person, right?

• 40:32Kevin Roose?

  Like collaboration, creativity, leadership, these kind of soft skills.

• 40:37Kevin Roose?

  As they're sometimes called, that really are things that AI can't do nearly as well as humans right now. And so my theory, and I'd love to hear what each of you thinks about this, is that as the white collar workforce gets more and more automated, there's going to be more of a shift back to the office where people can actually prove to their colleagues that they are in fact a human and not just like three chat GPTs in a trench coat, because that is where their value comes from as a worker in an age where a lot of the rote value of their job has been siphoned off by AI. What do you think?

• 41:11Casey Newton?

  How about you go first?

• 41:12Speaker 05

  I totally buy that. In terms of vibes based RTO, you know, theorizing, that seems completely on point to me. First of all, I've been talking with a lot of companies that are making these big investments in AI, whether it's like IBM or KPMG or Thomson Reuters. And I've been asking a lot of them with these investments that they're making, like, what are the skills that they think are going to be considered even more essential when they're hiring, once they've like completely integrated generative AI? And they always highlight like those interpersonal skills, like especially if you think about like a front office person, an executive assistant, a paralegal, like the, the work that can't be automated is like the personality you bring to the office, the way you interact with your colleagues,

• 41:56Speaker 05

  the brainstorming you enable, all of that. And, and I've experienced that too in interviewing workers about whether they're worried about being displaced. They always go back to like what they feel can't be replaced in their job is just what they bring as a human to their office. And I totally agree. I think part of that is tied to the in-person work. And anecdotally, I actually experienced this because I did a little experiment that I, I talked to you about, Kevin, where I tried to chatfish my colleagues. So just speaking to everyone only via chat GPT for a week.

• 42:28Kevin Roose?

  So you use chat GPT to like do all of your office communication for a week?

• 42:32Speaker 05

  All of like my Slack, my email, my pitches to my editor, everything. Uh, my responses to sources and I found that like people I'd never met in person were totally willing to buy like my chat GPT voice was me and they were like, oh, great. Like you seem like a very formal professional person. And the people who sit near me at work when I slack them via chat GPT immediately responded and they were like, that's not you. Like that doesn't sound like your voice. Like the editor who sits next to me within seconds was like, that's not your voice. I know what you're doing. So I, I think that there is like an in-person relationship. Uh, you know, there, there's something, there's a quality to that that distinguishes us from the robots.

• 43:11Casey Newton?

  Yeah, I, I think there is a lot to that theory, Kevin. I think I pushed back on it slightly. I think there are forms of creativity that do not require coming into the office or collaboration and that companies may still find value in hiring people who have particular skills that are maybe a little bit more single player. Um, I do think that to the extent that these are collaborative jobs, that yeah, it makes sense that those are probably going to be less remote. But I just keep coming back to all the reasons that we wanted to go remote in the first place. One of them is just, it's expensive to live in big cities and it's hard to have the kind of life that you want in most American big cities today. And unfortunately, that trend

• 43:55Casey Newton?

  seems like it's only accelerating. So I think you're still going to have a lot of very talented people who want to live in the suburbs or the exurbs, or maybe even small towns somewhere. And I think that they're probably going to have the leverage to get the jobs they want, at least for some time. But I definitely think that that one is worth watching.

• 44:14Speaker 05

  I also think that part of this is a question of what an individual person values most. Cause I, I think in what both of your, you are saying, there's a question of like, is your life centered around your career? Is your career or job centered around your life? And I think a lot of people have reevaluated that question in the last few years. I think that, you know, a really tight labor market helped. There were like two job openings for every unemployed person. You know, almost 50 million people quit their jobs in 2021. A lot of people switched industries. I think a lot of people did some digging to ask. Like, did they want their job to serve their life? Or do they want their life to serve their job? And I think some of it in the end might come down to that because maybe someone doesn't want to be like

• 44:55Speaker 05

  completely integrated into the fabric of their company and they'd rather live somewhere more affordable and spend more time with their kids. And I think maybe finding a remote role is going to work really well for them. And I think for other people, their job is going to really fire them up. It's going to like light up their life in a different way and they're going to want to make certain sacrifices so that they can continue to be like really embedded in the culture of their company.

