Latest / Computer Says Maybe / The Kids Are Online: Are Social Media Bans the Answer? w/ Cori Crider and Pamela Wisniewski

Transcript

The Kids Are Online: Are Social Media Bans the Answer? w/ Cori Crider and Pamela Wisniewski · Computer Says Maybe

• Hey there, I'm Alix Dunn, and welcome back to Computer Says Maybe, where I talk to the people pushing back on tech power and fighting for something better. For the last month, we've been talking to lawyers, researchers, policymakers, and parents about the ways that technology is failing kids. We've been focusing on what that failure looks like in schools, but the issue of kids and online safety extends obviously far beyond the classroom. So how do we make technology and the internet a safer place for all of us, adults and kids alike? Trying to answer that question is like opening a very complex can of worms. So in this episode, I'm talking to two guests, Pamela Wisniewski, a researcher focused on helping kids and parents navigate the risks of online and digital tools while maintaining good relationships with each other, and Cory Kridler, executive director of the Future of Technology Institute. Cory is focused on fighting for safer social media design frameworks and a smarter social media ban. And if her name sounds familiar, it's because she is a really good friend who I mention sometimes on the show. She's also been on the show before, and she's been working in this space, broadly, digital rights policymaking, legal frontiers, for a very long time, originally in the national security and human rights space, but now more recently in the policymaking space as it relates to these core issues that we've just got to get right in the next few years, particularly in Europe. So it's really great to have her back on to talk about this overwhelming policy challenge that we're facing that probably has no perfect answer. So this is the final episode in our back to school series, exploring the tech industry's impact on kids and education. But if you want to keep the conversation going, we would love to have you join us in the Maybe Collective. It's a space where you can connect with other people who are thinking about and working on these issues. You can join events with guests from the podcast, and you can dig deeper into tech politics conversations we're having on the show every week. We will drop a link to join in the show notes, so do check that out, and I hope to see you there. Now let's get into my conversation with Pam My name is Pam Lisiecki. I'm a principal research scientist at the International Computer Science Institute, or ICSI. It's an independent research nonprofit, uh, affiliated with UC Berkeley. Uh, prior to this position, I was a tenured associate professor at Vanderbilt University and the University of Central Florida. My research is in human computer interaction, and I do research at the intersection of social computing, so how we connect with others online, and then privacy and online safety, the unintended consequences of when we do. Um, and within that space, most of my research is centered on adolescent online safety and risk, uh, taking a resilience-based approach to helping teens navigate online risks so that they can benefit meaningfully from online participation. There's obviously huge policy push to change the way that platforms work vis-a-vis kids, like age gating and age verification. Um, do you wanna share a little bit about what problem you think that's trying to solve and kind of how you think about that sort of solution set? Sure. Um, I'll, I'll s- I'll start with, you know, I was vocal when there was a ruling saying that, uh, age gating was required for pornographic, uh, websites. Um, and I was a proponent of that because the idea there is legally children should not be have, have access to those platforms, and the age verification mechanism was making sure that the person was an adult. And under that it made sense, but I also, when I posted about it, uh, I also articulated that I was worried that this was going to be a slippery slope, meaning that once age verification, age gating mechanisms were in place for pornography and websites that are targeted towards adults, that it would then become required for social media. Lo and behold, you know, less than a year later, that's what we're starting to see from regulators, is that they're moving towards a structure of trying to do age verification on all social media platforms. Um, this is rolling out slowly state by state, um, and that's making it hard for the platforms to comply for one thing. My bigger problem with age verification are the privacy implications. It means that we no longer have the right to remain anonymous online because we have to provide these age indicators to platforms so that they can verify age to grant access. And how people are doing age verification varies widely. But many of them are using facial recognition to predict or estimate age. Some are doing neutral age verification by just, you know, putting in your birthdate on a site, to others requiring personal ident- identification, but you know, kids don't have a driver's license like you and me. So the concern is now all of these companies or the third-party providers that they're outsourcing age verification to now have a store of data about us and our children that laws like the Child Online Privacy Protection Act were trying to avoid having. Um, and so I think that there's a little bit of a paradox of like we're saying that we wanna protect children online, but then we're giving up their privacy and doing facial scans to do age verification. And then my question is, to what end? So is it just to age gate them so that they can't get on platforms where they can connect and, you know, get support from their friends? Or are there other mechanisms that, you know, can the government say that the difference between a 13 and 14-year-old is X to where you can't get on a platform. I actually have experienced this with my own daughter, where she wanted to be online to get crochet patterns. And so I created a duplicate account for myself on Facebook, which is a platform that's so uncool with kids that nobody uses anyway, so that she could get crochet patterns. And then I was like, "Okay, well, let's create a teen account for you because that would be even safer than just having this account that's, you know, a duplicated account for me that I check all the time." And when I created the teen account for her, it immediately blocked her account because it did a facial scan saying that she wasn't 13, which she's not 13, she's 11. And so in some ways, it, it worked against what I was trying to do as a parent, is having some of the safety mechanisms in place so that she could do something that was positive and beneficial. Like, she doesn't have any friends on