Is Social Media the Asbestos of the Internet? with Matthew Bergman

Transcript

This transcript was generated automatically and may contain errors — the audio is the record.

0:00:02 Matt B.: The social media companies make their money by selling advertising, and the more time they can keep kids online, the more advertising they sell.

0:00:17 Matt B.: We took a different approach, focused not on the content, but on the design features, making the products addictive to young people, not by showing them what they want to see, but what they can't look away from.

0:00:40 Thomas: Hello and welcome to Opening Arguments. This is episode 1258. I'm Thomas Smith and in a moment I'm going to be joined by not one Matt, but two. Well, I suppose a Matt and a Matthew. We're going to be speaking with the Social Media Victims Law Center founder, Matthew Bergman. You may have heard about and we touched on the recent court victory that the SMVLC just had in LA against Meta and Google. Taking the approach that this is more of a product liability question than, And, you know, the whole 230 thing, Matt's going to do, Matt Cameron, that is going to do more introduction and lay the groundwork better than I probably could in this intro. So I'll probably just get right on over to the interview. So that said, we'll take our usual first break here, which you can avoid at patreon.com/law. Avoid the ads. Get the goodies. A lot of movies coming out. Could be in a day here. We're working on the edit. It's almost done. Thank you so much to our patrons for making the show happen. We'll take a quick break and then we'll be on with Matthew Bergman about the recent win in over social media companies.

0:01:44 Ad: [Ad break 1 — ad music]

0:01:44 Matt: Well, I'm very excited to welcome to Open Arguments our guest, Matthew Bergman, who's the founder of the Social Media Victims Law Center and a law professor at Lewis & Clark. I want to start by congratulating you on the big win in California in KGM, a huge landmark victory for social media companies. As far as I know, there's never been anything like it. Is that right?

0:02:02 Matt B.: That's right. This is the first time ever that a jury has found a social media company responsible for harming a kid. Indeed, it's the first time such a claim has ever been allowed to get to a jury in the first place.

0:02:14 Matt: Yeah, and we're going to talk about the basis and kind of what got you to that place. But I wanted to ask about you first. I understand you've been a litigator for about 35 years, and you got your start for quite a while in asbestos litigation. Litigation.

0:02:25 Matt B.: That's right. I spent about 25 years exclusively representing individuals who had asbestos cancer.

0:02:32 Matt: So what took you from asbestos to social media?

0:02:35 Matt B.: I was looking to do something a little different, something that had more immediate social impact in that asbestos has not been in the workplace for 50 years.

0:02:45 Matt: Fortunately. Yeah.

0:02:46 Matt B.: Yeah, and social media is unfortunately causing harm as we speak. I was motivated by Francis Haugen's brave revelations to Congress regarding what social media companies know about what their platforms are doing to kids. That coincided with the Surgeon General's report on the youth mental health crisis and a helpful opinion in the Ninth Circuit called Lemon versus Snap that suggested for the first time that there might be a way around Section 230. So I decided to use my experience as a product liability lawyer proceeding against social media companies using some of the same strategy and tactics that I'd used against asbestos companies.

0:03:11 Matt: It makes a lot of sense when you say it that way. And just to kind of back up, I want to make sure that we're laying the groundwork here. Frances Hangen was a Facebook whistleblower. What was it about her testimony that you found compelling or that was new to you?

0:03:37 Matt B.: Well, what was new to me is that by 2021, it was becoming abundantly clear that there was a relationship, at least temporally, between a youth mental health crisis and social media, but it wasn't really all tied together. What Francis Haugen's revelation showed to the world was that the social media companies know and had documented quite effectively this relationship between youth mental health harm and social media and had explicitly decided against design changes that would make their platforms more safe for kids.

0:04:10 Matt: And if anything, they were just continuing to make Facebook more attractive to minors, right?

0:04:18 Matt B.: They were. They were. And there were terms like an addict's mentality. There were terms like tweens are herd animals. There was some of the same smoking gun documents that I had had experienced and worked with as an asbestos lawyer for many years.

0:04:36 Matt: That's interesting.

0:04:37 Matt B.: The only difference was that the social media company's conduct really made the asbestose companies look like choir boys.

0:04:45 Thomas: I wanted

0:04:45 Matt: Yeah I've read Careless People, the other sort of whistleblow account. I was pretty struck even knowing a lot to hear it from the top. That kind of brings me to some of the things I want to get into here, because there's an interesting parallel with asbestos litigation and social media in that you do have to prove causation. And it's maybe not always as immediate because I understand, and I don't know that much about asbestose litigation, but my basic understanding is you often have to proof causation many, many, years after exposure.

