Meta & YouTube Liable in Landmark Case, OpenAI Shutters Sora, How Claude is Gaining on OpenClaw

0:13Welcome, everyone, to The Information's TITB. My name is Akash Pasricha. It is Thursday, March 26th. We have got some big news this morning. The Information has published its 2026 list of general partners. We scoured our best sources across Silicon Valley to investigate who the most likely people are who will become the next general partners running some of the top venture capital firms. We've got 15 names on there from firms like Andreessen Horowitz, Lightspeed, NEA, Lux Capital. And next week, we have got a special few days of coverage lined up for you. We're going to be speaking with five of these investors.

0:49I'm very excited for those discussions. Be sure to check out the list. It is on our website right now. First up on the show today, Meta and YouTube have been found liable in a landmark social media addiction case. We're going to dig into the ruling and what it means for future cases. Plus, the OpenAI Disney deal is no more, and OpenAI is also shutting down Sora. We'll talk about what these decisions mean for the creator economy and how that business and sector is shifting in the AI era. We'll then discuss some of the key risks that AI poses for society with the co-author of the now-famous Cetrini article from a few weeks ago about he is out with another essay of his own.

1:30And we will wrap with a look at how Claude is gaining ground on OpenClaw as it adds new features for coding. It's going to be a great show, so let's get right on into it. A jury ruled Wednesday that Meta and YouTube harmed a user's mental health through negligence, defective product design, and a failure to warn of the dangers of their products. The landmark decision could affect the outcome of many other cases alleging social media companies are creating addictive products. Here to talk about the outcome in future cases are Jennifer Huddleston, Senior Fellow in Tech Policy at the Cato Institute, and Rebecca Haw Allensworth, a visiting professor at Harvard Law School.

2:09Welcome to you both. Thank you so much for being here. Rebecca, I want to start with you. What was your reaction to the verdict? Well, I think this is the right outcome. I mean, I think that these products are dangerous, and I think that Meta knew that and designed them to be so, and I think it's appropriate for them to have liability here. I think it's a little unclear going forward what it is that they need to do to comply with basic safety and to not be negligent, but I think this is a step in the right direction. Okay. Jennifer, what do you think? I think this is a hugely consequential decision.

2:39not just for these two companies involved, but for social media more generally. We have to think about what this new form of liability might mean for smaller and mid-sized platforms as well, what this might mean for free expression and the availability of these tools for all users, as well as the broader questions that we're seeing and the ongoing debates about the potential unintended consequences of policy approaches that could be trying to restrict access to social media in the name of protecting young people but actually have significant consequences for the speech privacy and information rights for all users of the internet including adults jennifer you say smaller platforms which platforms are you talking about here so while this particular ruling is only about meta and google google's youtube was also involved in this particular case one of the questions i think that will arise is what does this mean for say, platforms like Reddit or Discord or any other platform that may be starting up, that's now going to have to consider the question of, are they going to have to be prepared for a potential product liability issue that might arise, even if they find that their product is safe, that they might have to absorb the cost of that litigation.

3:54One of the advantages of Section 230 has been for particularly small platforms, if you're offering a product or service and someone uses it for something you never intended it to be used for, that your next call doesn't have to be a lawyer. You can try and focus on improving the product, on improving safety features. And even if you're vindicated in court, the cost of that litigation can be particularly costly to small and mid-sized platforms. So, Rebecca, Jennifer is bringing up Section 230, which is central to this. And we know that the tech companies, I think if they haven't said already, they likely will appeal the verdict here.

4:33And I wonder how you think that appeal could go, how important Section 230 will be to their appeal. How are you thinking about it? Well, I think it will be very important. How it will go is really hard to say because this is kind of unprecedented. There's not a lot of precedent here that really is going to tell us one way or the other. There's two things to think about. One One is the tobacco litigation, where there is product liability and strong product liability and led to major changes in the way that the products were designed and marketed to youngsters. And then there's firearms liability that kind of went the other way, that said that, look, the manufacturers of guns are not going to be liable.