• 45:17Casey Newton?

  I, I do think there is something to this idea that if you're a manager and all of your employees are in the office, you can delude yourself into believing that they all really want to be there and they all love their jobs. If you can't see them every day, I think absurd paranoia starts to creep in. And, and, and I, and I think it can become all consuming for some of these managers. So it's, it's really, I don't know, interesting to think about.

• 45:43Kevin Roose?

  You know, my, my other theory on this is that a lot of this backlash to remote work was caused by these viral TikToks, these day in the life office worker TikToks. Have you seen these?

• 45:55Casey Newton?

  What, you're thinking a lot of the backlash was due to a viral TikTok?

• 45:59Kevin Roose?

  Honest to God, I think that, so these were, for, for background, these were videos that went viral on TikTok maybe a year or two ago, where it would be like, you know, I'm a product manager at Meta or I'm, I'm an engineer at Amazon or something. And it would be like these people filming themselves a day in the life. And the entire day was like, you know, at 11, I do a Peloton class and then I go to the cafe and I get my latte. And then I take it back and I relax poolside for an hour. And then I do a little bit of work in the afternoon and then, you know, around 5:15, I take off for the day. And these went viral. People used them as sort of symbols of like how lazy and bloated these tech companies were. And I think honestly, I, I've heard several executives at large companies cite these

• 46:44Kevin Roose?

  as something that made them very upset because they're like, this is what people are doing when we let them work from home? Like, we're definitely making them come back to the office.

• 46:53Casey Newton?

  It's a true, like, they hate to see us winning scenario where it was like, they gave employees great jobs and the employees enjoyed them too much. And so now the music is going to stop.

• 47:02Speaker 05

  I think some of them were hashtag acting my wage, which I think is a factor, but I think certain people were like, you know what? Like, I do not get paid enough to grind as hard as they want me to. And also, like, it was a real worker power moment that people were experiencing. I don't think it was just like product managers like coasting over to get like a matcha latte. I think people were also kind of like getting a little bit more empowered to draw lines about like how hard they were willing to work and when they just wanted time to themselves. So I don't think that was all like just people coasting. I think it was also people feeling empowered and saying like, I don't want work to completely control my life.

• 47:46Kevin Roose?

  Yeah, you talked about worker power and we talk a lot on this show about zero interest rate phenomena, these things that like happen because interest rates are low and money is basically free and companies are flush with cash and we're in a good labor market and workers have a lot of leverage. So is remote work a zero interest rate phenomenon? Was it something that was very easy for companies to sign off on when the economy was booming and they were all bursting at the seams with money and they could sort of afford to let some people slack off a little bit, but now that interest rates have risen and money is a little tighter, now they really need to batten down the hatches and get people back to the office? Is there anything to that?

• 48:30Speaker 05

  I think that's part of it, but I also think that's very dependent on like which CEO you ask, because there are certain CEOs, like I remember talking with Jeremy Stoppelman at Yelp and he's a big remote work advocate. And Yelp is permanently remote. People have scattered, they've moved around the country and they're still churning out product, they're still working hard. I do think that for some companies, remote work was part of the easy money economy. And I think that, you know, maybe in 2021, they felt like they had a little bit more margin for error. In terms of like, if there were a couple of people slacking off, they could sort of swallow that. And I do think that like, once companies had to start doing these layoffs, once the economy tightened in general,

• 49:13Speaker 05

  I think CEOs did feel a little bit of a need to kind of tighten things up and make people feel like they were under a little bit more of like a watchful eye. So I think it was, I think that's part of it at some companies. I also think there are CEOs who genuinely believe in remote work and are gonna, you know, ride this phenomenon out.