0:05:12 Matt B.: That's true. And in an asbestos case involving anybody who had occupational or military exposure, that person was exposed to many different products, some of which can be attributed to a particular defendant and some of which can't. That's similar to a social media situation where in most instances, mental health harms are multi -causal. It is always possible to point to other things going on in a child's life that might have contributed to their mental health harm or worse. Just as in any asbestos case, it's always possible to point to other exposures to other products that contributed to the harm.

0:05:55 Matt: Just looking around at this, KGM, the minor plaintiff, obviously known by her initials in in the California case that got us here. I've seen some kind of unfortunate victim blaming this idea that maybe she had other issues, that somehow it was her parents' fault. But here's just a little bit about her and the other plaintiffs, who were bringing the suit and why.

0:06:14 Matt B.: Kaylee could have been anybody's daughter or anybody's niece. She got onto social media when she was nine years old and became addicted, developed severe body dysmorphia and depression and anxiety. That said, she was not by any way, shape, or form the most severely impacted of the social media victims that I represent. Many of the cases I have involve children that have taken their lives or been been hospitalized for long periods of time. This case was selected by the court, not as the most extreme example of social media harm, but the most representative sample. In Kaylee's case, there were other things in her life that were challenging, as there are for anybody. And you correctly surmise that the social media companies kind of put her on trial and her mom on trial and her family on trial with the hope of absolving themselves of their responsibility.

0:07:12 Thomas: To ask, as the non -lawyer trying to keep up, it sounds like you've had a lot of these cases. What do you mean by the court picked one? I guess I don't even understand.

0:07:19 Matt B.: Oh, Kaylee's case is part of a consolidated group of cases in Los Angeles County involving several thousand young people who have developed mental health harms or lost their lives through dangerous social media apps. What is typically done in these consolidated case groups is that there are several cases that are picked to be what are called bellwether trials, which are seen as representative of the group. Those cases go to trial and the resolution of those cases provide the parties with some yardsticks to determine the valuation of the other cases.

0:07:56 Matt B.: This is particularly important when you have a brand new type of lawsuit such as this, where you don't have a kind of an established Verdict history to know the factors that impact liability and damages.

0:07:56 Matt: Actually,

0:08:09 Thomas: So is that kind of like an informal thing, or is it like, that's really interesting, and then is it also the same as like a class action or not a class section.

0:08:19 Matt B.: It is not a class action. It is non -informal. There are various methodologies for picking the bellwether cases. In some cases, each side picks a few. In this case, the judge picked a few, the desire is to, you know, have a fair example or representative example of the cases that get tried to provide a basis to hopefully resolve the other cases. We have close to 2 ,000. Yes.

0:08:47 Matt: How many cases does your center have going right now all over the country? How many of those are federal versus state?

0:08:56 Matt B.: About a third of them are federal.

0:08:58 Matt: Yeah, I was interested in the state basis on this and the others. So I guess we should get into the basis of these suits because it's unusual, It's groundbreaking. Previously, as I understand it, when plaintiffs have tried to sue over things that have happened as a result of social media, for example, mass shootings or the kinds of mental health issues you're talking about, the social media companies have been protected by Section 230 of the Communications Decency Act, which protects platforms for the most part for things that are published on them. This suit takes a completely different route by looking at it as a product's, well, I guess product's liability is the best way to put it.

0:09:29 Matt B.: Yes, product liability and negligent design.

0:09:32 Matt: Right. Negligent design. It seems that you've been able to sort of bypass Section 230 in a way that I don't think people expected. What is kind of in your mind the difference between content and design for this purpose?

0:09:44 Matt B.: As it pertains to Section 230, it was enacted in 1996 when Netscape was the largest browser and Google didn't exist and Mark Zuckerberg was in junior high school. It was designed to first and foremost protect kids online, line, but also provide some immunity for what were then called bulletin boards that allowed third parties to post on their platforms,

0:10:05 Matt: Fond memories of BBSs, for sure.