5:13So I feel like this case, you know, we'll see on appeal which category it ends up in legally. As far as 230 goes, 230 is going to be relevant to the appeal, but I think there's a really good argument that the judge and the jury got it right here, which is to say that this is not really about the content. It's about the design of the products. And that's something that the platforms have great control over. They're doing what they're doing, they know what they're doing, and they know it's dangerous. Rebecca, what did you make of the amount of the punitive damages that we heard about yesterday? So it sounds like a lot to us, you know,$3 million is certainly money that i have lying around and it's that as much as was found for the liability phase so the total was six this is obviously just peanuts compared to what meta's bottom line is this is not a problem for them unless you know hundreds or thousands of these claims go forward and succeed and there are we know thousands of these claims pending so the question is what does this number look like when it's multiplied across the whole market and then that starts to look like real money to the company and it looks like maybe they would need to change the way that they operate But the fact that there was punitive damages itself shows what the jury thought of this conduct, and I think that that's important.

6:26Jennifer, what do you think that this verdict says about those other thousands of cases that we know have been filed? I think that the fact that there are thousands of cases that have been filed shows how this could open the floodgate to litigation and raises serious questions about what that may mean for expression online more generally, what that may mean for the availability of different platforms, not only for young people, but also for adult users. Additionally, I think there's a lot of complicated questions around youth mental health, and we want to make sure that we're taking those questions seriously and that we're not just rushing to point a finger at social media platforms as the problem causer and not looking at the broader context and really asking those deeper questions of young people themselves and considering how, in some cases, social media can actually be beneficial.

7:18This isn't a always harmful, always beneficial situation. It's something that's very nuanced to every potential user, every particular child, every particular family. And that's why it's so important that parents be empowered to make the decisions that are right for their children and right for their families. So, Rebecca, what I'm trying to figure out here is we know that we have these thousands of cases coming, or they've been filed already. I mean, the simple question is, is it a slam dunk now? I mean, does this case tell us exactly what is going to happen in those cases? Are there nuances that we should be thinking about?

7:57We don't even know how the appeal is going to go. What do you think? Yeah, so this is not a slam dunk. This is what's called a bellwether case. So given that there's these thousands of cases going, often they'll put a few of them up first to see how they go. And they're like test cases. And so what I would say is this is relevant information. This is a data point, but it's just one data point. And as you point out, there's going to be a lot of specifics to the cases, particularly about the person who was harmed. Having said that, I actually think that this case is a particularly strong signal in favor of liability in those other cases, because this was not a plaintiff who otherwise was extremely happy and well-adjusted, found social media, and suddenly found herself to be experiencing depression and other mental health issues.

8:41She had a lot of complicated things going on in her life, as many of us do. And so, in some ways, it's actually kind of a weaker case, excuse me, for the plaintiff than many of these might be. So actually, you know, it is just a data point. It's just one case out of many. And yet, I think it does mean something. So, Jennifer, let's now talk about this in the context of the other technology that is so popular right now, which is AI, and the precedent that you think that this could set for similar questions that could arise with these AI chatbots. And we haven't, you know, this is an issue that I'm sure will play out over the next few years.

9:20Do you think that this ruling sets a precedent for those cases whenever they come up? I think we're definitely seeing a debate about where the line between expression and product may lie. And that's going to be really important when it comes to some of these questions around AI. There are certain expressive activities that come into account on the developer side when determining how to program your AI chatbot. The same way that there are certain expressive activities that come into account when a social media platform is determining the rules for content moderation or the way that content may or may not be prioritized.

9:55So if we see that product liability is attaching based on expressive conduct, that could raise a lot of questions and a lot of potential points of friction and raise new questions in the AI space as well. Rebecca, what do you think? I mean, I think that's right. I think this distinction that Jennifer just made between product and expression is really core. And I think that the way that gets decided is going to matter a ton. Because to me, this case is about it being a product. But if it crosses over into the expression line, a lot of other things are at issue, including the First Amendment and Section 230.

10:29So that question is really in play when it comes to AI. What's the product and what's the speech? Jennifer, can you explain the product versus expression distinction? Just explain to us what you mean by that. a little more closely. So when I'm talking about the product versus the expression, it's a question of, is this an inherently faulty product? You made certain claims about what the product is or isn't capable of doing, things that might be more akin to our traditional fraud elements. Or is this about an expressive choice, the fact that the jury or a plaintiff did not like the type of expression that a particular platform was allowing or was prioritizing.