there, she just wanted crochet patterns. Another example is, um, she was using a VR headset to be able to play I Am Cat, and it also identified her as a child and tried to kick her off the platform. So my husband had to go in and do age verification as a, you know, 49-year-old man just so that she can play I Am Cat. And so in many ways, age verification is taking the rights of parents away and moving it to regulators and platforms on saying what is appropriate and what is not appropriate for my kid to be able to do online. So I think because of the privacy implications of age verification, and then also the question on, well, who's to say what is developmentally appropriate for a child, um, based on age, age verification seems like You know, a magic bullet solution, but nothing that I've seen about age verification shows me that it's actually going to work. I mean, as someone who's been doing research in this space for a long time, like, how are you processing how much information there is right now, how many stories there are right now, how much the sort of general public is engaging in discourse around kids online? One of the things that we study in HCI is value-sensitive design, and it's by Jeff Friedman. And the whole idea behind value-sensitive design is that we embed human values in the design of everyday technologies, whether we think about what we're doing or not. So for instance, parental control software. The values we're embedding is that parents have authority. They have the right to surveil their kids. There might be a lack of trust, and, like, and really the value we're designing for there is control. And why there's disconnects is that not all parents are authoritarian and want to wield control over their children. Some of them are more autonomy-granting and want their teens to self-regulate their online behaviors. And so this idea of value-sensitive design is related to the moral panic because when we have a disconnect between what we value as humans and what technology is doing to us as a society, I believe that's where it becomes a moral panic, where it's like, oh my God, my kid's, you know, glued to the screen eight hours a day, and I don't agree with that. They need to be outside playing. Or some of the content or interactions that they're having are online don't meet the same values that we have. And so that's one piece of it is that incongruence in values. The other piece of it is going based on, o- on feelings and fear rather than evidence-based research. If you look at the statistics, you know, there's still more kids, like, that are being challenged by teen pregnancies and automobile accidents and, and other, like, real world risks that, um For some reason, since social media hit the scene, we've kind of forgotten about all of those other risks, drugs, alcohol, and everything's been focused on social media. And maybe the reason is, is because the tech is the new-fangled thing, the new and shiny thing that we didn't grow up with and that we didn't understand when it hit the scene. So I think it's really both that incongruence in values, along with the lack of evidence and, and focusing on our fears rather than, you know, the data. Can you unpack that a little bit about, like, where you see presumption about what makes kids happy or myths about what keeps them safe, um, that help kind of tease out a little bit more about how we get into this, um, into this moral panic structure? Some of the myths that I consistently try to dispel with people are, one, that teens are, are purely victims and they have no agency, and that the internet is therefore a danger and a threat at all times. When we've worked with teens, we've actually seen the opposite. They make calculated decisions about privacy decisions. They are pretty effective at regulating their boundaries and reporting things when they're wrong or going to get help or saying no. So these are some of the myths that we hear about teens all the time, is that, you know, we treat them as victims online instead of active participants in our digital society. Another myth that I hear very often is about technology and, and addiction, where we're all addicted to technology I really wanna be careful of using this metaphor of addiction. Like, yes, like there are things that increase mindlessness and habituation towards technology use. It's so much easier to scroll on my phone and, and keep looking at reels. But if we use the addiction model, we're moving away from a model of like agency and self-regulation of like, these are things that I can control and that I can work to do better. I think the addiction narrative often brings us to a place of shame, um, and embarrassment, and that's not serving our youth well. And then a third myth that I often hear is that we often assume that teens are up to no good and that they're online seeking out these risks, and therefore we need to keep an eye on them at all times, um, and that we need to prevent this risky behavior. For the most part, what I've seen with teens is that many times they're trying to avoid risk, but they're trying to seek social connection. And so the teens that tend to get in more riskier situations are the ones who are, who are lonely and, and seeking love and affection. And so we wanna be careful that we don't mistake pro-social behavior of, of trying to connect and belong as risk-seeking behavior. In your research, like how would you articulate the problem? Well, I think first and foremost, it's a social problem. The idea is that we are trying to raise teens to become healthy and productive members of our society, and as parents, we worry that there's external factors that are preventing us from effectively being able to do that. And so when it comes to social media and tech use, cell phones, you know, AI companions, it's really these are pathways that we don't quite know how to parent. And to some extent it could be, um, over-dependency. For instance, it could be our kids are using AI too much and so they're robbing themselves of an opportunity to learn. It could be contact related risks of being able to make sure that we know who our kids are talking to online and, and that they're not unsafe actors. It could be content exposure of seeing things that, you know, if the internet didn't exist, they wouldn't have been exposed to otherwise. And, and really it's this, this amplification of both the positive and negative things that we experience in everyday life at a scale because the internet puts it at their fingertips. So it's really just a change and a disruption in the way that we socialize with one another and how we interact with this world that we can't pinpoint to a single risk. But I think the easiest way to operationalize that is at the gateway of screens, right? And so I think that's why we have so many conversations about