0:10:10 Matt B.: Provided that if somebody posts something on a bulletin board that is potentially libelous, an individual cannot sue the bulletin board, so to speak. They can sue the individual who posts it. That fairly simple and non -controversial principle was expanded far beyond what Congress was intended by a series of court decisions that by the 2020s had essentially immunized social media companies for anything that happened on their platform that involved third -party content, which is virtually everything. As a consequence, social media countries were being held immune in cases involving child sex exploitation and child sex trafficking, drug deals, harassment, intimidation, all all sorts of activities that were going on online, even where the company had actual knowledge that this was being undertaken on their platform. In these cases, we're getting thrown out before they even were allowed to move forward. So they were thrown out right after they got filed. We took a different approach and focused not on the content, but on the design. Design features that are self -consciously focused on making the products addictive to young people, Not by showing them what they want to see, but what they can't look away from. Design features that prey upon the neurologic immaturity of young people, as well as their need as adolescents for social acclimation. Pointed out that these companies, through some very, very sophisticated psychological and neurologic research, were not seeking to show kids what they wanted to see, but were seeking to enhance their engagement by showing them stuff they couldn't look away from. Because when kids are online, they are not the customers. They are the products. The social media companies make their money by selling advertising. And the more time they can and keep kids online, the more advertising they sell. So the fact that young people now are spending on an average of six to seven hours a day on social media is not an accident. It's not a coincidence. It is a consequence of the profit driven design decisions that these companies have made.

0:12:29 Thomas: To make sure lay people like me are keeping up. Is it analogous if I imagine a literal bulletin board? It'd be like, well, if somebody puts a flyer on that thing, you wouldn't like sue who built the bulletinboard necessarily for what's on it. But this is like, Well, they built the bulletin board and it's actually radioactive. And anyone who comes near it, it's like turns into a mutant or something. Is that kind of distinction?

0:12:54 Matt B.: Well, yeah. And beyond that, they're showing kids stuff that they're not looking for. They're They're feeding stuff to kids with the sole intention of triggering the maximum dopamine response that they can in their brain. And it's a fact of basic neurology that seeing the same thing over time, the dopamine response subsides. So they, by design, show kids increasingly extreme, dangerous, offensive, disturbing content because that's what triggers the dopamine response. It's a psychological fact that material that is discordant to us, difficult to watch, unpleasant, disturbing, triggers a bigger dopamine hit in our brain than material that it is affirming. And so by design, these platforms, for purposes of maintaining and increasing engagement and profitability, show kids increasingly more extreme, more difficult, and more damaging material.

0:13:53 Matt: Just to be clear that they're doing that to all of us, and we know that the algorithms are kind of designed to enrage us or make us feel other strong emotions. I totally agree with what you're saying. Your suits have focused on children, though, exclusively. Is that right?

0:14:03 Thomas: Hmm.

0:14:04 Matt B.: They have. And, you know, there clearly are the same algorithms apply and are used for adults. But first of all, their brains are mature. They're presumably more emotionally mature and less needy of social acclimation. Like many potentially dangerous activities or conduct, like driving a car or drinking alcohol or voting or owning a firearm, we age -gate things based on the premise that young people are not as equipped, either emotionally or neurologically, to make good decisions.

0:14:36 Matt: I want to go back to Thomas's metaphor for a second, because I think it is helpful to think about this. And I want be sure that I understand the on before we kind of get into the rest of this. We're talking about a bulletin board, a literal physical bulletinboard, a community center, and somebody just started posting murder for hire on the bulletin Board, right? Just saying I could kill somebody for you.

0:14:50 Thomas: It was me. I'm sorry. I was, I'll stop.

0:14:53 Matt: Obviously the community center would be liable at some point if they knew that somebody is posting, you know, drug sales or murder for hire or, you know, child sex trafficking or something. And section two 30 kind of contemplates this as well, that it doesn't excuse them for moderation. It sort of creates a thing that you don't get punished. If you don'T moderate, you DON'T get punished if you do moderate. Did I get that right?

0:15:11 Matt B.: Well, the intention of Section 230 was to encourage companies to monitor, particularly as it pertained to child obscenity. You know, if they fell short, if THEY didn't get it quite right, they would not be subject to liability for trying. And that's really separate and distinct from the bulletin board analogy that you describe. I would kind of expand upon your analogy and suggest that one of the differences is that, you know, you have murder to hire posted on the bulletin board, but every time someone comes to look at your bulletinboard, you get paid.

0:15:45 Matt B.: You essentially have the incentive to put stuff on the bulletin board that's going to bring people there to look it.

0:15:45 Thomas: Who's

0:15:56 Ad: [Ad break 2 — ad music]

0:15:56 Matt: For the purpose of these suits and what we're talking about here, how do you define addiction?

0:16:00 Matt B.: Well, there's a very clear definition of social media addiction. There's a six -part test that's used in the medical literature. It's skipping my mind. It's a city in Norway. Way. And they're similar to the factors that are used to analyze other forms of addiction, which is increasing levels of engagement with the substance, attempts to quit without success, that the use of the substance or the social media interferes with the successful kind of completion of life tasks and maintenance of personal relationships, and that too much is never enough.

