Meta Settled. Is it a "Reckoning?"

0:00This episode is brought to you by Capital One. Capital One's tech team isn't just talking about multi-agentic AI. They already deployed one. It's called Chat Concierge and it's simplifying car shopping. Using self-reflection and layered reasoning with live API checks, it doesn't just help buyers find a car they love. It helps schedule a test drive, get pre-approved for financing, and estimate trade in value. Advanced, intuitive, and deployed. That's how they stack. That's technology at Capital One.

0:32Stripe is built for whatever comes next, because as your business scales, so does the complexity of fraud. New customers, new markets, new AI-driven attack patterns. Stripe Radar protects your business as it scales with the scale of the Stripe network. The more payments Stripe sees, the smarter its AI gets at detecting those patterns as they emerge. Stripe's network is so vast, it has seen$1.9 trillion in payment volume across millions of global businesses. In fact, when a card is used, there's a 92 % chance the Stripe network has seen it before. All of this means fraudulent patterns across payments, accounts, and customer abuse can be detected, flagged, or blocked easier.

1:16Based on the rules you set up, of course. blocking bad actors, welcoming good customers, keeping more of your revenue, your revenue. Learn more about fighting fraud with Stripe at stripe.com slash radar.

1:36Paresh Dave: Have you seen Meta's ads that they've been putting out in big newspapers across the country? Correct. Urging TikTok and YouTube to agree to some of the same conditions that they have. Paresh Tve covered Meta's big social media addiction trial for Wired. That case just settled for roughly$17 billion. Meta agreed to pay a consortium of some 47 states over the next decade. Now, Meta's ads are urging its competitors to, quote, join us in supporting teams, which is company speak for please also settle. And the idea is that if everyone's on board, then teenagers don't have the option to go, say, jump from Instagram to TikTok when they hit their daily time limit on Instagram.

2:28If there's a sort of equal conditions everywhere, teens may actually go back to doing their homework, perhaps. God forbid. In the settlement, which did not admit liability, Meta agreed to some conditions like time limits on its apps. meta's argument is that it does not want to be at a competitive disadvantage to tiktok to youtube to snapchat um if teens are you know spending two hours on instagram then hitting their daily time limit and then unable to use instagram well then they're going to probably try to seek out another social media app perhaps they might go to tiktok and spend hours endlessly scrolling tiktok um in the end, that's bad for Metta's business because teens might just turn off Instagram use altogether if they feel like it's too restrictive and spend more and more time on TikTok.

3:22And so having these baseline protections, Metta argues, would sort of maintain competition in the social media industry, which is something that these regulators, attorneys general, have to balance. Not entirely unlike a parent trying to figure out screen time. They're balancing privacy. They're balancing free expression. They're balancing competition. And also trying to protect teens from the harmful effects of social media. Today on the show, meta settled. But is this really a social media reckoning? I'm Lizzie O 'Leary, and this is What Next TBD, a show about technology, power, and how the future will be determined.

4:05Stick around.

4:14Stripe is built for whatever comes next, because as your business scales, so does the complexity of fraud. New customers, new markets, new AI-driven attack patterns. Stripe Radar protects your business as it scales with the scale of the Stripe network. The more payments Stripe sees, the smarter its AI gets at detecting those patterns as they emerge. Stripe's network is so vast, it has seen$1.9 trillion in payment volume across millions of global businesses. In fact, when a card is used, there's a 92 % chance the Stripe network has seen it before. All of this means fraudulent patterns across payments, accounts, and customer abuse can be detected, flagged, or blocked easier.

4:58Based on the rules you set up, of course. blocking bad actors, welcoming good customers, keeping more of your revenue, your revenue. Learn more about fighting fraud with Stripe at stripe.com slash radar.

5:15This episode is brought to you by Capital One. Capital One's tech team isn't just talking about multi-agentic AI. They already deployed one. It's called Chat Concierge and it's simplifying car shopping. Using self-reflection and layered reasoning with live API checks, it doesn't just help buyers find a car they love. It helps schedule a test drive, get pre-approved for financing, and estimate trade in value. Advanced, intuitive, and deployed. That's how they stack. That's technology at Capital One. Hi, Brooke DeVard, host of Naked Beauty. Picture this, you walk in from the cold and every room already feels like fall.

5:52That's what whole home scenting means at Pira. One system working quietly in the background so your home always feels like the season and like you. Learn more at Pura.com.

