Pod Insights The Vergecast Facebook executives Adam Mosseri and Andrew Bosworth on splitting up Facebook, privacy and more About this episode Facebook executives Adam Mosseri and Andrew Bosworth sit down with The Verge’s Casey Newton at Code Conference to discuss antitrust and the prospect of breaking up Facebook, Instagram and WhatsApp. They also discuss Facebook Portal, and how the company is adopting new approaches to privacy, content moderation, and more.
Learn more about your ad choices. Visit podcastchoices.com/adchoices
Transcript 8,431 words · 12 speakers
Speaker names marked ? are inferred from speaking time and the show’s known hosts — they are not voice-verified.
0:01 Speaker 00 Company spending is complex. Managing it should be simple. This is AMEX Corporate, a single platform connecting cards, expenses, and payments to streamline the way you work. Flexible controls for when priorities shift. Support around the clock. All backed by American Express. So you can stay focused on what matters most. Built to simplify, designed to scale. This is AMEX Corporate. Get started at go.amex/ammexcorporate. Terms apply.
0:34 Speaker 01 Soft, cushy, plush, cozy. Soft, smooth, buttery, marshmallow-y, warm. Soft. Socks, underwear, tees, and more. Bombas. Everyday comfort worth obsessing over. Visit bombas.com/audio and use code AUDIO for 20% off your first purchase. That's B-O-M-B-A-S dot com slash AUDIO. One purchase equals one donated, always.
1:05 Nilay Patel? Hey, everybody. It's Eli from The Vergecast. Got a special treat on this week's interview episode. Casey Newton, who's in the studio with me right now. Hey, Casey.
1:12 Speaker 03 What's up, Eli?
1:13 Nilay Patel? Casey and I were just at the Code Conference. I was a mere peon in the seats. But Casey, you were onstage. You did a bunch of interviews. And importantly, you interviewed Instagram's Adam Mosseri and Facebook's Andrew Bosworth. How'd that go?
1:26 Speaker 03 I really enjoyed it. You know, we write about these companies all the time. But how often do we get to sit down with top leaders and ask them anything we want to in front of a live audience? And then the audience gets to stand up and ask about anything they want to. So my only complaint about this interview is that I didn't get to do it for four hours. But hopefully, in the time that I had allotted, we got to talk about the stuff that was at top of mind for me. Like, should we break up Facebook? And why doesn't Facebook think that it should be broken up, you know, beyond their own self-interest?
1:59 Nilay Patel? And real quick, before we get into it, just give a quick history of who Mosseri and Bos are.
2:03 Speaker 03 Yeah, so both of these guys have been at Facebook for a long time. Baz was Mark Zuckerberg's TA at Harvard. He's been with the company since 2006. He helped to invent the News Feed. His current job is overseeing AR and VR at Facebook, which means he's overseeing all their hardware products, so like Oculus and the Portal home speaker. And then Misery joined a little bit after Baz and most recently ran the News Feed team where he was working on all of the problems related to Russian misinformation and platform attacks. But after he became the chief product officer at Instagram, as Mark Zuckerberg was kind of trying to install some of his loyal lieutenants at these big platforms that Facebook had
2:45 Speaker 03 acquired, Instagram's co-founders quit last year and Misery was put in charge of Instagram. So I think a lot of us see Instagram as the future of Facebook, and Adam has a huge role in shaping how that works. So both of these guys are top lieutenants to Zuckerberg and just have a really good insight into how he thinks and how Facebook is going to try to reshape the future.
3:09 Nilay Patel? And one thing I want to contrast, we spent a lot of time last week on Friday talking about the Susan, which sick interview. Susan is the CEO of YouTube. I think the consensus at the Code Conference was that she did not do a good job. But the substance of what she said is very similar to the substance of what Misery and Baz said, very similar to the substance of what the executives from Twitter said. The performance was very different. And Misery and Baz are on Twitter all the time. They're applied to you, they're applied to me, they're applied to randos. This is just my take on it. It seems like that experience gave them a ton of confidence in this interview.
