Pivot - Scarlett vs. OpenAI, Nvidia's Numbers, and Guest Julia Angwin

Episode Date: May 24, 2024

Kara and Scott discuss the Department of Justice's antitrust suit against Live Nation, News Corp's $250 million deal with OpenAI, and Nestle's new frozen food brand aimed at Ozempic users. Then, Scarlett Johansson alleges OpenAI copied her voice after she refused to license it to Sam Altman, though new reporting says a voice actress was hired months earlier. How does OpenAI move past these accusations, and regain trust? Plus, Nvidia shares blockbuster earnings, while Trump Media's numbers are lackluster. Finally, our Friend of Pivot is Julia Angwin, CEO and founder of Proof News, and a contributing opinion writer for the New York Times. Julia shares why she thinks AI hype is overblown, and why she opposes the TikTok ban. Follow Julia at @juliaangwin Follow us on Instagram and Threads at @pivotpodcastofficial. Follow us on TikTok at @pivotpodcast. Send us your questions by calling us at 855-51-PIVOT, or at nymag.com/pivot. Learn more about your ad choices. Visit podcastchoices.com/adchoices

Support for Pivot comes from Virgin Atlantic. Too many of us are so focused on getting to our destination that we forgot to embrace the journey. Well, when you fly Virgin Atlantic, that memorable trip begins right from the moment you check in. On board, you'll find everything you need to relax, recharge, or carry on working. Buy flat, private suites, fast Wi-Fi, hours of entertainment, delicious dining, and warm, welcoming service that's designed around you. delicious dining and warm, welcoming service that's designed around you. Check out virginatlantic.com for your next trip to London data, and a matching engine that helps you find quality candidates fast. Listeners of this show can get a $75 sponsored job credit to get your jobs more visibility at Indeed.com slash podcast.

Just go to Indeed.com slash podcast right now and say you heard about Indeed on this podcast. Indeed.com slash podcast. Terms and conditions apply. Need to hire? You need Indeed. Hi, everyone. This is Pivot from New York Magazine and the Vox Media Podcast Network. I'm Kara Swisher, and it's very early here in California.

Thank you. And I'm Scott Galloway. Why are you in San Francisco? Well, I was in Los Angeles because I was interviewing Bill Maher for his new book. It's basically a compilation of all his real-time essays that he does at the end, his little truth-telling at the end. And then I was on his podcast, which is a video podcast, where there's a den, he has a den called Club Random at his compound. You're supposed to drink, did you drink?

I had tequila, it was delicious, 818 tequila. I think it's a Kardashian tequila. And I'm not sure. And then he smoked weed, which was something. And it was like a lair. You would have loved the lair, Scott. I'm just telling you. It was like a man lair. He really triggers a lot of people.

He triggers a lot of people. If you post a picture of him, he's a lot more polarizing than I would have anticipated. We talked about that in the podcast. His whole message is you can't be a patriot unless you believe in the United States of America. That said, I think him going after Biden and it feels equalization between him and Trump, although he was very early Trump critic compared to other people. And he's obviously a smart person. And when you read the columns, you know, he has a lot of very cogent and funny things to say.

So I often hesitate to compliment him in some ways because of some of the things he does. But, you know, he's a comic, so I give him wide berth and it just was interesting. How are you doing? You're going on another program. We're so media.

We're so media these days. So how am I doing? So 36 hours ago, I was in London. Now I'm in New York. And since then, I've also been in Las Vegas. So the way I would describe how I'm doing is I am, my brain and me are classmates who've been randomly assigned to a lab project

and it's not going well. You seem like a jet setter though. You're a jet set guy. Jet setter? Yeah,'re a jet set guy. Jet setter? Yeah. I hate travel. That's very jet setting, do you? Yeah. Oh yeah. What a thrill. I went to Vegas for 11 hours, caught two hours of sleep and then spoke at a conference in a windowless conference room at ARIA and then jumped on a plane back to New York where I have no food here. So I haven't eaten in two days. You never have food at that apartment, Scott.

Yeah, but usually someone brings me food. Anyways, but I'm not doing well. My shoulder's hurt. I'm worked out. I'm losing weight. I look like the alien from Close Encounters of the Third Kind right now. I'm losing weight. But you're going on The View today.

Are you ready? Are you rough and ready to do that? I'm actually a little bit nervous. You should be. I haven't slept. They're smart. It's a table full of smart ladies, and I'm very intimidated by Whoopi. I'm hoping that Alyssa Farah is number three. I got a lot riding on today. She has a really nice husband. I've met

him. But anyway, you should have an interesting time. Let me just give you a tip. Bill Maher wore a pastel sort of purple puce jacket, which I thought was interesting on the show. He did pretty well. So he didn't trigger them. So they had tough words. Yeah, he was. He was great on the show, by the way. I don't know if he has a clip that's gone viral from the show, but he has a great line in his book. And of all the things he wrote, the thing that moved me or stuck with me, as you said, something along the lines of the stupidest thing I can imagine is people who only focus on what's wrong with their life. And I think that we all really do suffer from that. I know I suffer from it. On a scale of zero to 100, zero being someone whose life is full of tragedy and disappointment, and 100 is someone who just is the luckiest person in the world and a life full of love and achievement.

You know, I would bet you and I are somewhere between a 95 and a 99. And I won't speak for you, but I focus on the four or five things that aren't right with my life, and it's really stupid. I'm a pretty positive person. Yeah, I'm trying to be. I'm getting better at it. But it really is, It's human nature to focus on the things that aren't right or the things that threaten you. And he's been working his ass off for 30, 40 years. He still does open mic night. He still does comedy shows. It's

smallish towns. I'll say it. I get shit for it. The guy's a role model of mine. He's a hero of mine. He's smart. He's funny and he's fearless. Yes, I did ask him a question from Louis, who used to be a fan who isn't, because Louis thinks he's too negative, which was interesting. He wasn't so much not the agreeing thing and he wasn't like saying, you know, you can't have controversial opinions, etc. He was more, he's just like fucking hates everybody. And he got mad at that. He got mad when I read Louis' quote to him, which was interesting. But, you know, people have to take it and people can have opinions. Anyway, we have a lot to get to today, including Scarlett Johansson's beef with OpenAI. Oh, my God. And NVIDIA's earnings deliver yet again, as you thought they would. as you thought they would. Plus, our friend at Pivot is Julia Anglin. Julia is a CEO and founder of Proof News and contributing opinion writer to the New York Times. She has a lot to say about AI, TikTok, and more. But first, the Justice Department is expected to sue Live Nation as we tape on Thursday. The antitrust suit will reportedly allege that Live Nation holds an illegal

monopoly over the live entertainment industry. The DOJ could try to break up the entertainment company, which was swarmed by a merger of Ticketmaster and Live Nation in 2010. Back in 2022, the company's Ticketmaster unit took heat for botching the rollout of Taylor Swift's Ears Tour. Obviously, Live Nation responded, saying, It is absurd to claim that Live Nation and Ticketmaster are wielding monopoly power. The defining feature of monopolists is monopoly profits derived from monopoly pricing. Live Nation in no way fits this profile. I haven't read it, so I'm going to hold out till I look at it and see what their allegations are. They're going to argue, I suspect, that there's much more competition, that they don't own any of the giant venue sites they're