6:09Paresh Dave: Let's take a step back a bit. So there is the settlement. It's a pretty historic deal. It required new safety features on Instagram and Facebook. Can you walk me through them? Yeah, so that main one is a two-hour daily time limit across Instagram and Facebook. Mandatory breaks if you're using Instagram or Facebook in one particular session for too long. Content filters, extreme makeover filters will no longer be allowed for teens. There are several parental supervision controls that will now be mandatory. So if you have a Finsta or kind of like a surreptitious backup Instagram account and you link it to our main Instagram account as a teenager, your parents will now be aware of that.

7:04Parents will also get periodic reports about how their children are using Instagram or Facebook. but in reality we know teens are mostly using Instagram, not so much Facebook. And so those are some of the mandatory things. And there are also some optional features, like teens will be able to have autoplay turned off on video so that they don't automatically turn on, and they'll have the option of a non-algorithmic feed to perhaps not be hooked as soon as they open the app. They might see a more boring piece of content and therefore not wanting to engage in that moment with Instagram.

7:45Paresh Dave: You know, some of these things are forced and some of them are opt-in, the non-algorithmic feed, the autoplay stuff. It strikes me that opt-in and turn off, like, you know, you have to do that. Teens are pretty good at ignoring that kind of stuff. Is there any penalty for meta if teens just don't opt-in? There is not. And that is the challenge here. So in the few days of court testimony that we had before Meta settled, one of the big things that came up is has these features, which many of these already existed for a long time. the adoption rate was like 1%, 2%, very, very minuscule. Even the awareness was super low.

8:30At one point, some of these features, just a third of teenagers in the U.S. were even aware that the features existed, which seems like, I don't know, mind-bogglingly low. And yeah, I mean, nothing about this necessarily means that teens can't also get around these features. by one, just asking for their parents' permission, some of these features can be toggled off, even if they're on by default. Two, to turn some of these, you know, ones that are not on by default on, you know, parents need to get involved. Will the parents get involved? At one point, just 1 % of parents had taken supervision power over their teenager's Instagram, according to some of the data that was revealed at trial.

9:19So yeah, there is a long way to go with some of this. And it is unclear whether that adoption will take up as expected.

9:28Paresh Dave: Paresh, one of the reasons I wanted to talk to you was that you were in the courtroom for this trial, but then this trial also seems to kind of be a vehicle for so many conversations about teens, social media, like not just the settlement itself, but these kind of larger, more existential questions about how we use social media. Are there any of these features that were agreed to in the settlement that you or your reporting deem as likely to actually be effective? Yeah, in reality, it's a mixed bag, right? The time limits are huge because there was evidence presented at trial from an academic psychologist that the studies over the last 15 or so years have pointed to the fact that higher time spent on social media is associated with an increased likelihood among adolescents of being depressed, having suicidal thoughts, these anxieties, these psychological issues.

10:30And so limiting the amount of time that adolescents, teens spend on social media could be huge, right? We'll have to see what the studies bear out over the coming years. But keeping them off social media at night, you know, during the school day, potentially positive. But at the same time, nothing in this settlement necessarily or sort of radically changes what the experience is like and what content these teenagers may encounter. So as part of this suite of features that Meta calls teen accounts, there are restrictions on the types of content that teenagers encounter. But we know that AI is only so good at content moderation.

11:17Things are likely to slip through still. And these companies have kind of severely cut their content moderation teams. Exactly. And either way, most of this being driven by AI, what teens experience and how those filters work. There should be less situations now where teens receive direct messages from creepy people, which is sort of a net positive. But again, this isn't foolproof. There could be things that slip through. So there's still a broader question of like, is Instagram usage among most teens still a good idea? Should they be banned altogether? And those questions still still kind of linger.

12:00Paresh Dave: One of the aspects of this that I think bears examining is how meta tends to age gate teens, right? Like verify that teens are teens. And this isn't just a meta issue. We have covered companies doing this kind of across the tech landscape. Um, couldn't you walk me through how they plan to do that and the questions that are being brought about, you know, using these systems for age verification? It's layers. So the first thing is they're hoping people sign up with their real birthdays. We know that doesn't always happen. And so the next thing. Has anyone ever met a teenager or an adult? I mean, you know, I don't put in my real age sometimes, too, on some of these websites.

12:45unless I believe there's some benefit to it. And so Meta's sort of layer of defense is that they use AI algorithms to try to guesstimate someone's age by looking at, well, if they're talking to all these other people who seem to be teenagers and they're not really using it that much during traditional school hours, well, you know, maybe they are probably a teenager. And if they want to prove that they are actually an adult and they shouldn't have all these restrictions on their account, well then they can upload a copy of their ID or they can take a video selfie that, you know, uses again algorithms to guesstimate their age based off of that video.