limiting screen time. We talk about the effect of technology on kids as though there are answers, or that there should be answers, or that adults should have the answers while we ourselves are processing how these technologies are affecting us and not necessarily coming up with super great solutions for us as individuals or as society. And so we, we want so badly to have the answers for kids, and I think that's basically impossible when we don't have the answers for ourselves. That's true. That's like my main take away from this conversation. That's human nature. Like we did some co-design sessions with some kids to design safety features for TikTok. Mind you, these were kids like under the age of 13, so they technically shouldn't have known what TikTok was, but they did, and they were able to, to design some features for how to protect other kids. But that was the key, is they were designing when they did their first round of designs to protect other kids, right? And so they designed things like a po-po button for calling the police or, like, parental controls. And so we flipped it around on them and we turned these designs into, like, higher fidelity prototypes, and then we had them critique them for if these were being used with them. Then the story changed of just like, "Oh, this is too, like, creepy or surveillance based, and I wouldn't want this because this and that and the other." And it was just so telling 'cause, like, when we as humans are trying to come up with solutions on how to keep other people safe, we tend to be More heavy-handed and protective than when we realize, oh, wait, this also has to apply to me. If the platforms were to proactively build systems that were appropriate for kids, like, what would you, what would you wanna see? Like, where, where do you think platforms could improve the online experience for teenagers? I'm a big proponent of safety by design, which is taking proactive efforts to embed safety in the design of a platform a priori, like before, instead of just having re- uh, reactive features to, like, report harm and block people after the fact. Like, I think safety be des- by design is a paradigm that we could all use where- Yeah ... you know, features on a platform that we don't wanna use, we can turn off and toggle and, and customize to, to, you know, our best interest, and that's for adults and kids together. For kids specifically, I think one of the things that we're missing are spaces that are designed just for kids. So my daughter was, like, really upset to hear that, like, Club Penguin Island, you know, got canceled, and she wanted to, like, code to rebuild it herself. Um, and th- and that really made me think of just, like, where are places kids can go to play? We can't let them play outside because it's too dangerous. We can't let them go to social media 'cause there's predators. So I guess the biggest question to me is, like, where can our kids go to have positive experiences? And so building platforms that cater to the needs of kids to not only connect them with others, like, in the digital space, but also helping them find ways to connect with others in the real world. I think one of the things that we don't have in schools, that especially after the pandemic that was needed, are, like, scaffolding social skills so that our kids, you know, have places to belong. Because what we're telling kids is, "You don't belong anywhere, and we don't think you're safe anywhere." Like, malls are dying. Kids don't go to the mall and hang out anymore and watch a movie. Like, where do we expect them to go? I get really, like, twisted up about, like, where is the solution that is evidence-based gonna come from when you're in this kind of context where there's very little incentive to get it right for the individuals. Um, there's more just, like, this kind of performative level of discourse. So I don't know if you have thoughts on, like, if you have seen these types of evidence-based solutions being proposed successfully and, like, what it looks like to do that well, or if there's any, I don't know, anything to point to that, like, might be an inspiration to, like, what it would look like to systematically push back or center kids in solution sets. I see in our country as polarized, and because we're so politicized, even the topic of how to keep kids safe online- What I would love to see is people from different sides of the table coming together with, you know, the AI companies and the social media companies and the smartphone providers To have a, a, a meeting of the minds and instead of trying to push one agenda or another, trying to meet in the middle to find ways that we can agree. But I, I, I just haven't seen that happening. And, and the reason I haven't seen it happening is because even in my own research, when I talk about the middle of the road and call out the problems on both side, I make everybody mad, um, because everybody just wants you to side with them. And if you don't sell out and side with a certain party, then, then your voice is often lost in these debates. One of the things that I've seen both sides kind of agree on is that we should have safer defaults on platforms. So for instance, remember when you first joined Facebook way back in the day, it defaulted to sharing your newsfeed with everyone instead of just friends, right? One way that we can promote safety without restricting autonomy is setting kind of these, these more conservative defaults that allow a user to override them later on so that we're not just opening everybody up to the whole world at the beginning. And so I think that's one of kind of the bipartisan solutions of, of looking at platforms and making sure that they're not always set to the most open and promoting frictionless sharing over, you know, meaningful social connection. If you were to do that, if you were to have the kind of space to make the design choices that you would want to see, like what, what would, what do you think would be effective? Um, ultimately if we can design platforms that bake in more user choice so that they can customize their experience to what they want, I think that would be in the best interest of everyone. And this is not just for kids, but for adults as well. And so mostly it's more about helping people self-regulate how they use these platforms in a way that is beneficial to them. And so we, we need to get back to the place where we go back to the basics of how people connect with one another, and we need to scaffold ways to help people form those meaningful relationships. I don't think the biggest problem in our society is social media or AI. I think the biggest problem in our society is loneliness, and that we don't have social ecological support structures that we need to thrive As humans, because a lot of that's been replaced with