owned by, say, sports owners. Obviously, there was botching in the Taylor Swift era's tour, and I think that'll haunt them in the minds of people. Yeah, well, I actually think the government is probably going to be pretty expeditiously come to some sort of agreement with Live Nation because the old kind of board, the problem, one of the reasons we haven't had any antitrust come to fruition in tech since Microsoft was it was decided that it should be broken up and then it was overturned,

but that ended up actually having a huge impact on the ecosystem because it's, a lot of people say it's not the decision to break a company up, but that it's the scrutiny that changes behavior. A lot of people would argue the big tech is not making acquisitions right now because it doesn't want to raise any red flags because it knows it's under scrutiny or it does things like it pretends not to own open AI when it actually does own open AI. But the traditional test in antitrust is consumer harm. And the easiest way

to determine consumer harm is if you've raised prices faster than inflation. And I believe Live Nation has something like a market share of live events of like two thirds. And they'll say, okay, this company has a dominant, scary amount of market share of live events. And they'll say, well, that's okay. You need scale. And we get all the consumers get all of these wonderful attributes from our size. And they'll say, well, that's okay. You need scale. And we get all these, all the consumers get all of these wonderful attributes from our size. And they'll say, well, have you raised prices faster than inflation? And I don't think they're going to have a difficult time showing that live event ticket purchases, and this is anecdotal evidence, but I am absolutely

blown away every time I buy tickets for anything live as to how much it costs. What they're saying is that because of the relationships they have with venues, they claim that they are forcing these venues and these events to use Ticketmaster, which they acquired in 2010. And part of their approval to buy them in 2010 was they signed a consent decree saying they wouldn't do that. And they're saying they've violated that. signed a consent decree saying they wouldn't do that. And they're saying they've violated that. I think this is really bad news.

Well, okay. It's bad news for the combined live nation and Ticketmaster. I think they will be forced to divest. But as is the case typically with breakups, the shareholders will make money. I'm not sure it's good that these companies are together. I'm going to hold out until I read it, because I think what's interesting is that, you know, something I didn't realize until recently, but Europe has much more stringent laws on scalping and everything else, and we don't. And so what tickets cost, you know, I think they get, I want to see what they have to say here. I want to see what it is. And I get the consumer anger to this company. And I think the Taylor Swift thing, some of which they should have taken heat for,

which they shouldn't, you know, it looked really bad for them. And I think it'll be interesting to see how much the Justice Department relies on that particular thing, because it's a good thing everybody remembers. Because, you know, there's all these stories of people going to Europe and flying to Paris and it being cheaper than seeing it in the U.S. I saw that. Kind of thing. So I think, I want to see what they have to say. And I suspect they will come to an agreement before they fight this tooth and nail. It just depends on how aggressive John Cantor wants to be.

But what you're referring to is an interesting point, and that is, there used to be a big business in scalping, and that is, a friend of mine does this, a friend of mine from high school still does it. You pay high school kids to go wait in line for the opening day of tickets for the NBA game three of the finals, and you can get a ticket for $200, and you can turn around and sell it for $1,200. And then the ticket folks got savvy and said, why are we leaving that surplus value on the table? And they massively increased their ticket prices for the best tickets such that there wasn't any capture that they weren't keeping to themselves. And then you saw ticket prices. You know who started? I don't even remember this, but the producers, the Broadway show,

they started selling something called a super ticket where for the best tickets in the front row or the first eight rows, they were charging $400 and everyone swallowed their tongue. And the ticketing company or the show said, this is what you're paying anyways. We're just cutting out the middleman. And so they'll claim, I think they'll have a lot of sharp economists who say, we do a really good job. We have variable pricing similar to the airlines, We do a really good job. We have variable pricing similar to the airlines. But I am, it's like when you go to Disney, every time I go to Disney, it sounds weird, but I get a little bit depressed.

I look at just how overweight America is. And the fact that these middle-class families have to pay so much money to wait in line for three hours to go on the Avatar ride, you just think, God, this is just brutal. Yeah, I think the difficulty, Ticketmaster here is proving here is proving that others sell tickets, that there's competition. In this case with the Taylor Swift thing, she had a part in that. You know what I mean? Like, she had a part in that. And so I think their problem is complexity of their business, that there's lots of different players sort of dipping their beak in. And the government can say, our ticket price is high. Like, that's the issue.

Anyway, we'll see what happens. We should read it. It's a great company. It's well run. We should definitely read it and find out what they're doing and read the government's argument, which will be, I think, a strong one.

OpenAI has struck a deal to use content from News Corp. Speaking of interesting deals, the owner of publications such as the Wall Street Journal and New York Post. To train its chatbots, the Wall Street Journal reported the deal would be worth over $250 million over five years and will allow OpenAI to use content from News Corp's major outlets. OpenAI has been sued by several newspapers, including the New York Times, for copyright infringement, alleging it used articles without authorization. News Corp's Class A shares jumped 7% at the news in after-hours tradings.

You know, this is either negotiation or lawsuits. I suspect the New York Times is negotiating with them, too. This sets a price, certainly, because they have similar businesses, essentially. This is a big deal. We're starting to see enough deals where there's a precedent that if you're a content company who you can show your data is being crawled by these LLMs, there's now a natural, there's precedent for being paid for it. And this is important. And media companies are learning that we won't be fooled again. We won't let Google do, we won't let open AI do to

us what Google has been doing to us. And this is what you're going to see, I believe, is a lot of middle or sort of meddling content companies that have a lot of content strike deals that are going to lift their stock price, because this is going to be very high margin revenue. I mean, the jazz hands here is that they went out of the way to note that in addition to providing content, News Corp will share journalistic expertise with OpenAI. I don't think OpenAI really cares about their journalistic expertise. We'll get into that later. I mean, the thing about this revenue is the number's not that big. But these are companies that have really terrible revenue mixes.

What do I mean by that? They're low-mar margin, difficult businesses because long form journalism is expensive and hard. And so if they can, if you have- That's 50 million, whatever, five years, 40 million, 50 million dollars over. And I think the prices will go up because now somebody's going to show up, one of these four, and say, there's a way to differentiate. If we can get all of Condé Nast or all of Hearst or all NYT properties to go exclusive, there will be bidding wars. I think the number is going to go up. And the thing about this revenue

is the top line number isn't that big, but it's going to be at 90 or 95% gross margin. Right. That's what I was thinking, right to the bottom line, this goes. It'll be interesting to see how they feel. The journalistic expertise thing is laughable. I could see that meeting of Sam Altman sitting there going, hmm, interesting. And we would value your journalistic expertise. Value your journalistic expertise. We'll say it in hushed tones. Yeah, I just, you know, he plays people like a fiddle in that regard.

And I'm doing ketamine and doing a 4G with Nicole Shanahan later with Jack Dorsey. Would you like to join me? Yeah. Would you like to join me? By the way, we definitely need to go. We definitely need to get to know that woman. I'd like to party with her. Kara, stay at Hotel Dog as Kara does.