• 49:31Casey Newton?

  Yeah. Well, that sort of gets to my question, which is, I just wonder where you think we find equilibrium here? It feels like some kind of hybrid situation is going to be the near term future for many companies. That also seems like it's really uncomfortable for a lot of the managers at these companies. Do you think it kind of stays in that zone? Or do you think there is another twist still to come?

• 49:55Speaker 05

  I think that a lot of companies are finding that when hybrid work works, it works like fabulously well. I think you just have to think about it. And so I do think some companies have figured that that out. And I think other companies are going to catch up. I just think that we're in that kind of like awkward growing stages, period, where everyone's sort of, you know, throwing spaghetti at the wall to see what sticks.

• 50:17Casey Newton?

  Right. But one thing we know for certain, Lizzo is gonna clean up. I mean, I don't know how many of these concerts she can do, but she's gonna be making as much money as she wants, I think.

• 50:27Speaker 05

  I am still waiting to walk into the New York Times and see her in the lobby.

• 50:33Casey Newton?

  Well, my fingers are crossed for you, and please invite me if that happens.

• 50:37Kevin Roose?

  Thank you, Emma. Thanks for coming.

• 50:38Casey Newton?

  Thanks, Emma.

• 50:38Speaker 05

  Thanks for having me on.

• 50:42Kevin Roose?

  We'll be right back.

• 51:08Speaker 00

  Support for this podcast comes from Dell Technologies. It's time to put AI to work. Is your business ready? The challenge isn't getting started with AI. It's scaling beyond pilots while maintaining control of your data, governance, infrastructure, and costs. Dell AI Factory provides the integrated foundation you need to scale AI securely and efficiently with one trusted partner. Dell, AI on your terms. From idea to impact. Learn more at dell.com/yourwaytoAI. That's dell.com/yourwaytoAI.

• 51:42Speaker 03

  AI spend is growing fast, but most businesses still can't see exactly where that money is going. Ramp gives businesses visibility and control across AI providers with real-time guardrails by team and model, helping ensure spend goes to the highest ROI areas. Stop guessing where your AI spend is going. Get started with Ramp. And if you're new to Ramp, you can get $250 off at ramp.com/hardfork. Go to ramp.com/hardfork. Ramp.com/hardfork.

• 52:14Speaker 06

  Today's AI builders need infrastructure that scales with their ambitions. iRen is a vertically-integrated AI cloud platform, delivering data centers, compute, and software for AI training and inference, underpinned by its expansive portfolio of land and grid-connected power in renewable-rich regions globally. iRen's AI Cloud gives direct access to Nvidia GPUs, scalable from hundreds to thousands. For builders, that means direct access to AI compute, rapid scaling, and predictable pricing. iRen, AI Cloud for the next big thing.

• 52:48Casey Newton?

  Kevin, you know, from time to time, we like to look away from our laptop screens and look at bigger screens to sort of wind down at the end of the day. Have you been watching the new season of Black Mirror?

• 52:59Kevin Roose?

  You know, I watched the first episode of the new season last night, and I am ready to talk about it.

• 53:05Casey Newton?

  Yeah, and full disclosure, the idea to watch the first episode of this show came from a listener who emailed us saying, hey, Kevin Casey, this is totally up your alley. So thank you for the suggestion. So if you haven't seen Black Mirror, it is the show that asks the question, what if computer is bad? Or as Daniel Ortberg at the Toast many years ago, very memorably, put, what if phones, but too much, which really I do think gets at a lot of the dystopian ideas in this episode. Before we dive in, we should say like, we are going to spoil the episode. So if you would rather be surprised by all the twists and turns, feel free to come back to this segment after you have finished. But let's get into it, Kevin.

• 53:48Kevin Roose?