3:43 Speaker 03 Absolutely. They do the work, which is not totally necessary, of going out and seeing what real people think about their platforms. I think if you're Jack Dorsey, you understand this problem really well because people are tweeting you all the time telling you that you're running Twitter terribly, right? But until recently, I don't think Facebook or YouTube have really kind of had to get in the mix on Twitter in particular, but Misery and Baz have. And while they'll be the first to tell you, like, it kind of sucks, their mentions suck, they do at least understand what people are mad about, and it lets them think through how they want to talk about it. So it is good, I think, for Facebook that they're on there doing that work.
4:22 Nilay Patel? All right. Well, that's enough preamble. Let's listen to this interview.
4:30 Speaker 03 All right. Hey, everybody. How's it going? Please welcome the VP of Instagram, Adam Mosseri, and VP of ARN Hardware at Facebook, Andrew Bosworth. All right, so lots to talk about today. And I want to start with an announcement that Mark made earlier this year, which was that Facebook was going to move to a privacy-focused vision of the future. Your core product right now is a public-facing feed. And Bos, you make hardware. So where do the two of you fit into this private future at Facebook?
5:01 Speaker 04 So I think when people think about Instagram, they often think about feed, because it's the heart of the experience. It's where we started. But there's also, obviously, Stories, which is growing very, very quickly. And what people don't often think about is messaging. Direct is actually driving a lot of our growth. So the way we think about it at Instagram is that there's a range of experiences, from the most public, me posting something for everybody to see that will be on my profile forever, to the most private, which is a conversation between just two people and messaging. And so certain types of expression make more sense on the feed side and certain make more sense on the private side. And we want to build a range of tools to enable all of that.
5:38 Speaker 03 All right, what about hardware?
5:40 Speaker 05 Yeah, for us, it's really convenient and consistent with where we already are on hardware. Certainly, the VR market is growing, but it's still relatively small and focused on a lot of single-player activities or kind of social experiences that are dedicated to that environment. And on Portal, we went really hard in the direction of private communication. You know, there's a lot of commentary when we launched Portal, like, hey, is this the right time? It was the exact right time, because Portal is a product that is exactly about what Facebook is at its core, connecting people directly. And it's entirely about private communication between two endpoints.
6:13 Speaker 03 But it's also a camera and a microphone inside the home that Facebook has at least some access to that data. I know when the product launched, there was a lot of commentary, including from a lot of the folks probably in this auditorium right now, that said, "I would never let Facebook put a camera and a microphone in my home." It's been on sale for a couple of months now. What have sales been like? And do you feel like this trust deficit that you have to reckon with has affected that?
6:34 Speaker 05 Been really good, and we actually have, you know, a lot more that we're going to unveil later in this fall, new form factors of Portal that we're going to be shipping. And what's interesting about it is, you know, it is a camera and it's a microphone, so it's capable of recording. You know, this gives a good insight into how much we were prioritizing privacy and user trust. We didn't ship the ability to record things. People can't record live videos, they can't record videos to send to their friends. It's really for calls, and those calls are encrypted. And so, we left functionality on the table to make sure that people felt like they understood what this device was. And then, kind of, not to put too fine a point on it, but remember, the reason we're doing this to begin with is we think there's a whole new generation of hardware coming out. The mobile platforms are relatively mature at this point, hardware's coming to the home,
7:18 Speaker 05 and you want to make sure that human connection, connection between two people, is a first-party experience on that hardware. And that wasn't what we were seeing with the kind of smart speakers and smart displays being put in people's homes. They were focused on these other kind of use cases that weren't very social. And so, not only do I am I happy with what Portal's done on its own rights and what the product is, but also, what you're seeing now in the industry is, smart screen manufacturers are adding cameras. They're refocusing on the ability to make calls and connect with people. When they were kind of, a year ago, they were going the other direction.
7:49 Speaker 03 Right.
7:50 Speaker 05 They were taking that stuff out. So I think we've had a really outsized impact for just being just a few months in.
7:54 Speaker 03 How many have you sold?
7:55 Speaker 05 I'm not going to answer that question.
7:57 Speaker 03 All right, what form factors are you thinking about?
7:58 Speaker 05 Well, I've, um, we're going to, we're going to announce those in the fall.
8:01 Speaker 03 Okay, all right, fair enough.
8:02 Speaker 05 I think they're very exciting.
8:03 Speaker 03 What about Quest? You also made like a VR console, it just launched, I played around with it a couple days ago, it is really fun.
8:10 Speaker 05 What were you, what were you playing when you threw?