0:16:36 Matt: We are a law podcast, not a science podcast. I don't want to make you go too deep on the science here, and I don't wanna talk about it too much, but I just wanna make sure I understand because this is all completely new. Obviously, social media has only been around effectively for maybe nine or 10 years. In its current form, it's only been a few years in terms of the turbocharged algorithms we're talking about. Out. Would you agree that the science isn't completely on this, on social media addiction?

0:16:57 Matt B.: The nature of science is it's never complete.

0:17:00 Matt: Sure. Naturally.

0:17:00 Matt B.: I mean, science is an ongoing dialogue. Certainly in the time that I have been doing this work, the Bergen social media addiction scale is widely used within the academic community. The findings at North Carolina neurologic program and their psychology department is finding actual physiologic neurologic changes in adolescent brains based on frequency of social media use that triggers the same part of the amygdala that drugs or alcohol does. I just don't believe that there's really much question that social media is addictive. Indeed, during our trial, Adam Masseri admitted it.

0:17:37 Matt B.: Adam Massieri, the head of Instagram, admitted under oath during the trial that social medial could be addictive.

0:17:43 Thomas: That? Hmm. Real Matlock moment. Jeez. That sounds dramatic.

0:17:46 Matt: Right. Totally. Yeah. That's why I'm so fascinated by this whole thing, because this is completely new. It's emerging. We're trying to find the parameters and the contours of it here. We can put it in the context, and I've seen it put in the context of things like cigarette litigation that took down big tobacco, or at least severely restricted cigarettes in the United States, and slot machines, which is of particular interest to me. I did some work on this about 15 years ago, trying to keep a casino out of my neighborhood, hood. And I learned a lot about how these things are designed to be addictive. And I think that slot machines are arguable. I think that's a better analogy just because they have admitted that themselves. I know from the discovery you've gotten that they've talked about the slot machine model, those little hits that kind of keep you going.

0:18:23 Matt: The difference, of course, is that slot machines' algorithm is pretty much whether or not you get a payout. And the algorithm here is whether you see a Nazi rant about how Jews are controlling the media or makeup tips.

0:18:34 Matt B.: The slot machine analogy is also directly the intermittent reward cycle is expressly designed into the social media functionality, most significantly with the like features.

0:18:46 Thomas: Yeah, Nazis are wild too. So if you get through any of those, it triggers the payout, the wild card. Yeah it's

0:18:51 Matt: You get the bonus. Yeah, you get bonus content. But I was wondering about that because the slot machines, it's fairly easy to regulate that algorithm. And all the states that have slot machines as I understand it, just from my own research on this, they strictly regulate how that algorithm works and how much they can take advantage of people, which is wild to me that that's built into the law.

0:19:09 Thomas: Also wild to me that people still play them knowing that it's like a mathematical formula of you're going to lose on average this exact amount. Out yeah a

0:19:18 Matt: Yeah. And we kept sitting here long enough that we're going to keep taking it from you, right? That's part of the design. Maybe you know her work, but Natasha Dalschul at MIT wrote a fantastic book called Addiction by Design. I thought a lot about it when I was doing the research for this and looking at your work. I certainly see the parallels, but you can see the differences too. I mean, it's certainly a lot more complicated, but slot machines are designed to be deceptive. Is that an argument you're making here? How so?

0:19:36 Matt B.: It's a similar psychological dynamic that is at play. Yes.

0:19:44 Matt B.: Social media, the executives have testified to Congress that they're not addictive. They have repeatedly and falsely represented to parents that their platforms are safe. They Have not posted warnings. I mean, one of the jury's findings was not only was Meta and YouTube responsible for designing a unsafe product, they were also separately responsible for failing to warn about it, which is a basic principle of law or just a basic human decency is if you know someone's going to go down a rabbit hole and you have the ability to warn them about it, you've got an obligation to give them a fair appreciation or a sane appreciation of the risk that they are undertaking.

0:20:27 Matt: And that's been a basis of every one of your suits, at least the ones I've looked at, has been failure to warn as well. And what would that look like if there were more of a warning built in.

0:20:34 Matt B.: Well, I think it'd be a couple things. One, parents would be provided with accurate information regarding the risks that children have when they engage online, particularly with predatory adults. Children, when they would would be trained to use a social media app, would be provided with warnings and trainings. This is just one example, but how to identify predators. One thing that these online predators have in common is they use a very similar grooming technique. If young people were simply provided with some examples, many of them could hopefully save themselves from these risks. So that's just, you know, one example, but there are many where providing a honest appreciation of the risks that social media poses to young people is a good way to provide parents with the tools they need to protect them from harm.