  So this episode of Black Mirror is called Joan is Awful, and it's about a woman named Joan, who's sort of like a mid-level manager at what appears to be like a big Silicon Valley tech company. And she discovers one day that there is unbeknownst to her, a TV show being made about her life, starring Salma Hayek as her. And basically, this show, which is running on this fictional version of Netflix called Streamberry, is using some kind of like generative AI quantum computer to like, take the things that she is saying and doing, and put them into a TV show that is basically like a one-to-one representation

• 54:32Kevin Roose?

  of her life. She's very surprised, obviously, to discover that the show is being made without her knowledge. But she then discovers that she in fact did sign away her life rights, her story, to Streamberry in this like, very lengthy terms and conditions that of course, like everyone who has ever seen a terms and conditions screen, she did not read, she just scrolled to the bottom and clicked OK. But now, Streamberry is using all of the data from her Devices to kind of collect a version of her life and use generative AI to like quickly turn it into a TV show. That ends up revealing all kinds of unpleasant things about her to the people in her life.

• 55:07Casey Newton?

  Yeah. And so that's already enough to chew on, I think, for one Black Mirror episode, but then they decide to throw in even more onto the pile. And one thing that they add is, well, what if we're all living in a simulation, right? Because as the episode goes on, the characters learn that they're not even in base reality. And in fact, the character of Joan, who you were first introduced to, who's played by Annie Murphy, who a lot of people will know from Schitt's Creek. She's not even the original Joan. There's actually a base reality Joan who you meet later in the episode. And then, of course, there's just the basic labor anxiety of are all creative writing and directing and acting jobs about to be automated away by AI? So a lot of stuff that was in there, Kevin.

• 55:52Kevin Roose?

  Yeah, it was sort of like talking to like your most conspiracy theory minded friend about like all of their ideas about, you know, your phone spying on you. And maybe we're all in a simulation and, you know, maybe aliens are real. Like it didn't go into aliens, but it might as well have because it really did get in like a sort of a grab bag of all of the tech related conspiracy theories from the last few decades. I mean, one fear that I have about this episode of Black Mirror is that some like startup entrepreneur is like out there right now saying like, oh, I could probably get funded to make TV shows of people without their consent that are like produced using generative AI. Like kind of in the same way that like, you remember the Black Mirror episode where like you work for these like merits?

• 56:33Kevin Roose?

  And you have to like ride on stationary bikes to like accumulate more merits that you need for basic daily activities. Like, I feel like multiple startups were launched because like someone watched that episode and was like, that's a great idea. I could make a billion dollars doing that.

• 56:47Casey Newton?

  I mean, to be clear, we already have TV that is made without people's consent. One of the most buzzed about shows this year, which is called Jury Duty, it's on FreeVee, which is owned by Amazon. Oh my God, these names. But it's about essentially a fake trial where everyone is an actor, except for one juror, who was apparently not told in advance that the whole thing was fake. So one of the challenges in making these sci fi shows is just to stay one step ahead of reality for at least as long as it takes to get the thing on the air. And I think, you know, Black Mirror succeeded here, but there'll be other cases where, you know, the auteurs don't make it.

• 57:24Kevin Roose?

  So what about this idea of a TV show that is created using generative AI? Do you think this is actually going to happen? Do you think there will be shows on Netflix two or three years or five or six years from now that are entirely created by generative AI?

• 57:39Casey Newton?

  Yes, do you remember Nothing Forever?

• 57:42Kevin Roose?

  Yeah, this was the like 24/7 endless episode of Seinfeld.

• 57:46Casey Newton?

  Yes, and it was not a hit in the way that Black Mirror is a hit, but it definitely had a moment online because people liked the idea of a never ending episode of Seinfeld that was just generated by something like ChatGPT. Now, it is true that part of the appeal was just the novelty of it, right? This was a curiosity, oh, look, they're making a Seinfeld-like show using AI. But, look, you know, one of the things that the Black Mirror gets right is it's expensive to make TV. And so if you could just have a computer do every single bit of it, there are some people who are going to want to use the computer to make every single bit of it. Now, I expect for the foreseeable future, anything made by generative AI that represents itself as a TV show is probably going to be pretty bad.