11:10And so there's going to be a lot of questions of, is this about an expressive conduct choice or is this about a product that is actually faulty and liable? Rebecca, last question for you. Do you think that these rulings will ultimately change anything about the behavior towards social media or towards these tools? I mean, liability is one thing. I'm thinking of the people using social media and the extent to which they know that these trials are even ongoing. The young people and how they interact with the platforms, the popularity of them. I mean, I'm thinking about how the platforms could change.

11:50We obviously know there are things about the product that, you know, maybe they have restrictions. Maybe they're more disclaimers. Maybe they have to take down some content. But do you ultimately think that it will change behavior towards social media in the end from the user's perspective at all? It might. It depends on the changes that a company like Meta feels like they need to make in response to this. So one thing that could come out of this is a warning. I think a warning, you know, failure to warn was a part of this case. I think a warning is very unlikely to change behavior, especially if we're talking about children clicking through like a warning about this addictive problems with social media.

12:25I think that's unlikely to work. On the other hand, you know, in an extreme case, if enough of these cases come through and enough of them are clear that children are particularly vulnerable, it could go the way of tobacco and there could be a ban on, you know, a certain age or like major age restrictions on who can use it. I think that would really change the way that people engage with these products and the way the products are designed. Jennifer, any last thoughts on that? I think, again, what happens next is a big question because it's a big question not only for how this could impact these particular companies, but what does it mean for the availability and accessibility of speech?

13:00There's also an ongoing set of cases involving challenges to the type of regulations in some ways that Rebecca just described, state-level regulations that have sought to engage in age verification or other limitations on social media platforms. When we're considering those, there are clear First Amendment concerns about how they could impact anonymous speech, privacy, and other rights of adult users as well as of young people themselves. Well, I want to thank you both for coming on. We really appreciate it. That is Rebecca Ha-Allensworth from the Harvard Law School and Jennifer Huddleston from the Cato Institute here on TITV.

13:36elsewhere in social media news open ai has shuttered sora and relatedly it has done away with its disney deal which seems to be great news for any creator who has been afraid that social media will just turn into a zoo of ai generated slop of course that's not to say that it hasn't happened already meanwhile there are also some big shifts underway with respect to where people are watching creators whether it is on tv on mobile or whether they're watching short short form or long form content. I want to bring on someone at the center of these dynamics. Aaron Dubobois is the founder and CEO of Spotter, a company that helps finance creators and connect them to different brands.

14:14Aaron, welcome to the show. It's great to have you here. Thank you so much. Appreciate you having me on. So I want to talk to you about the Sora news this week. So OpenAI says we're not doing Sora anymore and the Disney deal goes out the window. You talk to creators all the time. How How are they thinking about those headlines this week? Yeah, I don't think that creators are surprised that Sora was shuttered. I think that mainly the vote of confidence it gives them is that people are really valuing human creation with AI assist rather than just pure AI created video. And that AI purely created video doesn't really need its own platform to just be rolling out tons and tons of videos and that YouTube will continue to thrive for creators who are focused on building great story or focused on storytelling.

15:12So people weren't happy or sad. They just basically didn't think that the AI-generated content was ever going to catch on type of thing? Well, I don't think they thought AI-generated content will never catch on. I think it's more that they didn't think it needed its own platform. And clearly that showed, right? audiences didn't gravitate after the initial release to continue to be on that platform. And that, you know, creators are absolutely using AI and are seeing AI content show up on platforms like YouTube, like Instagram and TikTok. But that is a less, I think this is an event that gives them a lot more confidence that people are seeking human created content.

15:56Right. Now, Tell us a little bit about spotters business in the AI era. I mean, I think about the marketers that you talk to. I think about this push towards AI ads. Now that the platforms now have their own advertising business. I mean, has that at all affected interest in your own platform in selling ads via the creator economy at all? Talk a little bit about that. Yeah, I actually think it's enhanced the opportunity for the creators that we work with. We spotter mainly focuses on creators that are making long form content, 22 minute plus content that's showing up on TVs and is being consumed in the living room.

16:39And I think that the value of that human connection is increasing as we see more and more content come on platforms. And so I think that if you can actually work with a creator to create custom ads rather than AI generated ads, even if it's AI assisted ads, we know that individual creators have a connection to their audience that is a relationship and a trusted relationship and can actually cause audiences to act rather than just optimizing traditional advertising across a large platform. So you're saying the consultative approach of, you know, working with an advertiser to create sort of a bespoke ad as opposed, you know, that in your mind is sort of an enhanced product compared to just a, even though it's easier to create an ad, it's just like you don't really know if it's reaching the right audience.