technology and we have to keep up the pace and, and be more productive and get more done. So I would really like to go towards design patterns that help people find other people that make them feel like they belong, and that can give them a sense that they're not alone in the world. What our kids need from us more than anything else right now is just to know that like, "Hey, the Internet's kind of this Wild West, and I know you're gonna see some stuff that I don't want you to see, that you didn't wanna see, and I'm not gonna blame you for that." So I think we just really just have to have that sense of humility and realize that- Mm ... like, you know, our kids are navigating things that, like, they don't wanna see either, and we don't want them to see it. And, you know, we can't push solutions on other people that we wouldn't want for ourselves. Maybe we should be trying to solve some of these problems for ourselves as adults, and then seeing how that feels to see if that really makes sense for our kids. Because sometimes, again, we're trying to hold our kids to higher standards than we hold ourselves. Thanks to Pam for helping us think about what kids actually need from parents and teachers and lawmakers, and even the platforms themselves. For the second half of this episode, we are going to ask the million-dollar question. If we know all the risks that these technologies pose, what should we actually do about it? What laws can we put in place to protect kids without doing all the things that Pam warns us against? Blanket bans are one response gaining popularity around the world, basically saying, "If these platforms are harmful, they're harmful to kids, let's just keep kids off of them." But those bans quickly fall flat when they run up against the reality of how young people actually use digital tools, and they also prevent them from learning the resilience tactics that Pam described. Cory Crider is trying to open up a broader conversation about what could work, because we have to do something. Inaction is not an option. She's looking at solutions that take concern seriously while asking what policies could work in practice for kids and for everybody else using these platforms. So let's get into it Hey guys, I'm Cory. I am the executive director at the Future of Tech Institute, and we are a think tank that believes that Europe can have a different future, uh, for technology that isn't controlled by a handful of American monopolies. So we wanna see something that is sovereign, uh, but in a good way, not in a chest-beating jingoistic way, uh, and is developing technology in the public interest. We think that we can have a completely different tech economy from the one that we have today if we just take a little bit of power out of the hands of a handful of corporations. We work with not only officials to try to kind of come up with compromise solutions on thorny issues of tech policy, but we also work with people who are trying to build something better. So we will talk to people making an alternative social media service, or we'll talk to people who are trying to make, like, a small green cloud service. You know, we are interested in, I don't know if you wanna call it the non-aligned stack, but anyway, some other version of, you know, a, a, a tech- you know, a technology stack that doesn't answer to the same handful of companies. Because we don't have to have what we have now, and there are people who are doing some building, they just don't have enough attention and enough investment. What is the broad sense in Europe of what people want done? Is everybody clamoring for bans? Like, what is the energy and interest of the population? I mean, just to zoom out for a minute, yeah. Yeah. The... It's definitely the case that across Europe there is really mass, like north of 70%, touching 80% energy to do something, and we can talk about what that something is, to protect teenagers from the toxic end of social media. And I would say that parents also, the polling tends to suggest that they don't wanna have to be responsible for it themselves, right? Anything that kind of keeps the responsibility with kids and families rather than pushes it back onto monopoly social is really, uh, unacceptable to parents. But that said, uh, what at least we at Foti have been trying to do is to try and work with that concern and that debate and to say, well, look, ex- it's a bit like what you said with absence education, right? Which is to say blanket bans don't work. The early evidence out of Australia is that kids tend to get round them. I would say actually that there's a, that's a bit equivocal because there's a suggestion that a bit like with cigarettes, um, newer accounts, like younger, fewer younger kids are generating those new accounts. So it's not that it's had no effect, but it certainly hasn't just like completely transformed the space. Um, the other thing I would say is that, that really kinda surprised me is that Arcom, one of the French regulators, actually polled French kids between the ages of 10 and 14, and 69% of them thought that the thing should be taken out of their hands, right? Really? So I mean, what... Yeah, no, totally. Like 70% of- Interesting ... French teens were like, "Oh yeah, no, take this thing away." But all of that said, if you turn around and you ask parents, should the platform have to prove that it's safe and should toxic features be restricted, that also hits between, you know, kinda 69 and 75% as well. But, you know, and there was a poll run over the summer that did five major European jurisdictions. It did France, Spain, Germany, Italy, Poland. And support f- that people want those toxic features turned off. What are toxic features or like w- in that polling, like what are... Is there like a, I'm just curious 'cause I would- So the, I'm gonna use some slightly wonky words, but I, you know, this is a special audience, so w- like engagement based rankings, so the thing that's like, well, you looked at it, you clicked it, you liked it, you posted it, right? That, you know, just keep juicing that and giving it to you again and again. Auto play, uh, you know, targeted profiling, like the thing that got me to buy that stupid thing for the script. Um, but, and you know, and kind of infinite scroll, stuff like that. So that things, the, you know, I can't remember who it was who said this, but somebody some years ago said that the world's best engineering minds for the last 20 years have been focused on what will get you to look at something and click on an ad, and that has transmogrified into a very addictive set of design features, right? That keep people hooked, that keep people scrolling, and that every now and again get you buying something that you don't need to buy. But the side effect of that has really been pretty problematic