• 58:27Casey Newton?

  But as with everything we talk about with AI, the question is just how long will that be true for?

• 58:33Kevin Roose?

  I'll take the other side of this. I actually think I've changed my mind on this. I think I would have once said that like, fully AI-produced TV shows and movies are kind of inevitable. But I've been thinking about something in social psychology called the effort heuristic. Do you know about the effort heuristic?

• 58:49Casey Newton?

  I skipped that day of class, so no.

• 58:53Kevin Roose?

  So this is a concept that psychologists use to basically talk about the way that how much we value something depends a lot on how hard we think someone else worked on it. You know, they've done these studies where they give people identical pieces of candy, but some of the candies have a nice note on them that says like, I picked this just for you, hope it makes you happy. And the other pieces just have a note like, I picked this randomly. And then they have people eat the candy and rate how it tastes. And interestingly, researchers have found that people prefer the candy that has the nice label on it. Because to them, it signals like someone cared, someone worked hard to make this candy selection for me, even if that's not true.

• 59:34Casey Newton?

  And you know, Kevin, this is a great time to point out that every episode of Hard Fork takes over two years to produce by a team of more than 500 people. But anyway, go on with your story.

• 59:45Kevin Roose?

  So basically, this applies to all kinds of fields where part of what we value about watching, you know, a show like Black Mirror, for me, it's like succession. Part of what is so fun about watching that show is just you can kind of see the craft in it. You can see how hard the writers worked to like, get the exact right line or the exact right shot. And you can just sort of appreciate the mastery of it. And I think that one effect that generative AI is going to have is just making everything it touches look extremely cheap to us, right? I wouldn't watch an AI-generated TV show in part because I would assume that it's garbage. But even if you told me that it was good, I would probably still be reluctant

• 1:00:28Kevin Roose?

  because part of why I watch TV is to like, see the effort of other talented people who are making things for my enjoyment.

• 1:00:35Casey Newton?

  Yeah, I can understand how you would resent that. And here's why I think there is something to this idea. When the chatbots all came out, they took over all of the social networks as people would type anything you can imagine in them just to get them to write scripts or tell jokes or whatever. And whenever I would see one of those screenshots of text, I would just immediately dismiss it. I would think, there's no way I'm going to spend some of my precious moments on earth, reading the like, random firings of a neural network. Like, because to your point, there was no effort put into it. So why should I care? Way before AI, I once read a line that stuck with me, and I believe it's something like, what is written without effort is read without pleasure. And this was in the context of writing

• 1:01:19Casey Newton?

  advice. Like, you know, you actually have to put work into it if you want it to be good. But I think it's doubly true in an age where the quote unquote, writing is just a series of predictions.

• 1:01:30Kevin Roose?

  Yeah. I mean, I actually wonder if the advent of generative AI will kind of lead us back to a more artisanal way of producing stories. Like, you know, sort of the way that like, fast food produced slow food. Like, after the fast food industry sort of proved that you could mass manufacture food and that it was like, decently good and that People would buy it. There was sort of this backlash that resulted in like all these farm to table restaurants where they would like come to your table and explain where the vegetables came from and tell you how the chicken was raised and like, you know, really like put in a lot of visible effort to try to communicate like we are not mass producing this. This is not fast food.

• 1:02:14Kevin Roose?

  And so I wonder if there will be something similar with sort of entertainment where you'll have your AI generated TikTok videos or whatever that you watch for sort of cheap pleasure. But when you really sit down in front of a big screen to watch something that you want to blow you away, you're going to go for the kind of laborious human versions.

• 1:02:32Casey Newton?

  Yeah.

• 1:02:33Kevin Roose?

  I mean, I love watching my favorite part of like any, um, do you watch like nature documentaries ever like?