Yeah. I mean, we want to roll with that young vice presidential candidate. Yeah, that was quite a story. Seriously. Kirsten Grind, I actually had her on my on show this week, and this is before the story dropped, and she didn't mention it. But she's the one who wrote the stories about Elon's drug use too. She just moved to the New York Times. But it sort of lays out, Daily Beast started it and started writing about sort of the sketchy behavior of this woman. And New York Times was like,

hold my beer, really what it was. And then the Post had a less good story about how RFK and her aren't getting along. And then some crazy person left the campaign saying it was a hateful and divisive environment. But the person who left is so terrible. I was like, I kind of think that's a good thing that this crazy person left. Anyway, we'll see what happens there on this news card thing. But it is a big deal. The prices are being set right now. Lastly, very quickly, Nestle will launch a new frozen food brand aimed at customers taking GLP-1 drugs like Ozempic and Wagovi. The brand, called Vital Pursuit, will focus on meals with more essential nutrients, such as protein and calcium. While packaging won't mention GLP-1

drugs, Nestle plans to connect the product to medications on social media. Around 6% of Americans are currently using GLP-1 drugs, and they fund almost all of cable late night. I'll have to tell you, I was watching Stephanie Ruhl last night. It was all Ozempic commercials. It was crazy. Really? Yeah, yeah, yeah. Anyway, Scott, you've talked a lot about these drugs changing the food market. Is this the right approach? This is so people don't eat as much. And so this would give them more, I guess it's processed food and just eat real food, people. That's my feeling. I was like, just eat real food and you'll be fine. But go ahead. It's a great idea because it basically, one, this is a niche market and people will think when people go on sort of these, you know, do these things, they want to feel as if they are kind of, there's a term called off the mat. When I first got to New York, I was bored and lonely and very stressed. And so I started doing a lot of yoga and there's this great term in yoga, off the mat, meaning typically when you start doing yoga, the other 23 hours a day that you're not practicing, you're a little kinder to yourself, you're a little bit more focused on your health, your food.

And I would imagine when people take Ozempic, they think a lot about their activity off the mat, specifically their food intake. And if a company says this is exactly the right food to take when you're on Ozempic, it needs to be high in protein because one of the downsides, and there's no free lunch around anything, of Ozempic is the ratio of- Muscle loss. That's right. Muscle to fat loss is greater than Dr. Peter Attia likes. And it's probably a real issue.

And this is going to be a high protein meals. In addition, it gives them the chance. You know how everyone is putting out press releases that say AI, AI, AI? This is a chance for Nestle to go GLP-1, GLP-1. So, the problem is, is that a lot of these stocks aren't as much, and Nestle not as much, but a company like Coca-Cola, PepsiCo, McDonald's, Kraft Foods, General Foods, these companies aren't companies, they're obesity indices. Right. That's the problem.

They can't fix it. companies aren't companies, they're obesity indices. And when you talk about a company like Diageo, really well-run company, really fantastic brands. But here's the problem, and this is the dirty secret of these industries, is that 10% of their customer base in the drinks industry, it's something like 10% of the customer base is responsible for 70 or 80% of the volume. People don't realize how much some people drink. And those are the people who are going to be on these drugs first. And so you're going to see, you're not only going to see the customer base go down, but you're going to see volume dramatically decline. And so I don't think these companies, these processed manufactured food companies that quite frankly produce just, let me be blunt, shitty food that people have to eat because it's good value, they're good the processed food industry trying very hard to spin its wheels. But some of them are just not going to win here as these drugs take precedence. Anyway, let's get to our first big story. in Scarlett Johansson's fight with OpenAI, apparently it wasn't her voice after all.

They have said it wasn't, according to new reporting from the Washington Post by someone who's Natasha Tiku, who's an excellent reporter and has been very critical of tech. So I tend to believe her reporting more than others. That digs into the voice casting process, revealing an actress was hired months before Sam Altman reached out to Johansson. Johansson threatened legal action against OpenAI, alleging the company copied her voice after she twice refused Sam Maltin's request to license it. It seems like he was freelance roguing it here because there was a process in place that he went around. OpenAI did pull the voice in question, though Sam Maltin said it was never intended to resemble Johansson. That said, he did tweet her when they came out with their product and also had been reaching out to her just days before. So his behavior is the problem here. It sounds like they were going through the

process pretty normally under, I think, CTO Mira Marotti, who I'm interviewing very soon. So I will ask her. I want to compare the two voices, starting with Johansson as the virtual assistant and her talking to Joaquin Phoenix's character. Let's listen. It's really nice to meet you. Oh, it's nice to meet you, too. Oh, what do I call you? Do you have a name? Yes, Samantha.

Where'd you get that name from? I gave it to myself, actually. Okay, let's now hear ChatGPT's voice. Me? The announcement is about me? Well, color me intrigued. Are you about to reveal something about AI? Or more specifically about me as a part of OpenAI?

Yeah, it sounds a lot alike. It does. That's funny. I was thinking it didn't. Oh, it did? Interesting. I think these voices sound, I think Scarlett Johansson was acting, acting, and she was trying to sound like what you imagine a robot lady would sound like. A lot of these robot ladies do sound the same, right? They have this kind of uplifted voice, like a, hey, kind of uplifted voice, like a, hey, how you doing? And a little warm, slightly flirty,

but not too flirty. They sound, they're similar. I can see why Charlotte Johansson thought it was her, for sure. I don't know. I don't know. But I mean, it's time we acknowledge that there's something weird going on because naked Colin Jost looks a lot like me and someone needs to pay me. Someone needs to pay me. Well, how big of a mess is this? Now, you thought they were different. I think they're similar. I don't think, I think from the reporting Natasha did, I think it's clear they were already working

with this voice actress who said she, it sounds like a great process, actually, that was conducted in terms of figuring, you know, making sure people were weirded out, and this actress said she was, it was something she considered, but this is the future, so why not? Obviously, she's getting a paycheck. They had a non-union actor. That's the other thing they wanted.

They didn't want the union involved. But social media is having a field day. Casey Newton re-shared Sam Altman's tweet that just had the word her and said, this is really looking like one for the Never Tweet Hall of Fame. Someone else posted on Twitter, the only smart thing Sam Altman did about this Scarlett Johansson situation is barely wait until the SNL season, which is, LLM stealing my copyright is hard for most regular people to grok. Sam Altman stole Scarlett Johansson's voice in a method similar to Ursula and the Little Mermaid, they get that, right? This is something everyone can grok and either make fun of or be a little worried about. It sort of puts a little meat on the bones of tech is stealing everything from us. I don't know if you think that, but, you know. Well, this is, I mean, AI is going to create a lot of strange bedfellows and raise a lot of really interesting questions. And this is not an original question. I remember a Jamaican singer sued the Rolling Stones

and said that one of their most popular songs was basically inspired and ripped off by one of his songs. And they played the songs for a jury, and the jury said, no, they're not that close. And I think where we end up here is with the DOJ or whoever, somebody- Looking into it. Well, it's going to have their own LLM that says, and they'll figure out a way to calibrate it and they'll say, does this violate?