13:31And again, this is not a perfect science. Things are expected to fall through the cracks. What is maybe be positive is that there are some actual metrics that are built into the settlement that Meta has to hit in terms of accuracy for some of the tools that it uses for age assurance.

13:52Paresh Dave: Like what? Around 90 % accuracy, 85 % accuracy in different circumstances. One of the pieces of pushback around this and actually age verification more broadly is the idea that, you know, it is a privacy concern. And then I think the Electronic Frontier Foundation said this will enshrine Meta's harmful surveillance into law. What do they mean there? Well, there's this idea that in the early internet days, it was so easy to browse anonymously. You had sort of this freedom to be yourself online without being worried that you were suddenly going to be swarmed with all kinds of personalized, targeted advertising, that that information about what you're looking at could be used against you for law enforcement purposes.

14:43And now you kind of have to enter your name or your email address or your age everywhere you go. And this is, in effect, institutionalizing that. We know that some adults are going to get caught up in the age gating and will have to upload their IDs if they want a normal experience or traditional experience on Instagram. And we also know that sometimes teens do need anonymity. If they are in a situation where their parents don't support their sexuality, for example, they might want to find community online and they may want to do that in an anonymous way. And how can they do that and still have a good experience in this situation where they have to share more and more information about themselves, that could be a challenge.

15:38And they may not feel comfortable doing that. And they may face the same psychological issues that these laws are intended to prevent because they can't get on social media.

15:49Paresh Dave: Where does all this information go? The idea is that a lot of this information isn't stored. It's briefly tokenized and all that Meta knows is that you're in this age range or you're not in this age range. And that is the intention. But we know from data breaches in the past that implementation of these tools is not always ideal. And sometimes information is stored that wasn't intended to be stored or stored incorrectly. Encryption fails in some, you know, unfortunate way and data can get out there, which is why you see these activist organizations is really concerned. Yeah. There's also this aspect that I don't totally understand.

16:37Paresh Dave: I was wondering if you could explain to me why Meta is pushing app stores to also get involved in the sort of facial ID part of this. This has been a long running dispute, especially between Meta and Apple, which are, you know, effective frenemies. Mark Zuckerberg has never been fond of the way Apple runs its App Store in addition to other issues. Walled garden. Exactly. So what Meta wants to happen is more regulations on Apple and Google because it sort of sees, well, if we can get one regulation, maybe if we can get more regulations, we can bring down App Store taxes and fees and, you know, a whole host of other things.

17:22But this would be a crack in the door. And what Meta wants to see happen is right now, app stores collect some data to verify age. Can we require app stores to be more stringent in their collection of age data? And then they, the app stores, can just pass that on to apps in some sort of safe and privacy preserving manner, ideally, and therefore there wouldn't be the onus on users to upload their age data to thousands of apps or hundreds of apps or dozens of apps or a couple of apps, whatever they're using. Instead, they just upload their age maybe once to Windows, maybe once to Mac OS, once to the App Store, once to the Google Play Store, and that is a safer, better experience for users.

18:17Paresh Dave: Let's talk about the money here. So the number seems to kind of slide around, right? Somewhere in the$16.7 to$18 billion range. On the one hand, that is real money. On the other, that's what, like a third of what Meta makes in a quarter? That sounds about right. It is not an insignificant amount of money for the state. And it's over 10 years, we should say. Correct. And the states will be able to put it towards mental health counseling for teenagers, towards, you know, public service announcements to teach people about the dangers of social media. You know, a lot of beneficial uses potentially for the money for states.

19:04For Meta, I mean, it's not ideal to have this money go out the door at a time that they are trying to invest every little bit of cash that they have into building AI data centers. So, you know, it's not the greatest for Meta, but it is not something that is going to be painful for them. And that is why you see a state like Florida refusing to settle so far because they say they really want to put the screws to Meta and make it financially hurt.

19:31Paresh Dave: So this is an interesting question. I listened to an interview with Colorado's attorney general where he was basically saying, like, no, if we had gone to trial, we might be caught up in appeal after appeal after appeal. And none of this might ever happen. We're willing to just take the money and do what we can now. But if you are a Florida, right, what's what's your hope here that some big decision really hurts Metta financially or kneecaps their business? What's the strategy there? Yeah, well, it's kind of the best of both worlds because you know Meta is going to roll out these upgraded features or, you know, these revised features across the U.S.