• 1:02:39Casey Newton?

  Like Planet Earth? Sure. Yeah.

• 1:02:41Kevin Roose?

  So my favorite part of nature documentaries is always the behind the scenes episode that they put out after the last episode where they tell you how they actually filmed the thing. And I just always love that kind of behind the scenes detail because it's so easy to appreciate the effort of someone who like sat in like a little hut in the wilderness for six months to like get the perfect shot of the snow leopard on the one day that it like emerged and tripped the camera sensor. Like that is the kind of delicious detail that just makes it so much more enjoyable, knowing that someone spent all that effort just to get you one shot that you saw on your TV. Like that is the kind of thing that makes me actually think that AI generated TV and movies may be a long way off.

• 1:03:23Casey Newton?

  Yeah, I think that could be, but also I think eventually some generation of children will be born that never knows anything different and just accepts the AI generations as as good. But hopefully that's a long time off.

• 1:03:36Kevin Roose?

  Yeah.

• 1:03:37Casey Newton?

  One final thing about this episode, uh, because I do believe that this episode of Black Mirror made a contribution to the culture, which is that, uh, the character's shorthand for the quantum computer in this episode is the quantum computer. And I just actually want to use that. Like, I think that's a good term for a quantum computer. And so I will be saying quantum.

• 1:03:58Kevin Roose?

  All right.

• 1:03:58Casey Newton?

  Thank you, Black Mirror.

• 1:04:28Speaker 00

  Support for this podcast comes from Dell Technologies. It's time to put AI to work. Is your business ready? The challenge isn't getting started with AI. It's scaling beyond pilots while maintaining control of your data, governance, infrastructure, and costs. Dell AI Factory provides the integrated foundation you need to scale AI securely and efficiently with one trusted partner. Dell, AI on your terms. From idea to impact. Learn more at dell.com/yourwaytoAI. That's dell.com/yourwaytoAI.

• 1:05:02Speaker 03

  AI spend is growing fast, but most businesses still can't see exactly where that money is going. Ramp gives businesses visibility and control across AI providers with real-time guardrails by team and model, helping ensure spend goes to the highest ROI areas. Stop guessing where your AI spend is going. Get started with Ramp. And if you're new to Ramp, you can get $250 off at ramp.com/hardfork. Go to ramp.com/hardfork. Ramp.com/hardfork.

• 1:05:34Speaker 06

  Today's AI builders need infrastructure that scales with their ambitions. iRAN is a vertically-integrated AI cloud platform, delivering data centers, compute, and software for AI training and inference, underpinned by its expansive portfolio of LAN and grid-connected power in renewable-rich regions globally. iRAN's AI Cloud gives direct access to Nvidia GPUs, scalable from hundreds to thousands. For builders, that means direct access to AI compute, rapid scaling, and predictable pricing. iRAN, AI Cloud for the next big thing.

• 1:06:08Casey Newton?

  Hard Fork is produced by Rachel Cohn and Davis Land. We're edited by Jen Poyant. This episode was fact-checked by Caitlin Love. Today's show was engineered by Alyssa Moxley. Original music by Dan Powell, Alicia Byut-Too, Marian Lozano, Sophia Landman, and Rowan Nemestow. Special thanks to Paula Schuman, Pui Wing Tam, Nellga Logley, Kate LaPreste, and Jeffrey Miranda. You can email us at [email protected]. Au revoir. Au revoir. Bonsoir.

• 1:07:17Speaker 07

  Companies are pouring money into AI, but few leaders can explain how those AI investments work with the systems you already have. You know, the ones that actually run your business? Without Systems, you don't have to start over with AI. You can choose the projects where AI will have the most impact. And OutSystems gives you the platform to make it agentic on your terms, with complete control every step of the way. So you can trust AI with the mission-critical systems that run your business. Own your agentic future. Learn more at outsystems.com/NYT. That's outsystems.com/NYT.