stuff, because what, what we engage with often isn't something that if you asked us what we wanna see, like if we had a set of features at the front end that said, "What are you interested in? Sports, politics," whatever, what you would click, like, as your, your intentional interest and what you just look at, uh, obviously aren't always the same. And, and it's actually that latter signal that the profit motive of an attention maximizing ad funded surveillance, you know, driven platform, like your Instagram, like your Facebook, has tended to maximize. So anyway, so just to zoom back, back again, 'cause we've only, at Foti, we looked at this in Europe and the UK so far. Um, so there's been this overwhelming pressure. Meanwhile, of course, Europe has had laws like the Digital Services Act. Can you define the DSA? So the Digital Services Act is one of Europe's flagship pieces of legislation that was meant to make social media and big tech platforms behave better. But we've been so slow on enforcement that you've got just, you've got a real sense of frustration, I think, and anger, uh, when you go and you talk to families absolutely everywhere, and they're like, "I thought that Brussels was sorting this out, and they haven't done it, so do it now." And so you get then the leaders of member states, whether that's Pedro Sanchez in Spain, here where I live, or Emmanuel Macron in Paris, basically saying, "Well, look, if Brussels isn't gonna sort it out, you know, we're going anyway. We're just gonna do it ourselves." And if there's one thing that gives Brussels a rash, it's like a regulatory patchwork, right? So it's like, "Oh no, you know, there's like, uh, like more than a dozen member states are doing the wrong thing, so we must have a centralized solution." And this has been... I mean, this year we've been kind of calling this the Spanish cattle prod. The idea that, like, the, what has focused Brussels' attention on this is pissed off parents and then their local national legislatures and their national kind of executives being like, "Well- You know, you say that the DSA is gonna work itself out, but meanwhile our kids are getting older and developing eating disorders and hurting themselves and all of the rest of it. So we're, you know, we're going anyway, even if it isn't perfect, it's just no answer. So what we've seen this year is, um, all of these member states kind of passing their own laws and, you know, they've had a various shapes and sizes. Frank- France tried to pass what we'd call a blanket ban that's just like, poof, cut off no, you know, no access for any kind of teens. That got struck down by their constitutional court, so they're gonna have to do it again. But anyway, in this di- And then Spain actually has changed their bill to, and it started as a blanket, but maybe in dialogue with the Australians, maybe with others, maybe with us. I don't know. They, they moved to target features and that's, I guess, what we've been saying at Foti is like I, I, I don't like the idea that we're kind of, instead of working on trying to make, deal with the toxicity itself, we're just gonna kind of wall some subset of people out and d- and not deal with the fundamental business model that has created this problem in the first place. So what we'd love to see is the defaults change for everybody, uh, actually. And then as well, I also think like we will have failed as a movement if five years from now we are still arguing with Elon Musk and Mark Zuckerberg about what happens on their platforms and nobody is anywhere else, right? So even if we change the defaults, we've got to help people move to somewhere better. There was a huge child safety kind of re- wave of revulsion this January. I don't know if you remember, but it was back when Grok was still kind of live as a chat component of X, the artist formerly known as Twitter, and like you could just get it to produce n- kind of nasty CSAM or s- sexual, you know, non-consensual intimate imagery of like anybody's photos and researchers showed that in like eight days it produced something like 23,000 child abuse images. But like on this bridges out point, I tried then, I tried phoning like journalists and politicians who are still on X saying, "How about now? Like if we didn't wanna move off of X when it was like the kind of far right Nazi platform, maybe wanna move now that it's the child abuse platform." And people just said, "Well, yeah, but everybody's still kinda here, and news..." I'm not gonna call anybody individually out, but it, I was like, whoa, this is a sticky network effect. This is a really, really sticky network effect. People do not want to start over with their network, with their followers, with their posts. And so until we build people that bridge out, like we're just gonna, we're stuck fighting on their land about the way this stuff happens. So anyway So on the defaults, I'm happy to say, although it isn't perfect, that the draft of the Kids Act... So Brussels went and said, "Okay, we have to avoid this patchwork where every member state is doing their own thing 'cause that's gonna be a hot mess." So they've decided to try and centralize it to create a single act about how kids will be, uh, how kids will be treated between the ages of kind of 13 and 16. Uh, they, and they are proposing kinda safe defaults as the starting point for everybody. Now, there's a bunch of problems with that, a bunch of loopholes. We can get into that. But the idea- Yeah ... is safety by design first. They then say, uh, and there's even, by the way, there's even a limited form of interoperability. They say that, like, safe maker, other makers of, like, certain kinds of safety tools need to have the right to plug in, which is a start, right? That, that, that potentially is, like, the thin end of the wedge that you can use to start to build something real, interoperability out for everybody, which would really, I think, help. But, and, and then on the, on the part that I think is the most contentious with my community, and, and rightly so, right? Which is, well, yeah, but how do you do age checks in a way that doesn't just create s- yet another piece of surveillance infrastructure on the internet? They do specify in the draft legislation at the moment that the age checks that are required cannot identify or track or profile. It has to use zero knowledge proofs. You have to use a s- you can only work from certified options on a public list. Because at the moment, right, like, there's just a proliferating market for these providers, half of whom seem to be basically to be data brokers, right? So those all are out, right, in the draft bill. Um, and member states have to offer some kind of free option for people without documents. I'm not saying that there are no problems or no trade-offs, but it looks to me at least like