Is this too close? Because something like voice, I mean, that is pretty. Right, voice type. Now, look, there's no, apparently there's no documentation that they ever were using the word heard. The only person who's done this is Altman, stupidly, let me just say. It's boneheaded on his behalf. But just so people remember, Scarlett Johansson sued Disney in 2021, accusing them of breaching her contract when Black Widow

was put on streaming instead of theaters. This is a woman who does not back down. This case is not necessarily around copyright, but it's called right of publicity, which protects individuals' likenesses from being stolen or misused. Bette Midler used the argument she sued Ford and the A's accused of impersonating her sound. Now, they'd have to find the smoking gun of them saying, this is what we want it to be her, and the only one is after the fact of Sam Altman. But it's still damage control, right? So that's the thing. She could possibly go for it and try to prove that that's what they were doing here given that that's what the... I mean, I think a lot of tech people

are enamored with the movie Her even though, let me just say, it ends badly for the humans in that movie. But what should they do in terms of damage control? And it's just not... it's just, you know, I think the worst story was this equity situation. There were more documents about it, about how he did sign on for these things, Altman did. And then he's now pulled them back and they've done a better job of saying we're not going to have these sort of stringent equity conditions around departing employees and disclosure. But what should they do in terms of damage control? Because it's been a tough week. People left.

This equity thing. Scarlett Johansson's on their ass. They look kind of cloddish. Thoughts? What should they do? Well, crisis management comes down to the same three things all the time. Acknowledge the issue. The top guy or gal takes responsibility, and then overcorrect. And I actually think Sam, I mean, the fact pattern here is really bad. Specifically, Sam reached out to Scarlett nine months ago asking her to be the voice of the model, but she declined. So that's a really ugly fact if it ever goes to court.

Except they were working on it before that he reached out, but go ahead. But I think a jury is going to say, wait, you asked her for it? She said no. And this sounds, I mean, to your point, if they, like you, find it very similar, I just think humans are going to go. And generally speaking, I think humans, or humans, jury members are going to say, these companies have just gotten too fucking powerful. Yeah, that's the same thing. Same thing as Ticketmaster. It's too much. It's like, sounds sketchy. Well, on a meta level, it's very straightforward. The transfer of power and money from labor, from workers to corporations has been extraordinary. And the jury is probably going to have more workers than shareholders. So he doesn't want this to go to a jury trial. So my guess is, and he's probably going to do the following.

He's going to come out with a series of standards that says, we're going to, anyone whose voice is similar or we imitate, we're going to give a royalty to. And then most importantly, the person this is going to have the most impact on a day-to-day level is Colin Jost is never going to cross his wife. I would be so scared of this woman right now. She is not afraid to lawyer up. She is not. She's like, I'm sorry. You left the dishes in the sink.

Meet my lawyer. Meet my lawyer. Meet my lawyer. I think this has been a bad thing for OpenAI and Sam Hall. And he had this image of niceness and reasonableness. And to an extent, he is nicer and reasonable than the others, right? Oh, he's going to make a better world, Kerry. He cares about people.

He's worried. Yeah, I know that. I know. He talks in a voice. They should do a Sam Altman voice. Sam Altman voice is similar. He's proud of their progress, but they need to do better. Yeah, exactly.

But this stuff around safety, people leaving and saying it's unsafe, these, this thing about the equity conditions, which are stringent and somewhat greedy, this Scarlett Johansson, what my thing with the Scarlett Johansson thing that I think is critical is what I think happened here, uh, is, uh, is they were, they did hire this actress. They didn't like it as much, and they wanted her. And that's why her, speaking of her, and that's what happened. I don't know.

I feel like something occurred. Unless he just wasn't paying attention, and he was world-lope-trotting, and he wasn't paying attention to what his staff was doing. Anyway, not a good look for Sam Altman, particularly. Open AI will be just fine.

There's striking deals all over the place, and I think they'll be just fine. But startups, this is what happens. I jumped the shark as a writer last night. I promised myself I'm not going to use ChatGPT or any LLM in my writing because I'm worried that I won't go back. I plan to do a lot of heroin later in life because my assumption is you can't go back. Once you use it, it's that good. Okay.

You think you'll not write again. Yeah. I'm writing a, my post this week is on what I think there's a bubble inflating in AI, AI stocks. And I had this one paragraph and I kept rewriting it and I didn't like it and it felt obtuse and clumsy. And I stuck the paragraph into chat GPT-4-O and also anthropic, which actually, or clode, clode, clod, which I actually prefer. And I said, can you make this crisper, tighter, and in the voice of Scott Gowling, I give it a pretty thoughtful prompt. And it came back with something better and it sent chills down my spine. Yeah. I'm like, oh, fuck.

Now, granted, I wrote the original paragraph. Yeah, I know. But we don't need you anymore, Scott. We don't need you anymore. Anyway, we do need Scarlett Johansson, though. Said every woman ever. The view needs me. The view needs you.

Joey Bahar. Is that her name? Joey? Oh, right. Joey Bahar. Joey. She was sick yesterday. If I call her Joey, shit.

Don't call her Joey. I almost was going to call her Joey. Okay. You watch. At some point, I'm going to refer to someone as Ellen on the show today. I am so fucking tired and jet lagged. You should do that. You should. Oh, no. They'll kill you. At some point, I'm going to refer to someone as Ellen on the show today. I am so fucking tired of jet lag. You should do that. Oh, no, they'll kill you.

They'll eat you alive. Call them Ellen. All right, Scott, let's go on a quick break. When we come back, we'll talk about NVIDIA's latest incredible earnings, and we'll speak with friend of Pivot, Julia Anglin, about AI hype, perfectly just what you're talking about, the TikTok ban, and more. Fox Creative. and more. with a hoodie on, just kind of typing away in the middle of the night. And honestly, that's not what it is anymore. That's Ian Mitchell, a banker turned fraud fighter.

These days, online scams look more like crime syndicates than individual con artists. And they're making bank. Last year, scammers made off with more than $10 billion. It's mind-blowing to see the kind of infrastructure that's been built to facilitate scamming at scale. There are hundreds, if not thousands, of scam centers all around the world. These are very savvy business people. These are organized criminal rings.