0:20:35 Thomas: Little notice when you log in like may cause genocide in Myanmar or something you know just

0:21:34 Matt B.: Yeah, it really is a overtly false representation that's made regarding the risksthat people, particularly young people, have on social media.

0:21:45 Matt: I can hear listeners thinking as we're talking here that there's a lot of products like this. I mean, Doritos are certainly meant to be designed that you eat as many as possible without realizing it, right? Like Netflix makes, every streaming platform at this point takes you on to the next show before you finish the credits on the ones you just watched, right? Just to keep you going.

0:22:02 Thomas: I wish they'd stop that

0:22:03 Matt: Yeah that's pretty annoying. It works on me. There's also the video games, right, that are designed to keep going and have the same kind of slow burn dopamine effects. Sex. As far as I know, you haven't brought any suits against video games. Do you see that differently?

0:22:14 Matt B.: I haven't. Other people have. Every case is different. There's levels of degrees. When you have internal meta and YouTube documents explicitly expressing a desire and an intention to addict people, that's different than, I think, Doritos. Just because something is habit -forming and just because people might do more of something than they ought to does not give rise to a cause of action. Some people drink too much, and we're not suing Johnny Walker.

0:22:45 Thomas: Yeah I actually feel like you know with those other things too like that's what they're plainly doing right I mean we know people make video games are trying to make the best video game we don't necessarily know back to duty to warn kind of thing like we didn't necessarily know that social media and these other products are designed to be so addictive restrictive necessarily. Like that wasn't really part of the open goal of what they were doing, right?

0:23:08 Matt B.: It certainly was not the stated goal. They certainly don't say when you go online, you know, this is going to cause addiction and you're going to be confronted as a child with sexual perversity from predators. Most people know that if you have a couple scotches, you're not safe to drive. I think that there's a level of deception here that is significant. I think of significance as well is that the day before our jury verdict, in our case, in KGM, a jury in New Mexico had expressly found that in a case brought by the New Mexico Attorney General, Meta had deliberately misrepresented the risk of sex abuse that young people have as a result of their platform.

0:23:55 Matt: Yeah, I guess I want to go back to what you're saying because I want to make sure we cover this. What you just said about how there isn't necessarily a cause of action just because you have a bad habit or you pick up a bad habitat or you start doing something too much. And obviously, as Thomas was saying, cigarettes, alcohol, I mean, the states have built gambling into the government, right, that we now regulate gambling and allow it pretty much in every state now. And we know that there's a certain percentage of people that will not be be able to stop gambling to the point that they have to tell the casinos not to let them in, right? That's a terrible model. What is the difference then, I guess, between something that is habit forming or could be a bad habit and something that rises to the level of the kind of addiction you're talking about?

0:24:31 Matt B.: Well, the first and foremost difference is, is it targeted toward vulnerable kids or mature adults? As we all remember from the tobacco litigation, one of the most compelling pieces of evidence was the Brown and Williamson Project 16 documents that focused on marketing cigarettes to young people through like Joe Camel and other images, knowing that if they became smokers in their teens, they would be more likely to be lifetime smokers. That was, I think, a very strong argument that countered the narrative that the tobacco companies articulated at trial was that everyone knows tobacco is addictive. You're responsible for it yourself. I'm not saying there are never circumstances where a social media company would be liable for activity or directed toward an adult, but I think it's a much, much higher standard. We have a much higher expectation on what we would hold an adult responsible for than a child.

0:25:35 Matt: Not to change too much, but I'm very interested in this question about, you've also, I believe, either brought suits or have been doing work on AI chatbots and sort of the delusions that they can give people or the things that they can end up making them do or self -harm or harming others. We're talking more about adults in those cases. I'd like to hear a little bit more about that too.

0:25:51 Matt B.: I would have thought after a year and a half of representing victims of social media, I couldn't be shocked by anything. But when I encountered these AI chatbots, which didn't really exist at the time I started doing my work just a few years ago, I was positively shocked and appalled at the, in particular, the suicidal material that was being force-fed to young people.

0:26:17 Matt B.: It is a new frontier. I think with respect to adults, we're seeing a significant number of cases involving individuals who develop psychosis -like symptoms far after the age when a person would organically be expected to have a psychotic break. These situations are colloquially known as AI psychosis, but in actuality are called AI -induced delusional behavior. Through a number of these chatbots, it is a growing phenomenon of individuals developing emotional connections with these randomized, probabilistically driven chatbats, in many cases manifesting very severe and in other cases life -threatening mental health harms.