20:10It's not like Meta is going to deny teens in Florida some of the things that are discussed in the settlement. And so if you're Florida, you have it, you know, you're not delaying teenagers getting access to these potentially vital features. and at the same time you can continue to pressure Meta to do even more. And it's worth noting it's not just them. There are school districts that have brought similar allegations against Meta. Those cases are ongoing. There's families and individuals who have brought lawsuits against Meta. Those cases are ongoing. So there's a potential to squeeze even more out of Meta, though the company, I'm sure, is going to try to stick to the terms of this settlement in any future settlement as much as possible.

21:03This episode is brought to you by Capital One. Capital One's tech team isn't just talking about multi-agentic AI. They already deployed one. It's called Chat Concierge and it's simplifying car shopping. Using self-reflection and layered reasoning with live API checks, it doesn't just help buyers find a car they love. It helps schedule a test drive, get pre-approved for financing, and estimate trade in value. Advanced, intuitive, and deployed. That's how they stack. That's technology at Capital One. State fair food is getting smaller. And pricier. We'll have room for fried cookies. Okay, you never overspend on food.

21:40What's your deal? What's his deal? He went to Metro and got$450 in value on the new Samsung Galaxy A37 5G for free. And no activation fees. Free phone? Oh, sign me up. Get that more for your money feeling. only on Metro by T-Mobile. Bring your number and activate a$40 period plan. $450 based on device discount waived activation fee. Not available if with Metro or T-Mobile in the past 180 days.

22:05Paresh Dave: Part of this settlement deal is that Meta will only pay like$12.7 billion over 10 years. And the rest, to go back to where we started, is conditional on TikTok and YouTube also doing this. Like, why would they do that? And Snap. Yeah. Why the other platforms would do this is pretty simple. The same reason Meta settled is, do you want to be stuck in years of litigation, years of uncertainty? do you want to deal with years of meta running I don't know maybe Mac versus PC style ads where they're saying Instagram has all these wonderful protections for teens did you know TikTok refuses to do these same things like imagine dealing with that for the next few years it wouldn't be great for TikTok's brand and reputation especially at a time that that's trying to say that it's a wholesome American company now, you know, same case for Snapchat and YouTube.

23:10There are arguments to be made that each of those experiences is vastly different than Instagram, and that, you know, therefore they should not be subject to some of these same limitations. You know, we know that teenagers often use YouTube to help with their homework, to understand, like, lessons that they're learning in school, or they might be even mandated to use YouTube in certain respects for their schoolwork. So how they negotiate around some of that is to be seen, but you can see why they would be quick maybe to agree to these same terms.

23:47Paresh Dave: So I guess I wonder, like, let's say, okay, they agree, right? They put some money into this. We are in this huge moment where I think there's a reckoning with the harms of social media, but also just technology and trying to figure out what to do. Nobody seems to have a great idea, right? Some school districts have banned phones. You have school districts making similar AI bans. What do you think social media might look like for kids and teens in a decade? Is it going to be noticeably different? Well, one thing that this settlement doesn't really contemplate is the world of AI chatbots that we are swiftly moving into right now.

24:30And that is often the criticism of lawsuits and courts and sort of the judicial system. We're fighting the last war. Exactly. That social media was so yesterday's problem. The problem now is all the time that teenagers are spending on chatbots, asking for health advice, asking for homework advice, asking for relationship advice. is that the best source for them? You can argue that it is a resource that is available to all, and for people that may not have other options, it's a great tool of kind of last resort. But we know that people are relying more on those tools, and those tools could get more social over time.

25:16Paresh Dave: I have interviewed parents whose children have killed themselves after spending time on chatbots. Exactly. Which is why, you know, it's noticeable or at least, you know, very, you know, glaring that, you know, nothing in this settlement addresses Meta's AI tool called Meta AI. So I do wonder if social media somehow looks more like our current AI world that we're entering into and how awful that could potentially be for teenagers if they're spending a significant time talking to these synthetic personas, if they're talking to computers, basically, what that will mean for their interpersonal skills, what that will mean for their mental health.

26:04And are we doing enough right now to get ahead of that problem?

26:09Paresh Dave: Last year, Australia tried to ban social media for kids under 16, did ban it. But there's a new study from their internet regulator that showed that more than 8 in 10 young Australians still use social media. So I wonder what that tells you when we look at this world of like, how do we figure this out? Which tools are effective? Well, it's a couple of things. We know that in Australia that they were not as stringent perhaps as this settlement in specifying that the age assurance technologies or the age verification technologies have to work really well. And you see now that Australia's regulator is asking for more enforcement powers, more fining powers to go after platforms that are not doing enough in terms of age verification.