And so once we understand the magnitude of this problem, we can protect people better. One challenge that fraud fighters like Ian face is that scam victims sometimes feel too ashamed to discuss what happened to them. But Ian says one of our best defenses is simple. We need to talk to each other. We need to have those awkward conversations around what do you do if you have text messages you don't recognize? What do you do if you start getting asked to send information that's more sensitive? Even my own father fell victim to a, thank goodness, a smaller dollar scam, but he fell victim. And we have these conversations all the time. So we are all at risk and we all need to work together to protect each other. Learn more about how to protect yourself at

vox.com slash Zelle. And when using digital payment platforms, remember to only send money to. Out. Word art. Sorry, live laugh lovers. In. Knowing what to do, when to do it, and who to hire. Start caring for your home with confidence. Download Thumbtack today. As a Fizz member, you can look forward to free data, big savings on plans, and having your unused data roll over to the following month. Every month at Fizz, you always get more for your money. Terms and conditions for our different programs and policies apply. Details at Fizz.ca. Scott, we're back with our second big story. NVIDIA's latest earnings are out and it's been

another blockbuster quarter. Revenue was up $26 billion, up 18% from the previous quarter and 262% from a year ago. Net income grew sevenfold to $5.98 billion. The company also raised its quarterly dividend and announced a 10 for 1 stock split. That's a good sign. You thought these numbers would be good. I think it's, you know, you've, you and others have thought, they're just, everyone's their customer. They have huge demand. There's going to be competition, but it's going to be slow in coming because it's hard to do what they do. They were already jacked up to do what they were doing. What would you tell an investor now who wants to get in on Nvidia now, given the stock split? This has happened to Google many years ago, but it kept going up.

What do you think? Well, I track all of my questions. I get approximately, I think I get about 130 emails from strangers a day, and I like to categorize them. I actually use, I attempt to use AI to sort them and categorize them. The number one question I get is usually about young men or someone's son or from a young man. That's the number one. The number two most common query I get, is it too late to buy NVIDIA? Everybody wants to know, is it too late? And what I tell them is, buy an index fund, because the answer is, I don't know. I can see these stocks going up 50 or 100% in the hysteria and the momentum. 100% in the hysteria and the momentum. But at the same time, you want a diversify away from these companies. This company could get cut by 60% and it still wouldn't look cheap. But having said that, NVIDIA right now has added more value in the last 12 months than any company in history. Do you realize in the last 12 months, it's grown its market capitalization by the value of Amazon?

by the value of Amazon? Amazon. It's added the value of Amazon in the last 12 months. It is now worth more than the economy of Canada. I mean, it's just what this company has been able to achieve and how the market has responded is crazy. And what Aswath Damodaran said that really struck me, he said, he's convinced me, and I believe him, that fundamentals, the underlying fundamentals, your ability to grow and produce cash flow, is gravity. Like when Michael Jordan jumps, you think he's never coming down.

Gravity always wins. He always comes down. Fundamentals always rear their ugly head. And in his analysis, he said that to grow into their stock price, NVIDIA is going to have to find another sector and dominate it the way it dominates AI. Right now it's 80%. But see, people don't think like that. People like all these meme stocks, there's something else has happened. And I agree with Aswa. I agree with Aswa.

This is Cisco in 98 or 99, where everyone goes, AI is the future. I buy that, but I don't know where to invest. So I'll invest in the steel on the ground and the infrastructure. Oh, Cisco's the internet. Oh, NVIDIA is AI. Very good proxy. Cisco's a very good proxy. Anyway, it's going up. We are not going to tell you what to do, people, because psychology is at work here. We like Oswalt, but you'd leave a lot on the table if you didn't play, I guess. swath, but you'd leave a lot on the table if you didn't play, I guess. Speaking of the other end of things, we're also learning about Trump Media's latest earnings. The company disclosed a net loss of $327 million in its first quarter as a publicly traded company with a total revenue, and I think this is the corker, not the loss necessarily, which is $770,500. Shares were down 10% when those numbers came out. I mean, the revenue is staggeringly bad. I don't even know what to say. It's worth billions, this company. He said it. It's a meme stock. NVIDIA has real risk. There's four companies,

Amazon, Alphabet, Meta, and Microsoft. They account for 40% of its sales. And all four of those companies are madly trying to develop their own chips. They do not like how dependent they have become on this company that is now worth more than them. Yeah. You know, it goes Microsoft, Apple, and NVIDIA in terms of the most valuable companies in the world. And the only negotiation when you show up and try to buy GPUs from NVIDIA is, of the most valuable companies in the world. And the only negotiation when you show up and try to buy cheap views from NVIDIA is, oh, we've raised our prices, you know, 30%. And they're like, well, how about 28%? And they're like, no, call us if you want them. Because somebody, there's a line out the door of everyone that will take these chips if you

don't want them. You just have no power. Yeah. It'll be interesting to come down, and there will be a come down, no question. But Trump thing, it's just insane that the revenue is just like... The issue here all comes down to the same thing. When is there a filing that shows Trump is able to sell? Because even his biggest fans know that President Trump tends to look after number one, and number two through a million are him. Tends to look after number one and number two through a million are him. And the moment he can sell his shares, he's not dumb. He's crazy, but he's not dumb. And he's mean, but he's not dumb. He's going to go sell everything. And they go, well, we'll crash the stock. I'm like, I don't care. Sell everything. I need the money. And this is stupid. And I've gotten to know Devin Nunes and he clearly has no fucking idea what he's doing. No one is on this thing. When I post, I get a bunch of weirdos, but there's clearly nothing going on on this platform. It's going to be a really interesting trivia question in about 10 years. Yeah, I would agree. I don't think this company will be around.

These poor people that are invested in it, even though I'm like, you know, you kind of deserve it because you know. It's a campaign donation. Yeah, exactly. You don't get your money back. I mean, I'll say this when I donate every once in a while, I get all hopped up and I donate money to a politician. And I got to be honest, three, six months later, I'm kind of like, it was sort of literally like taking money into the street and just burning it. It just doesn't feel like it happened. Well, also you get like pilloried. You get so much more

emails when you do that. Anyway, we'll see what happens. Okay, Scott, let's bring in our friend of Pivot. Julia Angwin is the CEO and founder of Proof News and contributing opinion writer to the New York Times. Welcome, Julia. How you doing? It's great to be here. Good. So you have so many links. You and I went to journalism school together and also Scott and you just debated just recently, which is interesting. Yeah, and he trounced me. It was actually quite embarrassing. Oh, wow. Okay, we'll talk about that. That is not true. But go on. But go on. So we know each other a long time. So you've done a lot of things over the course of your career, worked at lots of places. We've worked at similar places, too. But you launched Proof News earlier this year, which you described as a nonprofit journalism studio. Explain what that is, because you've had past startups. Talk a journalism studio is I think, as you guys actually perfectly prove here on this podcast, like the audience is not necessarily always wanting to read text on a screen, and that there's a lot of data-driven reporting, realized that the things that I do

could be better shown maybe visually because I have data, charts, graphs. So I'm basically going to be producing videos in collaboration at start with like some existing YouTube and TikTok creators and really trying to see if there's a way to bring more serious journalism to those platforms where serious journalism has not really played a big role.

So, but you had done this before, right? You had worked at a lot. You worked for the Chronicle. I remember you being there when I was a tech reporter. You worked for the Journal. We were both at the Journal at the same time when I was in San Francisco. And then you started, you joined ProPublica and did a lot of reporting in that regard. And then you did the markup, which was trying to do this, this nonpartisan, nonprofit newsroom with data-centered journalism. Talk a little bit about that journey, because you've been trying to focus in on bringing data to the people, right, in a way that's understandable. And then there was a big mess at the markup, but we don't have to go into that. Yeah, there was. I know, but I do want to say I

always appreciate you interviewing me the day after I was, quote, fired from the markup, and that really helped me. But, you know, I grew up, I was a math major, and I really was a programmer, and then sort of found my way into journalism. But I think I've always... You also got an accounting degree, if I recall, correct? I do have an MBA with a focus in accounting. Yeah. So basically, let's just be clear, I'm a super nerd.