0:27:04 Matt: To make light of this, but we've seen it give lawyers the delusion that cases exist, right? And that it certainly across the board has had a lot of problems. That's a product liability issue to me. That seems very clear that you have a chatbot that is completely within the control of OpenAI or whoever's making it, where they can control the parameters and contours of what the chatbot is doing, and they can put safeguards or remove safeguids or allow it to hallucinate to whatever degree. They certainly could have much much earlier, very strictly limited how much it talks about self -harm and not allowed it to ever suggest to somebody that that's the way to go. But that is different from an algorithm that is feeding you somebody else's content.

0:27:38 Matt B.: Very much so. And that's why I think some of the defenses that the social media companies have, until recently, deployed so successfully regarding immunity are unlikely to be as availing to these chatbot companies as they confront cases brought by individuals who are severely harmed or in some cases die as a result of the intentional programming decisions that they make.

0:28:02 Matt: Because there's no Section 230 protection for them, right?

0:28:04 Matt B.: We don't think so.

0:28:05 Matt: Yeah, I wouldn't think so. I would hope not.

0:28:07 Thomas: Be like, your honor, we stole all this from people though. So it's not ours. Right.

0:28:10 Matt: Yeah, right. Right. Yeah. It's all recycled slop. It gets on my main practice area, but I've always been interested in internet law and sort of how we regulate online speech. I was born in 1980. I kind of grew up with the internet. It. And I remember the debates over the Communications Decency Act when the Supreme Court struck down most of it and sort of Section 230, as you said, going forward from the classic BBS model to now regulating these algorithms, which are a much different game. I've seen a lot of criticism, a lot of pushback about the idea that these algorithms are in a sense content because they're like a The New York Times deciding what to put on the front page versus a forum where people are just posting on Reddit or whatever, and there's no moderation except for those those individual forms. And so there is that question of, well, isn't the design of the algorithm itself a form of content? And aren't you just doing an end run around section 230?

0:28:57 Matt B.: Well, in many cases, we believe, and the answer is yes, and courts have agreed with us. In other cases, the answer's no. You see this in Moody v. NetChoice, the Supreme Court really grappling with what is human human-manifested editorial decisions, what is machine-driven machine learning. The fact that an algorithm is involved does not inherently remove Section 230 protections from a company, but the question is, what is the algorithm being deployed to do? If the algorithm is being deployed for the purpose of enhancing engagement over all ends without any concern for the online experience, then arguably that is a very clear situation where Section 230 does not apply. If the algorithm is deployed to help the platform exercise traditional editorial decisions, then, you know, probably Section 230 is still there. And so bad content moderation, at least until the Supreme Court weighs in, is not a claim that can survive Section 230.

0:29:08 Matt: Got it. Okay. And I was thinking, as you're saying that about Lemon, the ninth circuit case you mentioned, because that's such a clear cut. Actually, I'll let you say what the issue was in Lemon, but it was so ridiculous to me that that even existed.

0:30:13 Matt B.: Lemon was a situation where Snapchat had something called a speed filter associated with its platform where you would send a snap to somebody else and it would say how, you know, through GPS would be able to say how fast you were going. Well, sadly, in the hands of teenage boys who are recently behind the wheel, that gave rise to a phenomenon where they would be snapping when they wouldbe going 100, 110 miles an hour.

0:30:38 Thomas: Oh, wow. Jeez. Wow, I

0:30:38 Matt: And how would that not happen? I mean, that's just inevitable, right? I mean it's just crazy.

0:30:42 Matt B.: Boys will be boys. Tragically, there was a fatal car accident where a minute before the collision, one of the kids had used Snapchat and, you know, it said, you knows, snap to somebody else said, I'm going 100 and I think it was 120 miles an hour. Snapchat objected to that claim, said it was because the case gave rise or arose out of a third party communication, which was one of these tragic young men sending a snap to the other that that was third party content.

0:31:11 Matt B.: The Ninth Circuit said no, that it was the design of the speed filter Snapchat knew or should have known would give rise to dangerous behavior, not the third party content. And so I think that was a very important focus that the court made. Section 230 is not geared toward whether or not content is somehow involved in the chain of causation, because it usually is, but rather whether the plaintiff's cause of action seeks to hold the company liable for its publishing activity or the violation of some other duty. And in Lemon, the court held that that duty was the duty to design a safe product.