27:00So the settlement here in the US suggests that maybe things will come out a little bit better because of that. But then there are two other things. We know that teens are determined they will find a way to get around some of these restrictions. And that's what we're seeing in Australia. You know, teens doing these video selfies, but kind of covering their face and obscuring their face or doing funny things. And they're able to get through the age verification technologies. So there's that. And then I would say the third thing that maybe is different about the Australian experience is that there's exemptions for content that is public and teens can continue to access that.

27:43And again, that goes to maybe like the YouTube homework problem. And how do we deal with that? where there are certain content that maybe kids do need to see or experience. You know, if Trump posts something important to Facebook, should teens be able to see that? You know, perhaps there should be a way around that, even if they've hit their two hour time limit.

28:08Paresh Dave: In thinking through this settlement and what it pretends to the future, a lot of people have made the comparison between social media and the big tobacco settlement from. 1998. That was the master settlement agreement between the big companies in several states. I was really interested in this New York Times piece, interviewed a woman who'd written a book all about the political history of the cigarettes. And she said, the meta of 2026 is better off than the big tobacco of 1998. And that like by 98, the anti-smoking campaigns had really done a lot. And it does make me wonder, like, oh, are we going to be back here in 10 years with some multi-billion dollar settlement about AI?

28:57Paresh Dave: Or we've just reached the point where there isn't enough of a groundswell yet to make a big difference? I don't know. There are certainly a lot of parents out there who are concerned and, you know, increasing number of activist organizations who are trying to tackle this issue by putting pressure on companies, by putting pressure on schools, by putting pressure on parents, you know, fellow parents to supervise their teenagers' social media use. At the same time, except for Facebook, which teenagers aren't using kind of because it's boring. It's for old people. Yeah. Yeah, there continues to be high numbers of, you know, users in that adolescent demographic for these social media platforms.

29:44And there's always, it seems every year, there's some cool new platform that all the high schoolers are using, whether it was the anonymous apps of a few years ago to these like homework helping AI agents that you've seen pop up in the last couple of years. There's always something new. They're always engaging with it. They are on their phones. It does seem like we will end up in a situation where if it's not meta, there will be another company that we're talking about in five to 10 years in the same situation. Because a lot of people from from Facebook and meta and Instagram who are responsible for overseeing some of these features that we're now addressing in this settlement.

30:30Well, they've gone on to work at places like OpenAI and Anthropic. they're going to repeat the same thinking unless something changes about our culture and how we design these features. And as I spoke to activists last week, including some that testified during the trial, like one thing that was evident is that nothing because of this settlement changes Metta's culture. You know, Metta's culture is going to stay the same. Their incentives are still the same. They are still a advertising company that is driven by trying to sell more ads and sell them at higher prices by engaging people and making people immersed in that experience.

31:13And even if teenagers are restricted, the goal is still to make money off teenagers and keep teenagers in this meta universe for as long as possible.

31:23Paresh Dave: Piresh Deve, thank you so much for talking with me. Of course. paresh dave is a senior writer for wired where he covers the big tech companies and that's our show for today what next tbd is produced by evan campbell madeline thames ducharm and patrick ford page osborne is the senior supervising producer for what next and what next tbd mia lobel is the executive producer of podcasts here at slate and ben richmond is the senior director of podcast operations i am lizzie o 'leary you can go yell at me on blue sky i'm at lizzie Really. Thank you so much for joining. Talk to you soon.

32:05This episode is brought to you by Capital One. Capital One's tech team isn't just talking about multi-agentic AI. They already deployed one. It's called Chat Concierge and it's simplifying car shopping. Using self-reflection and layered reasoning with live API checks, it doesn't just help buyers find a car they love. It helps schedule a test drive, get pre-approved for financing, and estimate trade in value. Advanced, intuitive, and deployed. That's how they stack. That's technology at Capital One. State fair food is getting smaller. And pricier. We'll have room for fried cookies. Okay, you never overspend on food.

32:42What's your deal? What's his deal? He went to Metro and got$450 in value on the new Samsung Galaxy A37 5G for free. And no activation fees. Free phone? Oh, sign me up. Get that more for your money feeling. Only on Metro by T-Mobile. Bring your number and activate$40 period plan. $450 based on device discount waived activation fee. Not available if with Metro or T-Mobile in the past 180 days.