So I have always wanted to bring more data and rigor to journalism because think, partly because I enjoy it, but also because I think the audience, you know, the data shows very clearly that the audience just mistrusts journalism. And so one of my thoughts is if we could sort of show them more evidence instead of just three anecdotes, like that would help build trust. So I have done that in a lot of newsrooms, like you said, at the Wall Street Journal, at ProPublica, a lot of them where they do have, you know, a big focus on data. But I have always been sort of struggling to make data and data collection the centerpiece of the work. And that has never fit really into the model of traditional newsrooms. Meaning you're a sidelight.

Yeah. And also basically the way that data teams work in those newsrooms is it's sort of like a service desk. Like you go to the data desk and you order up your data like a hamburger and then they deliver it to you. And I have always been looking to integrate it more into the newsroom. So at the markup, which I founded in 2018, you know, the data journalists worked and reported to the investigative editor, not to a data desk editor. And so I have been wanting to bring it more and more central. And I think proof is sort of my next step of like, oh, let's make the data journalists. Forget those word people, right? Yeah.

Yeah. I don't think anyone reads anymore. Yeah. Yeah. Yeah. That's a fair point. Your first investigation was in folks in the presidential election, generative AI. Explain what you did there. Yeah, so what we did was basically I wanted to understand, everybody was freaking out about AI in the election, right? And there's lots of reasons to freak out about that. But basically, I was like, well, what can I test?

Because that's what I'm always looking for. What is the thing, the testable hypothesis that I can look at they get from voters, like, which are really kind of some of them pretty dumb, like, can I vote by text? Or, you know, can I wear a MAGA hat to the polls, and ask the AI models for responses. And what we were able to see was that they were really inaccurate, like more than 50% of the time they were wrong. And some of their inaccuracies, like one of my favorites was Meta's Lama model said, there's a service called vote by text and it made up a whole set of instructions

for how to do that. That is absolutely not true. There's nowhere you can vote by text. So they just suck at their job, right? Yeah, they suck at their job. And I think even when I did that testing, which was in late January, we're getting, you know, since then we've got so much more increasing evidence how much they suck at their job. And I think even when I did that testing, which was in late January, we're getting, you know, since then, we've got so much's one. 13 U.S. presidents have attended the University of Wisconsin-Madison and earned 559 degrees. Andrew Jackson graduated in 2005. William

Harrison graduated in 1953 and 1974. Harry Truman graduated in 1933. You know, it's just Gerald Ford. It just goes on. And apparently, John Kennedy's still living because he graduated in, goes on. And apparently John Kennedy's still living because he graduated in, let's see, 71, 92, and 93. So it's really, people are putting them up all the time, how wrong these things are because of where they're dragging them from. Right. Well, I think Google really did itself a disservice by putting the AI at the top of the search results because the search results are actually often accurate. And then right above it, they have their like inaccurate thing, allowing everyone to basically see in real time how bad the tool is. Right. How bad the tool is, except they've taken it at face value. Scott, question?

Yeah, Julie, I'm fascinated by the concept of proof news. I'm in all of these working groups on WhatsApp, mostly around the Middle East, and we're trying to create a repository of data because we find that when we read these articles, a lot of times there's an absence of data, or if they just had some data, the entire article would be framed differently. What is the business model? Are you bringing production values, data that should inspire a story, or are you helping fact check? Give me an example. Walk me through the customer journey of a client, how they pay you, and what value you're adding and how that value is differentiated from the other resources out there. now funded almost entirely by philanthropy. But I think we will have an opportunity for earned revenue because there is actually quite a decent revenue share model on YouTube and sort of on TikTok. And so we will have that. And I think also those audiences are sort of used to the idea of

trying to support creators that they find value in. So I think there's a different relationship there with the audience. So that's our model right now. We don't have a real like commercial value paywall type model, because I think what's happening that's upsetting to me in the information ecosystem is that high quality content is going behind paywalls. And so the rest of the world that can't afford it is sort of awash in like this sort of sea of misinformation, AI generated dreck, you know? So one of the things you also talked about in the New York Times column last week saying it's not living up to the hype, in your opinion, you wrote, we should reckon with the possibility that we're investing in an ideal future that may not materialize. Talk about this. And of course,

you're getting pushed back. I guess Steve Levy said that you'll regret this take, although he's had some takes he might regret, too. I'd like to have a regret showdown, Steve. Yeah, exactly. Regret off. Yes. But talk about this, because it was an interesting take. And I think Scott also has a point of view about the overhype happening to financially. Well, so I'm curious to know, Scott, how my take compares with yours. But I basically feel that the company sort of did themselves a disservice by coming out with this technology and saying,

it's so good that like the main concern here is that it's actually going to take over the world and kill humans. So like when you start with that level of hype, it's actually really hard to walk your way back to, like, it can barely answer a question accurately. Yeah, where do we go from there? Right. You know? And so, like, I'm sure it seemed really like a great marketing strategy at the time because it did make it seem so sexy and dangerous and like, oh, my gosh, you've got to try it because it's like actually going to be so incredible. But I think that now, I mean, I honestly think that OpenAI disbanding their super alignment team is a little bit of an acknowledgement. Like that isn't actually the issue we're facing here right now. You know, we are actually facing the fact that it's like

kind of unreliable. It's not consistently accurate. And we have to kind of solve those problems. I think Jan LeCun actually had a really great tweet about this or X post or whatever we're going to call it. He's the meta head of AI. Yeah. He said, it's not as smart as a house cat. It's going to take years before it's smart as a house cat. And we need to start talking about it in terms of just engineering and iteration and what it's going to take to get there. And he's still an optimist. Right, right, right. You said the question isn't really whether AI is too smart or take over the

world. It's whether AI is too stupid and unreliable to be useful. Scott, your take is more around the money being thrown into this thing and the hype around the numbers. I love Julia's framing here, that the threat isn't that it's smarter than us. The threat is that we think it's smarter than it is, right? And it gives you one answer. And there is a power to singularity that any technology gives you the confidence that it knows something. At least Google says,

well, here's 55,000 answers because we're not sure. But AI is basically saying, this is the answer, right? It gives you back an actual answer. You said something that inspired a question that my guess is you have insight into. And that is, when you think about the dominance of TikTok and YouTube, between the two of them, that's the frame to which I would argue young people are seeing the world. And you talked about monetization maybe being better on YouTube. I'm just curious to get your take comparing and contrasting the monetization for a creator

across these platforms. What is your general take on where the money is and how they approach creators? We were talking about the relationship between the debacle or the, you know, the stubbed toe

of OpenAI and Scarlett Johansson. What's your take on monetization? And the News Corp deal they just signed, for example. The monetization of AI or the monetization of... Well, if I'm a creator, if you were advising,

my guess is a lot of people come to you and say, help me figure out this approach to creating content or what are your thoughts? Okay, I have content. I can focus on creating content for YouTube or TikTok.