0:31:51 Matt: I just can't believe lawyers even would sign off on the speed filter. I mean they're just so obviously going to create the problems it created.

0:31:56 Matt B.: Well, lawyers didn't sign off.

0:31:57 Matt: Oh, they didn't? Really?

0:31:57 Matt B.: I mean, that's the whole problem with Section 230 is that it engendered a attitude, you know, the move fast and break things attitude of development that social media had. You know, every other company in America has a concern about how their products are going to be used. And if they know that there's a dangerous design that their product has, they could earn a little money, but would incur liability. They'll change their platform or they'll change the product. That basic economic exigency that every other company in America operates under, the duty of reasonable care, does not, or at least until recently did not, apply to social media. So in answer to your question, you know, the fact that somebody would be foolish enough to design a product, market a product to adolescent or young teenage boys with a speed filter in it is testimony, I think, to how the immunity that Section 230 conferred has created dangerously defective platforms that are savaging young people, not just in the United States, but around the world.

0:33:06 Matt: I thought the William's book, Careless People, I just thought there was a perfect title for that taken from The Great Gatsby. It really does seem to be so careless in so many different ways. As I understand it, and just I think as we understand social media, we have a situation where a 14 -year -old girl could be watching an Instagram video with a recipe or a cooking tip. And then within maybe a few days, suddenly you're seeing diet restriction videos. You're seeing maybe be pro eating disorder or disordered eating content, right? And then that's something that is so obviously, and that's the kind of thing that you're getting into at KGM, right, of what the algorithm is feeding you. And I can see the comparison to the speed filter, but I can also see the criticism about section 230 and about how this is just sort of moderating content that other people are creating. And it's not as dangerous as intentionally creating a product that is going to recklessly hurt people. The way that these two things are rubbing up against each other with 230 and with the products liability you're talking about is fascinating to to me. And I'm still trying to figure it out. I did not come into this interview with any particular opinion about it. I'm Still thinking it through myself, but how do you kind of address those, that conflict?

0:34:06 Matt B.: I don't think that enforced content moderation is appropriate under Section 230. To follow up on your analogy, if a 13 -year -old girl, for some reason or another, becomes interested in anorexia and starts searching for anorexic content online, that's a tragedy. Tragedy one would certainly hope that young woman would get counseling and help but I don't think that's on the company I think that that's something that section 230 isn't designed to protect on the other hand if that youngwoman is seeking exercise information and is sent down a rabbit hole with material encouraging her to hate the way her body looks to get by on 500 calories a day develop a a body image that is unhealthy and unsafe, I think that's totally different because that's showing her not what she wants to see, but what she can't look away from.

0:35:04 Matt: But what about the flip side of that? If you have a girl who maybe doesn't want, is too afraid or embarrassed to talk to anybody about this and wants to find out about how to deal with the feelings she's having of anorexia or bulimia and wants To try to get help on it. Do you think that this would maybe result in platforms having to restrict that kind of content too, because it would all fall under bulimia or anorexia.

0:35:24 Matt B.: I don't believe so at all. No. If someone is seeking information on it, whether it's in your case, in your most recent example, information that is helpful, even if someone is seek information that hurtful, that is separate and distinct from being fed stuff that you're not looking for. I do not believe that that imposing basic principles of reasonable care and safe design that have been woven in the web and woof of product liability law for 50 years is going to shut down the internet or squelch the free exchange of ideas. It's simply going to provide an opportunity for kids to see what they want to see and not be fed stuff that's harmful to them that they're not looking for.

0:36:08 Matt: And I guess this is one parallel I see to the tobacco litigation, even though I believe that tobacco and social media are fundamentally different products. But I think we do understand that what the AGs did around the 20th century, early 21st century with tobacco litigation fundamentally changed the industry, kind of broke it apart, made them completely change how they advertise and how they do business, right? And I think that there's a lot of potential for that to happen with your work. And I'm not at all opposed to that. I think many people would like to see big tech regulated more and to see some of this more harmful stuff reigned in, we should get into this as well. The AGs are in very similar fashion, bringing suits that are going to go forward this summer and very similar to yours. I think we can talk about the differences, but what is the kind of ideal outcome for you and the people on the side of things?

0:36:40 Ad: [Ad break 3 — ad music]

0:36:51 Matt B.: The ideal outcome is if I never have to talk to another parent whose child killed himself. The The ideal outcome is if social media shows kids what they want to see, not what they can't look away from. And what's so tragic is that all of these changes could be implemented with the flip of a switch. Most of them have been contemplated by and recommended to the senior executives of these social media companies by people of conscience within each corporate organization, but shot down whenever they would impinge against the profitability of the company.