some of the drafters of the legislation are trying over time to kind of move towards and reduce the space of kind of invasiveness of this. From my perspective, right, like, every safe default feature, like, seems to me to shrink the space for age gating. Like, I would much, like, I would much, much rather get to a space where we're not age gating any part of it, to be honest. But it, but I think what our community probably has to grapple with, you know, and this divides our community, let's be honest about that, is the fact that there is a really strong, pretty aggressive democratic demand to move something for kids now. Not to work, you know, not to work out and to be litigating with the platforms about kind of safe features and tenures, right? And so I suppose the way that we've started to see it is it's difficult, th- y- there are difficult trade-offs with these, with these systems, but that this is a s- like this is a, I hope a temporary space that we would like to see shrink in favor of a safe default for absolutely everybody, not just for kids, but for everybody. The Kids Act has, uh, basically requirements for companies to turn off certain toxic features, but what those features are specifically and how that'll be implemented, we don't yet know exactly. There's interoperability in broad strokes, so saying that platforms have to invest engineering time in developing standards that would allow for there to be exit from their platforms to other social media sites that have similar features or structures. And then there is age verification for platforms that in the legislation have some type of age limit. Is that the kind of broad strokes of what the Kids Act includes? More or less. So they, so they say that you need to have safe defaults, and then they start to talk about some of what the toxic features are, so kind of, um, engagement-based ranking, feed profiling, feeds, infinite, you know, infinite scroll, autoplay. They actually do name some of the addictive features. Um, on age verification, I mean, again, there's lots of, there's gonna be a lot to fight about in here. Um, one of the things that they say is, for example, is that if the platform already knows that the user is an adult, so in other words, from the incredibly detailed profiles of all of us that they have, then they may well be able to get around verifying. So if, like, a- with the draft as it is, I fully expect Instagram to come in and be like, "These gazillion people are already adults, and we don't have to change anything that we're doing." Um- Yeah ... which is why I'm really saying, like, it, it's what's most important, I think, is to open up a conversation about the default feed for absolutely everyone. Uh- Yeah ... but where there is an age check, they do specify that it needs to be a zero knowledge proof, that it has to be a certified solution. And, you know, the most important part, of course, is that the, like there needs to be something separate, uh, that where the state can never see, obviously, where anybody went, right? We can all agree that that's absolutely mandatory, and, uh, in theory, that's what the text of the legislation requires. There are all kinds of other problems with it. So, like, what's the most, what's the classic problem with European, like, what is the classic problem with European legislation in tech? Enforcement. Like, we all know that if we just, if we pass another law and we don't have a good enforcement mechanism, it is like they say in The Big Lebowski, "It's just, like, your opinion, man." Right? So they, they... And they just... So Article 5 at the moment, like, I won't, let's not get into too much detail, but there's, like, a kind of self-certification regime that isn't robust enough, so the e- the, the before how they come in and have to say that they pass it isn't strong enough, so there's that. There needs to be some kind of separate audit set up. It needs to be much stronger, and it needs to be much faster behind the, behind the gates, uh, where people can come in and try and enforce when there's breach after the fact. Like, if you think- Yeah ... about the DSA, like- There w- there was a look at TikTok and addictive design that took like two years to bake or something before they did anything. And like nothing in the current draft would avoid that timetable again. And I guess this is one of the things that I'm really worried about, if I'm honest, and where we all as advocates like really need to zoom in here and, and do some pushing because like they've got, uh, they've got everybody's attention now, right? They're proposing that this is gonna do something for families, and Brussels is already fairly kind of accused of having a democratic deficit. This, at least arguably, is meant to be a response to demand and response to public pressure. If we turn around and produce something that is yet another kind of damp squib talking shop without tools for enforcement, then it is going to discredit the entire European regulatory enterprise in a way that I think will may be quite hard for us to recover from, right? So it's gonna be, it is gonna be really, really important that whatever requirement comes in has some proper teeth, and that, and that isn't resolved right now. That's really not resolved in the text of the statute. But the principle of the thing, which feels really important to me, is that instead of coming from a knee-jerk blanket kind of age gate position, it's like how do we default to safe design? Yeah. And that I think is the beginning of a conversation that we wanna have about everybody. There's a couple things I struggle with in this kind of solution space. One is that we're going from basically no generalized recommendation for parents to a ban, or like to constraints by platforms. Like has there been, I mean, you may not know the answer to this, but like has there been any effort by the medical establishment or people who are positioned to be, um, providing guidance, like I'm thinking in the UK, like through the NHS saying, "We recommend parents not allow their kids onto social media platforms until a certain age for these reasons." Like there, there has, it, to me it seems like- What ... there hasn't been like guidelines or like advice. So there definitely are some studies, I can't remember what the name of the French study is, but anyway, there've definitely been some studies suggesting like kind of what the effects are on adolescent brains- Sure of kind of unedited exposure to that. I guess there's a kind of interesting question about where that then sits, right? So I mean- Yeah ... a- and now I'm gonna speak rather than scientifically, I'm just gonna speak anecdotally and socially as a parent. Yeah. Yeah. So what, so what I feel like has happened if you talk to families about, like over the last little while, is that