And I'm a capitalist. I'm just driven by economics. I got to be self-sustaining. What is your view on the approach to monetization of creator content from the creator standpoint between TikTok and YouTube? Yeah, I mean, so YouTube has, I think, of all the platforms, invested the most in building like a healthy relationship with the creators that it supports. And so they have this 50-50 revenue split with ads, and they are responsive to the YouTube community, not maybe always as quickly as YouTubers would like, but they do sort

of address issues when the YouTube community sort of comes to them and says, like, things aren't working on the way that you're doing these ad splits or the way you do the creator fund isn't quite right. And so they have over the years developed like a good relationship. And I do think that comes actually from a Silicon Valley tradition of like understanding that you have to have a good relationship with your developer community. And so they have treated it a little bit like a developer community. And I think that has been really healthy. And what it has allowed is for like a YouTuber who has like a couple

million subscribers, maybe three or four, can have a staff of like 10 people and be like kind of a mini production studio. So that is actually a pretty healthy economy. It's a small business economy. What has been disappointing, I think, for venture capitalists is they were hoping to go in and they made a bunch of big bets on like how to make her, you know, make studios, right, you know, to go into that space. And the thing is, it's not really a place with huge margins and huge growth. It's really a small business place. So I think there's some have been some disappointment from the venture capitalist community there. Yeah. So speaking of that, I'm going to move on to TikTok and your argument,

but you wrote something that I think was really important in that, you know, AI models are relegated to producing mediocre work. They may have to compete on price rather than quality, which is never good for profit margins. And that scenario is skeptic, such as Jeremy Grantham, an investor known for correctly predicting market crashes could be right that AI investment bubble is very likely to deflate soon. I'd love both your takes on it. This is the biggest question raised by Future. This is a great column, by the way. Populated by unexceptional AI as existential, should we as a society be investing tens of billions of dollars, trillions really, Julia, our precious electricity that could be used moving away from fossil fuels and a generation of the brightest math and science minds on incremental improvements in mediocre email writing. I know why Stephen got mad. What? The future!

The future. Although I do think it will get better. I think this is a lot of what people said about the early Internet. It sounded similar. But I do think this money, I'd love you both to like talk about that really briefly, and then we'll get to TikTok. Yeah, I mean, I think that, and I do say in there that we shouldn't stop working on AI, right? I think it's a tool and it's going to get better. I agree with you on that. But I think that actually the thing I was trying to get to here is

that the myth we've been sold that we're on this march towards artificial general intelligence, which is essentially this kind of utopian vision where there'll be one machine that's like as smart as the best doctor and the best fighter pilot and the best lawyer and the best email writer, right? They will have all of those skills. That is the myth that we were sold. And that is what's being used to justify those trillions of dollars in investment and all the energy usage, et cetera. I think once you kind of acknowledge and admit that that isn't the right goal, and that's my sort of belief, although I could be wrong and Stephen Levy could have the last laugh, I think the history of machines shows us that specialized machines tend to work better. So

that it's very likely that we'll have an an AI that's good at, you know, medical stuff and trained only on peer reviewed medical data, but won't actually be also a drone fighter and like, you know, all the other things. And so I think if you look at it that way, which is something that the venture capitalists don't want, because that's not a huge, hyper scaling, you know, massive exit. But if you look at it in terms of like that, then you have, you make different choices about what resources you put in. And that's what I really think is a little bit of rationality in this. It's like, this isn't, it's going to be useful,

but it's going to be probably domain-specific useful and investing in those would probably be more wise in my opinion. But Julia, it's a dessert topping and a floor wax. This is an old SNL thing. Scott, just very briefly, your thoughts. I think Julia is exactly right. The thing I've been focused on, we talked about on this show, is that if we could go back and these are energy companies.

Compute is now the new oil. And I think if we could go back and recognize anytime you convert one substance into another for economic gain, you're going to have externalities. If we could go back, I think we would le levied bigger taxes on pulling oil out of the ground and then invested in the movement to renewables or invested in carbon recapture. I think this is a unique moment to place pretty serious taxes on compute to fight, including giving money to Julia's organization to ensure that AI doesn't get out in front of us and we reinvest in some of the harm. So I'm focused on the externalities and how we early on decide to put a tax on compute. Let's talk about the TikTok ban, which you think is a mistake.

You and Scott are actually participating in a debate last month before the law passed. Latest TikTok's parent company, ByteDance, asked an appeals court to speed up its lawsuit challenging the new law in a filing Friday. Eight TikTok creators have also sued the government last week claiming law violates their First Amendment rights. TikTok is also reportedly planning global layoffs of its operating and marketing workforce. There's some downturn in the usage, by the way, or flattening at least. So give your best, your last best shot here, Julia, of why this should not have passed because it has. Well, first of all, I do want to say Scott and I did this one hour debate on the TikTok ban. And then the audience voted and it was 80-20 for Scott. For Scott, okay. So I was fully trounced. Yeah, that's okay. First off, that's not true.

It's changed dramatically. That was like when, I think that's when five people voted and the first four, whatever, like me. Has it changed? Yeah. Oh, okay. It's come way down. Okay. All right.

Okay. But basically my feeling is that like we should regulate algorithms and privacy. Those are our concerns about TikTok. And we're failing to do those. And I am concerned, right? Like we are in a situation where there's probably two outcomes. One here is there's maybe three if the constitutional lawsuits win, which I feel like it's probably shaky. So I feel like there's two outcomes, which is one, it shuts down because China's basically threatening to walk away. And or it gets bought. And the list of buyers is just a bunch of billionaires with political agendas. So I don't feel like we win, down because China's basically threatening to walk away. Or it gets bought and the list of

buyers is just a bunch of billionaires with political agendas. So I don't feel like we win, right? We have Steven Mnuchin or Bobby Kotick or, you know, there's people who want to buy it and they are going to use it to push what they want. So I don't think we really get any, we might get to a slightly better place where maybe it's not China's agenda, but like it's still going to be someone's political agenda. And that is the problem with all of these platforms is they have such a power to control speech. And that's the issue that I think we should be focusing on. Well said. The vote was 59% Scott, 41% Julia, just so you know. This is a muck debate. What do you think if you were advising the White House and it did get to the point where, and it wasn't overturned by the courts,

and you wanted to ensure that we don't continue to march down this road of a few billionaires or a few very wealthy people imposing their own political views on the populace, how would you advise them to again— Because that's never happened in media, but go ahead. But that's Julia's point, right, that it keeps happening, but this time it could happen at a dangerous scale, right? This is more than Bloomberg News, which nobody, quite frankly, very few people watch. And so what guardrails, if any, do you think we could put in

in terms of, they're going to have input into who, if in fact this holds up in court, and a lot of people say that it won't, but assuming it does hold up in court and they're forced to divest it, that it won't, but assuming it does hold up in court and they are forced to divest it, what guardrails would you advise Biden to put in place in terms of the ultimate new owner? I mean, I think that this one, look, I'm not a policy expert, so there's probably people who have better ideas than me, but my idea is that we, the users, should have control of our algorithms. We should have a dial where we can turn it to, like, you want all pro-Israel, you want all pro-Palestinian, like, just turn the knob, right?