17:34You don't really know if it's sending the right message. That's sort of the distinction you're making. Yeah, I think that's right. And I think it's the other point I'm making is that the value of experiences that show up as true human connection is growing. So it's just like in sports, the value of sports is increasing over time because people are seeking the relationship between the real life moment and their own life. right um and so i think when we think about creators who have millions and millions of uh fans right um that have been following them for years they're going to appreciate advertising from those creators more than they will just appreciate a random ad that's generated um from ai but that's a little bit more about not just how the content is generated but really about what the content is saying what the ads are saying and how they're connecting to the audiences so So leveraging that relationship between creators and their audience is a really critical marketing tactic that I think will become more and more valuable as we see more and more content being created.

18:39Now, you published a report this week on the era of creator TV, and we've, of course, seen the popularity of platforms like YouTube on television screens. And now there are people on YouTube that I think are gunning for Emmy awards, etc. And what was the message that you were trying to send with that report this week? The message we were trying to send or we're trying to send is that, you know, in an era of abundance where we showed that 26 million creators are generating 4 billion videos a year that show up as 200 trillion views a year. that it's really hard to just now have a strategy that says, hey, there's YouTube and there's Insta and there's TikTok.

19:27So just general platform thoughts is just too broad and hard to digest. And so when we thought about Creator TV, we said, look, there are pockets that there's a lot of different type of content. There's shorts, there's longs, right? There's documentaries, there's podcasts. And what we really wanted to do is say, well where is the most attention happening just like you would on traditional kind of linear television over the last a number of decades you would ask most marketers and fans would ask where is everybody watching what is everybody watching um and no one's really asking that question from an attention perspective meaning if there are two 20-minute videos which video is the one that's grabbing the most attention and that's probably the more valuable important video and so i mean defined creator tv we said well how many creators out of the 26 million creators are generating 22 minute long videos that are uploaded on a consistent basis that are generating a significant amount of viewership 100 000 views per per video um and when we did that analysis and had and very much focused on it has to be showing up on the tv screen and influencing culture in the living room it showed up as 7 ,000 creators out of the 26 million that are performing that or making that.

20:49That are doing it consistently. That are doing it consistently. And it's been one of those things where you say, well, what's the difference between TV programming and what we call creator TV? Well, the big difference is TV programming does all the things like consistency and repeatable formats that fans could depend on, but they don't have the creator connection between the creator and the audience, right? So Stranger Things is an amazing show. People show up to the theater for it, but the fans don't necessarily listen to the creator about what to wear, what to buy, even just what to do. Whereas in creator TV, the creators have a deep connection with the audience because they're really showing off their lives.

21:32Right. So is the creators that you have on your roster, I mean, I'm sure you're always studying what is standing out amidst this attention economy overload. I mean, is it the long form content that is really standing out and standing apart? Is that where you are advising all of your roster of clients to sort of direction towards? Yeah, I think there's two areas of consumption that are probably equal in number of views, but not necessarily the amount of watch time. So one would be shorts. There's an amazing amount of viewership in shorts, probably equal to or bigger than long-form video. But the amount of time people are spending with long-form is greater.

22:21And where the economics are best for creators is in long-form. So when you think about, well, why is it that consumers are looking at long and short? um both of those very let's say 22 minute plus content and less than 60 seconds require a lot less decision making by the by the uh audience in terms of what to watch so I can either just scroll and not have to think about it or I'm intentionally picking a show that's 22 minutes or plus and I can lean back in my living room and watch the error of you know five to ten minute content trying to find five different videos in a 30-minute session is just too much to ask for the audience to search for.

23:08So, yeah. So, let me ask you one final question then, connecting it back to the previous segment that we had. I mean, we have all of these social media trials happening in the background right now. They will continue to play out. Do you think that they will change any of the consumption on these platforms? I mean, it's so hard to know. And we, of course, we live on these devices, on these platforms. It obviously depends on the product changes that the companies make. But I mean, do you think about that at all with respect to what content creators should make and how they should make it and stuff like that?