the social pressure to give somebody a smartphone has kind of trended downward and downward and downward and downward. And so a kind of effect has tended to happen in schools where you, you don't wanna be the kind of holdout parent, like causing your child to be- Exactly ... socially excluded in some way. Um, and you know, and I'm that parent, right? Like I r- like in going back to my child, I'll be like, "No, I don't want you in that photo because they said they're gonna put it on social media. GDPR." And my child's like, "Mom, you're making a thing here." My child has never been on the internet, you know? Yeah, exact- exactly, exact- yeah. We're like the ultra as a parent. But you can imagine, right? Like it, like that, that, you know, even when she was small- The amount of time and space and, uh, patience and control, and like I mean, that is like, there are not very many families that have the space to push for those kinds of boundaries, and it's not fair that those families have been put in a position where like that's harder and harder. Yes. Well, 100%, and I think families don't wanna be put in a position of so- of being socially invidious. Like, they don't wanna, they don't wanna be the- Sure ... person, the only holdout family saying to their- Sure ... 12-year-old or their 13-year-old, "You can't have this thing because I don't trust what's happening, and I don't think I have any control over it." But where I appreciate what Pam is saying about trying to empower kids as opposed to kind of create this relationship of potential mistrust is, like, for, like, let's take that, that part of the draft legislation at the moment that says some kind of third-party tool on a plugin. Like, if kids themselves, when you ask them, say that they know that this stuff is un- safe, then why not give them some of the ability to kind of go through and plug in some kind of tool themselves that would make it a bit more appropriate, right? Where they get the part that they want, but maybe not the part that it sounds like quite a lot of people agree is pretty toxic. And then you could just specify, right, with some- 'cause then I can, I can hear, I can hear all the moms and dads, like, in my ear being like, "Yeah, but I, I can't trust my 13-year-old to choose that off the App Store." And you just say that any tool so offered has to comply with the requirements of the rest of the act. And then, like, you, you put a little bit more, like... So for example, there's nothing in it at the moment, um, that, that offers, like, a plugin interrupt for a feed or anything like that. But lots of people, I mean, if you've gone to any of the AT protocol services, Blue Sky or Mu, which is the Euro Sky equivalent, um, you know, there's lots of feed design, uh, the, at work where people get a little bit more control over what they see according to their own explicitly expressed interest, right? Which is- Yeah ... you know, a, a signal that you've actually chosen as opposed to, you know, the addictive kind of feedback dope, trying to keep you in a dopamine loop, which is the way that the thing is currently designed. Yeah. And I think, like, there ought to be, it, it ought not to be beyond our imagination to come up with something there that, again, actually empowers kids rather than kind of set something up in the way that it's currently set up. But, but I think- I just feel really like there hasn't been an effort to create heuristics around what's safe and what isn't, and, like, a broad set of guidelines. Like, I just don't, I don't think, um... Like, I'm not saying this is the answer, but it does feel like we've gone from zero to 60 in terms of, like, no advice that is prescriptive to it's so prescriptive that we have to technically engineer it into the platforms directly. I wonder if it's because for so long we've been gaslit by the platforms themselves. It's like there's, you know, you have all the choice you need, like, there's no problem here, there's no problem, there's no problem, there's no problem, and then suddenly, lo and behold, you have Facebook settling for some $17 billion to avoid a jury trial in the United States- Yeah because their own internal documents showed that they knew that this thing was addictive and horrible to teenagers, right? But given your career, when we start talking about age verification, like, you are doing an extremely good job of, uh, supporting policies that are pro-kids, that are pro-progress in terms of making positive changes on these platforms. Like, put your, like, cynic's hat on for me for just a second. Tell me what you're worried about when you hear about age verification. Look, you know, for those of you who can't be bothered to Google, right, I spent the first dozen years of my working life suing the national security state. Like, I only, only, only sued the defense department, CIA, MI6, NSA, and GCHQ. Those were the defendants on the other side of 100% of my cases for the first dozen years of my working life, right? I, here's... So I'm gonna offer a reflection about us as a movement, and then I will come back to talk about what I'm worried about. I will actually answer that question. I would've loved it if I myself, and maybe some of us in our movement, as well as having a theory of abuses of state power, we had accompanied that with a theory about private power. And I, and again, I'm talk- I'm telling on myself here, right? And I, you know, I don't know how many of you all will see yourselves in this story, but I, you know, I really, I came up through the W- kind of George W. Bush administration's kind of excessive response to September 11th. And I, for me, like the source of harm really was the state. And while we were fighting with the kind of post-Snowden state surveillance regime, I feel like we all sl- And, and I'm sure there are people who are s- exceptions here, but many of us, I think, and I would put myself in this category, slightly slept at the wheel while a private surveillance apparatus that is more as powerful as that, I mean, they don't mostly have missiles, but as powerful as that in terms of its invasiveness, um, swallowed the entire political economy. And so now we find ourselves in the position of have- trying to unwind that, um, you know, from a, from a position of some weakness. Because these guys have so much money and so much infrastructural power that they kind of can lobby and kind of compli- like, you know, compliance theater this stuff into absolute submission. So, so, so I think I got to a point where, you know, I, I thought, well, what is a level of power that is gonna counterbalance against this private power, such concentrated private power over how we trade, how we learn, how we communicate, what we read, what we see? And I don't see it other than some version of public power. Like, I think that