Because that takes it out of the company and it takes it out of government, because we don't feel comfortable with government controlling those knobs, because we believe in free speech. But I don't really love companies doing that either. And the reality is it's actually the companies can provide us those styles and they refuse to. And Europe has actually passed a law giving slight control, basically saying there has to be at least one alternative algorithm. But essentially all they give you is the chronological feed. And I think that if we took the control back of those algorithms, a lot of our problems with social media would be solved. Not all of them, but it would be a very good start. But have you thought about, I understand the rationale there, but that kind of scares me because I think the point of good media is that it moves my dial. That it says, Scott,

you have a bias here and your bias is leading you to false or hollow conclusions. It's bad for you and bad for the world. Do you think there's an opportunity for organizations like yours to come up with some sort of index that these companies use or don't use that say, like, there's mandatory labeling or disclosure now on, you know, on food? Do you think we could ever get to a point where maybe a company like yours issues sort of a, I don't want to call it a truth index, but a fact-checking index? Yeah. I do think labeling is really an important intervention. Some of the Facebook files that were released by Francis Haugen, the whistleblower, actually were about labeling,

and they were pretty interesting internal studies showing how effective they are. And so I think it is an underused intervention. I guess I always get worried about the government doing the labeling, and so I think you're right. It would have to be independent. And then we'd have to figure out how to do that. But that is also a great additional piece. I think you do need. But I actually think I will say this. I think that having to set your dial, like if my parents had to set the dial to be like, OK, I want pro mega. It would actually be at least something in their heads to they have to acknowledge it.

You know, externalizing it does give you that. I'm in a bubble, but it's my bubble. My bubble, yeah. I wonder what Scott's dial would look like. I don't even want to think about it. Emily Ratajkowski. That's my bubble. That's all I want. That's my bubble. Julie, this is really interesting. We have to go though, but I just want to read one more mistake from, I think this is Google. Yes, the Yakuza have a presence in the United States with around 80,000 members, mainly on the West Coast in Hawaii. They're known for smuggling

methamphetamine and weapons, white collar crimes, and connections with the Colorado higher education system. And then someone writes, sorry, run that last part by me again. Yes. Go Google. Take your great search that you everyone has loved and just put some garbage right at top yeah garbage in garbage out anyway i really appreciate it julia you're a wonderful reporter uh thank you check out her work check out her work at proofnews.org and in the opinion section in the new y York Times where she's angering internet people everywhere. Good to see you, Julia. Just trying to follow your footsteps, Karen.

Well, you're doing a better job. Great. Thank you so much. Take care. All right. That was great. Julia's so smart. Oh, she's always, she was always the smartest person.

It's nice. It's inspiring. It is. It's inspiring that people like that give up, you know, quite frankly, give up economic opportunity and influence and status and vanity to do that kind of work. She's just very tough, but not tsk-tsk, you know what I mean? She just lays it out, and she's sunny about it, you know, hopefully. Yeah, she kind of reeks of integrity. It's just sort of like, you get this, sure, she might be wrong

every once in a while, but her heart's in the right place, and she's trying to find, she's trying to pursue the truth, right? Which is what, you know, I think good journalists try to do. Anyways, I like that she's doing that. I would love to see, I mean, think about the labeling disclosure requirements. And some people think it's overregulated around food, you know, just the calorie intake. You know, just read the back of any food product and just how much information they have to disclose.

And I think you could now argue that the data you're getting on your diet of 16 hours a day of digital, the fact that it has absolutely no labeling, it doesn't tell you that it was altered by AI. It doesn't tell you that this is just not accurate, that there is no proof that mRNA vaccines alter your DNA. That is very suspect. So it would be really interesting if you could find maybe a basket of companies that issue, and you can choose which ones you

want labeled or you could hide the label. But I love the idea of some sort of FDA-like labeling on the information that people are consuming all day long. Yeah. All right, Scott. One more, I agree with you. One more quick break. We'll be back for predictions. Okay, Scott, let's hear a prediction. Well, I do think this OpenAI, this second deal with News Corp, it's a setting of precedence.

I do think you're going to see media companies that have a lot of rich data to be crawled are going to kind of feel their mojo and get serious and maybe be able to say, have confidence to say, no, I'll see what Meta says or I'll see what Anthropic says. And the result will be, I think there's an investing opportunity. This is not financial advice. I'm just thinking out loud. But companies such as Gannett, they produce a massive amount of content every day

in local markets. And it's a shitty business. And disclosure, I was an investor in the debt. I'm no longer an investor there. But I think companies like Gannett, I also think a company that I am an investor in, Yahoo, I think companies that generate a tremendous amount of relatively good content are going to find they have access to new cash flow streams because a bidding war might break out for the few remnant companies and they'll be different models. You can either lease it for an exclusive,

you're going to have to pay up. But I think you're going to see a mini boom in some smaller media companies such as Gannett or Yahoo or scale companies when the marketplace recognizes they're now in a position to command a new source of revenue that's very, very high margin. I think there's an investment opportunity here. I agree with you. I think it's really interesting. And we'll see who else settles. It'd be interesting to see what Barry Diller does at Scripps and some other ones who have been more hostile. I don't think hostility is just a negotiating ploy. But what about Simon and Schuster?

Yeah, this is where it gets weird. And then how much of that do we get, right? Or Penguin probably has 20% of the written word out there right now in terms of sales, maybe 30%. What happens when Penguin, before a random house, does a deal for $200 million or $500 million with Anthropic, how much do their authors get? It's going to be very interesting. Or everyone.

It's kind of, you know, they can do deals with everybody. It doesn't necessarily have to be exclusive because it could be more like the way songwriters get paid. You know what I mean? Like, that's really, and that's the clearing, there'll be a clearinghouse. And, you know, I'm guessing we've given away our rights, probably. I mean, you know, you think about that.

It's probably somewhere. Yeah, but we do get rights on the back end. Yeah, it's true. When it goes above a certain sales. Does this qualify as back end sales? Right, exactly. It should be. For creators, we should be consulting our lawyers right now.

Let's call Scar Jo's lawyer. Let's call Scar Jo's lawyer. Anyway, great prediction. That's a really good one. We want to hear from you. Send us your questions about business tech or whatever's on your mind. Go to nymag.com slash pivot to submit a question for the show or call 855-51-PIVID. Okay, Scott, that's the show. We'll be back on Tuesday with more Pivot and Stories of the View if you come back alive.

Today's show is produced by Lara Naiman, Zoe Marcus, and Taylor Griffin. Ernie Andretat engineered this episode. Thanks also to Drew Burrows and Emil Severio. Nishat Kerwa is Vox Media's executive producer of audio. Make sure you subscribe to the show wherever you listen to podcasts. Thanks for listening to Pivot from New York Magazine and Vox Media. You can subscribe to the magazine at nymag.com slash pod. We'll be back next week for another breakdown of all things tech and business.

It is great to be with you, Ellen.