23:45Well, I actually, I do think about it and think that this connection between the audience and the creator that's trusted actually creates a better, more healthy relationship between content and the people consuming it. I think parents, when they see content showing up on their TV screen, they can actually, first off, actually see it, right? And two, they feel more comfortable that it's a repeatable show that they can depend on, right? Versus a lot of random content that, you know, something bad could show up and influence your children in a way that may not be positive. So I think, you know, seeing consistent creators show up is a really important thing in terms of how that impacts the viewer.

24:35Great. Well, Aaron, I want to thank you for coming on. That is Aaron de Beauvoir, the founder and CEO of Spotter here on TITV. Thank you. Alap Shah was the co-author behind the Market Moving Cetrini article that a few weeks ago painted a stark picture for the future of the AI-powered economy. Now, Alop is back with another essay, this time describing some of the solutions that he thinks could help address some of the key risks that AI could bring to society. Alop is the managing partner of Lotus Management. I want to bring him on to tell us more about his thoughts. Alop, welcome to the show. It's great to have you here.

25:11Great to be here. So I just want to rewind a little bit. What did you make of the reaction to the last article that you wrote, which obviously blew up quite a bit. It was surprising in sort of how picked up it got and how much it got folks talking. But it was sort of, you know, we had an inkling that people were going to care, obviously not to the extent that they did and the extent the markets moved. But ultimately, we write about this. It was, I think it's just a confluence of a lot of events because this is something that everyone is viscerally feeling in their day-to-day lives. And we were among the first to kind of put it together to say, hey, there's a set of dominoes that could kind of fall here.

25:48But it's just every week, every day, there's more and more of a torrent of information around how much AI is progressing. And I think that's the new world we live in. And so I guess in that sense, it wasn't as surprising. So how did the reaction to that piece inform this piece you just dropped this week? The main way is it just delayed it, more than anything else, because a lot of things came up. But originally, had there not been such a hubbub, the goal was to get Peace 2 out within 10 days or so. But ultimately, it was intended in my head, at least, as sort of one cohesive piece with, you know, this is the setup in terms of this is what I'm seeing in part one.

26:26Part two, this is how things could go poorly if we don't react soon. And then part three is, hey, there actually is a very straightforward set of solutions here that we, you know, collectively as a society need to start talking about and thinking about. Okay. And so what are the solutions that you're pitching? You know, at a high level, there are a lot of different ways to get to the other side of this. you know, if there's the world, there's two different worlds that can happen here. The first is GDP really accelerates as part of this productivity boom, and there are a bunch more jobs, in which case none of these things are problems.

26:56And, you know, we're going to be in an amazing world. The risk is in between that move from productivity to GDP, lots and lots of jobs are lost. And in that world, we have to sort of bridge the gap that requires some sort of an impact on taxation where the jobs are being lost. And so specifically, the way that we like to frame here is we're trying to put something out that's very centrist, very much something that could be passed in a Congress, not today, but I think a Congress a year from now when, you know, it's much more clear the writing is on the wall. And so there's a lot of different ways to structure it.

27:29And, you know, we're trying, we're not trying to come in and weigh in on what, you know, the long term structure is in a post-work world or where a world where humans are not working very much, but rather, you know, what are the solutions they could get passed in the near term based on the systems that we have and the politics that we have. And at a high level, if you take a look at that, what we conclude is the simplest way to do this is to kind of avoid the third rails of things like a wealth tax and, you know, moving up sort of income tax brackets for individuals, but instead focus specifically on where, what's driving the drop losses, i.e.

28:00where is the displacement happening? So corporates that are able to, you know, like a block or, you know, the next generation of a block where they lay off 60, 70, 80 % of folks, and that immediately turns into the profit margins, those margins probably shouldn't be taxed at the same corporate rate that we would otherwise. And if we have some sort of a system wherein those companies that are benefiting the most and obviously have the most upside are paying a little bit more back into the system, that is sort of a nice transition to get from one side to the other. Okay. So I hear you. And by the way, there's a lot of ideas in the piece, and I would encourage people to go and read it.

28:36I guess the broader question I have for you, though, is how realistic you think that these policy changes are? And I mean, we've seen this play out a little bit with previous technologies. We've seen how some of it has gone. I mean, I like the ideas, but do you think that they actually will get implemented in reality? I believe that in today's world, absolutely not, right? Let's be serious. But I think the core points of core message - That's why it's called thought leadership. Well, but thought leadership - I'm just saying, like, it's hard. Like, I get it, and I just don't know if it'll happen.