we need it. And so shaping that public power is now up to all of us, right? So what am I worried about? What am I worried about? Somebody said to me the other day, "What's then to stop A kind of mission creep of that infrastructure through lots of other random parts of the internet. Like, what's to stop it becoming like a cookie banner or a CAPTCHA, right? Like, if you're a website and you don't get any ad money out of kids anyhow, like, why wouldn't you just put in like a, some kind of an age check? And then s- somebody who is, again, very well-informed on this said to me, "Well, and also, like-" It's a- Yeah. I've got to be honest, I hadn't really considered it. And th- and then they said, "And also, not only that-" So then it becomes, like, two layers of internet entirely where- Yeah. Yeah, yeah. And, and, and then it'll be around any kind of co- I don't know, like a vape w- it won't just be social media and porn sites. Like, it might well be a vape website, it might be, I don't know. And then you can imagine it might be a whole bunch of other things where, you know, actually views c- you know, the... I, I said that this stuff, uh, has overwhelming consensus, but once you get down to the question of, like, what parts of information should teens be able to access and what shouldn't they, then you can imagine parents would start to disagree pretty quickly. So I would be lying if I said to you I have a clear answer on that particular trade-off, which is how to avoid this thing creeping out beyond the very limited kind of time-bound and space-bound space where it may be temporarily necessary as, or required as a response to public demand, because I don't think the public would really like to see that. Um, maybe it's just that you actually have to specify not only that this is how age assurance should happen, but nobody else can do it. I don't know. Like, there, there may well need to be some way of kind of not creating that, but I think the, it may well be that some of the debate that we have, like we all talk about this, but we're actually w- we mean to be talking about something else, right? Um, and I d- but I, I think we can all agree, it would be, it would be undesirable, I think, if suddenly, uh, a whole bu- like, in the effort to deal with Instagram Or Facebook or X, kids can't access, I don't know, information about, I don't know, reproductive choice or things like that on Reddit. So I think that, that, that we do need to think very carefully about what the limitations on the deployment of some of this stuff are. Do you have any early indication as to whether or not there's gonna be arm twisting from the US on the Kids Act? I think because it's called the Kids Act, it becomes a little bit more difficult, but you can bet, you can bet your bottom dollar that the, that the pro- that the platforms will try. You know, this is more kind of innovation killing, whatnot from the platforms. But the other thing is, I mean, look, let's not predict what will happen in the midterms, but I wonder whether if the Republicans are shellacked in the midterms and we're kind of coming in and having most of the legislative debates about this in a, you know, possibly a lame duck kind of presidency, if, if, if, um, then the European conversation starts to look slightly different again. So I don't know. Um, uh, they will try. They will try, particularly because there's stuff in there. Like, I mean, again, it's only a very limited form of interrupt, but that interrupt got lobbied out of the, the Digital Markets Act and the Digital Services Act. There's no interrupt for social right now. And so the idea that there's even a little bit of plugging in is gonna, is gonna give them absolutely the heebie-jeebies. Particularly because, I mean, you mentioned the kind of ag- the agentic future. I mean, I'm not a futurist by any means. I'm a late adopter of everything, but, you know, I've heard from enough people that you can imagine all social media having, like, being slightly superseded, or lots of people choosing to in- like, access it through a partly agentic interface that, like, the right to plug in is something that they could find very confronting and very threatening to the business model indeed. Okay. So if people wanna keep you honest, if they wanna follow this conversation, what should they be doing? How should they do that? So look, keep your eye on the way this law develops because we are nowhere near home and dry and there's some stuff that we have to fix. The good news is the commission has moved the conversation on from blanket bans to design, but we need to keep going. So we need to make the rules quicker and simpler to enforce. We need to fix the feed for everybody, not just, you know, teenagers. And we need to build people a bridge out so that in five years' time we are not talking to Elon Musk and Mark Zuckerberg about what happens on their land only, like we actually have built people a bridge out to something better. Thanks so much to both of our guests for helping us make sense of this really complicated and increasingly urgent issue. There's one question both Pam and Cory raised that I keep coming back to: if we can't navigate the tech well for ourselves, how are we gonna expect kids to? I don't have a clean answer to that as a person in the tech politics world or as a parent, but most of us are being nudged and tracked and hooked by the same systems that we're trying to protect kids from. So as we think about what solutions for our kids should look like, it's probably worth asking what we want from these solutions and what it might look like for us, too. What would feel empowering? What would feel restrictive? Because if it feels that way to us, it will probably feel that way to them. And with that, we're wrapping up our back to school series, but this is not the end of the conversation. If you are energized by this issue and want to be in conversation with others who are, too, please do check out the Maybe Collective, because we will keep this discussion going, but also hopefully get us into some tangible action. So what's next? With the US midterms just over a month away, we are turning our focus to the intersection of tech power and politics in what's being called the Data Center Elections. We are kicking things off next week with Carissa Velez, the author of Prophecy, about the role prediction has played in helping powerful people hold on to control throughout history, and how claims of AI inevitability are influencing this election season. In the meantime, you can catch up on this month's episodes wherever you get your podcasts or watch full episodes on our YouTube channel. Thank you to the podcast team, Kushal Dev, Marion Wellington, Van Newman, and Zoe Trout, for all their work making this podcast happen every week, and we will see you next time.