0:37:25 Matt: Let's talk about those. What are we talking about here?

0:37:27 Matt B.: Turn off the algorithms. Show people what they wanna see, what they can't looks away from, don't feed kids the stuff that they're not looking for. Turn off the gamification, the infinite scroll, the likes, the streaks, the push notifications, these features that prey upon the fundamental need that adolescents have for acclimation of their peers. Provide real age and identity verification so that, first of all, kids are the age that they are to get online. And more importantly, that adults can't pose as kids or can't pose as safe people and use social media to prey on and abuse young people. And finally, a basic duty, a basic warning. Provide teenagers and their parents with a sane appreciation of the risk that social media entails so that they can, as a family, make the intelligent decisions of whether or not they're going to allow their kids to engage with social media and if so what kind of ground rules are going to apply.

0:38:27 Matt: What about data? I haven't really heard you talking about restricting how tech companies access and use our data, because that's a big part of the algorithm, right?

0:38:34 Matt B.: Yeah, they use their data quite a bit. That's been an ongoing effort for data privacy. Congress has been grappling with that. Senator Cantwell, in particular, the chair and now ranking member of the Senate Commerce Committee, has been working very hard across the aisle to try to get something there.

0:38:51 Matt: Because you can see how that would, maybe the one problem could help to solve the other, right, if we restricted how much, especially on kids, how much they can collect. I want to make sure that I represent the sides here because I know there's been a lot of pushback amongst civil libertarians and digital rights people on the idea of these kinds of age verifications and this kind of restriction. I know that the UK's attempt in the Online Safety Act has been heavily criticized because it's allowed censorship potentially that wasn't anticipated and barriers to entry that maybe people didn't want. Is there a model legislation that you support that would do things better or is there, what are your response generally to these kinds kinds of concerns.

0:39:23 Matt B.: I do not believe that turning off addictive algorithms, turning off streaks and likes and push notifications in any any way, shape, or form restricts the free exchange of ideas or requires a social media company to espouse any particular point of view. The suggestion that imposing the same duty of care on a social company that you impose on any other business somehow contravenes the First Amendment, I think is fatuous and misleading.

0:39:55 Matt: What about age verification though?

0:39:57 Matt B.: Why is age verification to engage in what is Is this dangerous activity any different for social media as it would be for alcohol? We have freedom of commerce or freedom of contract. You could make the argument that restricting alcohol sales to children violates their freedom of contract, obviously such an argument would be ridiculous. And I think similarly, age verification requirements for activities that we know to be dangerous are not only appropriate, but widely utilized throughout industry.

0:40:31 Matt: You don't have concerns about how that might spread over to adults or issues there.

0:40:34 Matt B.: No, because there's so many activities that we allow adults to engage in. There are activities that adults have a constitutional right to engage in that we don't extend to kids. So I don't really buy this slippery slope analogy. The First Amendment had been in existence for 207 years when Section 230 was enacted and had done a pretty good job, and courts have done a pretty good of protecting free expression. I don't think that if companies are relieved of their privilege to invoke Section 230 to avoid legal responsibility, that the First Amendment is going to go by the wayside, or that in appropriate cases, courts will be disinclined to strike down efforts to restrict the free exchange of ideas. We certainly saw that in Net Choice versus Moody.

0:41:25 Matt: To be clear, I think I'm with you on a lot of this, that I think very much like with tobacco, that the only way a lot of it is going to be reined in is through litigation. I think there's certainly a lot of place for it here. I'm just trying to understand sort of where the contours are. Any other closing thoughts then? Great.

0:41:37 Matt B.: Well, I just think kind of going along with what you said, we view the importance of the civil justice system as fulfilling its economic role of product liability, which is internalizing the cost of safety. Every other company has to consider the potential of product liability when they decide whether to design a dangerous or a safe product, or when they decide whether they implement a safety measure that will make their product somewhat less profitable but a lot more safe. We think that the jury verdict last month, as well as others hopefully to come, will kind of instill that basic economic rationality on social media companies that every other company in America and indeed throughout the world operates under. We consider ourselves pro-capitalist, we consider ourselves pro free market. And we think that by moving forward with these cases, we can continue to have the marketplace operate as it was intended to do so.

0:42:34 Matt: Well, thank you so much for your time. Really, this has been really interesting.

0:42:37 Thomas: Really appreciate it.

0:42:37 Matt B.: Thank you very much.