8:11 Speaker 03 I was playing Beat Saber, which is a game where you have two lightsabers and you attack flying blocks.
8:15 Speaker 04 And you, is it true that you threw the controller?
8:17 Speaker 03 That is true. I did not follow the safety instructions, I did not use the wrist strap, and so that's a life hack for all of you in the audience. But talk to us about, you know, why did Facebook make a video game console?
8:28 Speaker 05 Yeah, virtual reality is really exciting for us. If you think about the history of Facebook, it's always been about human connection, connecting people, and arguably connecting people as broadly as possible. Even over like the finest filament of connectivity, people can connect to each other through Facebook. And VR actually is an opportunity to go deeper. Like you could have a really meaningful experience with somebody else who isn't there. The mission of my organization is to make sure people feel together anytime, anywhere. So when you can't be together, is there something we can do that can give you that sense of shared experience? And virtual reality offers an unparalleled opportunity for that. It's very early, and we kind of feel like with the Quest in particular, we just rounded out the first generation of virtual reality. We finally, you don't need a PC, you don't have a bunch of other wires.
9:13 Speaker 05 It's self-contained. You've got your hands. You've got this feeling that you're actually in a place. And it's very exciting, actually. We, just in the first two weeks since the Quest has been available, we've already seen $5 million in content sales. And that's important because now that we've got the hardware where we think it needs to be for the first generation, you want to build the ecosystem so that there's plenty to do, so that developers can engage and expect to make money on the platform. And so this $5 million is a big deal because that means we're kind of on that path to having this become a self-sustaining ecosystem.
9:42 Speaker 03 Right. All right, cool. Well, let's talk about another aspect of this privacy-focused future, which is some of the trade-offs that come with it. And you know, Mark has been very open about this, that when you build a platform that is end-to-end encrypted, you know, you invite people to do bad things about it. You know, a question that people are wanting me to ask you to on Twitter today is, when Facebook leans into a decision like this, how does it think through the unintended consequences? Is there a formal process for trying to evaluate what is going to happen as you, you know, maybe move into more encrypted products?
10:14 Speaker 04 So maybe not a formal process, but a series of very passionate, heated debates. One of the things that we don't talk about enough, I think, in general, is the tensions that arise. So there's a real tension between privacy and safety. The more data you have access to, the more you can keep people safe. You can identify bad actors of all sorts. But the less private everyone is, and vice versa. And so we put a stake in the ground and said we believed that messaging, the most private of communication media, that really should be encrypted. It should be absolutely private. And that was going to take time to build, yes, but it's also going to take time for us to rethink all of our different work around safety and integrity to work and be effective in an encrypted environment.
10:57 Speaker 04 And so it'll take time, but that was, that is the main tension. And that was, I mean, we talked about other things as well, but that was the primary topic of conversation.
11:05 Speaker 05 And even within privacy, there's tension. You know, some people, when they say privacy, they really mean privacy from other people. That's kind of how you think of Facebook privacy historically. Some people mean private from the government. Some people mean private from corporations. Some people even mean private from processing on the device. Those things have very different recourses, depending on what you care about. If you care about privacy from the government, you're probably not that well aligned with people who are lobbying the government to do privacy regulation. So I think it's a really, some of these issues get tied up behind big words like privacy or safety, but they're actually like exploded out into 10 or 20 different actually subcategories that don't all align.
11:44 Speaker 03 So like which of those categories, which kind of privacy is the most important to Facebook?
11:49 Speaker 05 Historically, privacy for Facebook, going back to 2008 period, was about privacy that people's control over their data and who was able to see it. And I would literally think that was the entire privacy conversation as an industry 10 years ago. We've really come a long way and that's actually good news. We don't have good answers yet. This is a global conversation in privacy. It's playing out not just in the US, it's playing out in Europe, it's playing out in Asia. And so for me, like I don't know that any one of those things is more important. It depends on who you are.
12:15 Speaker 03 Right.
12:16 Speaker 05 If you're in a country where your government is a greater threat to your personal safety than other forms of harm, that probably is the more important part of privacy. If you're in a more developed country where you're concerned much more about safety as it relates to terrorism or child endangerment, then you would take a different stance. So I do think there is an entire global conversation that's not going to have one answer of what's the one most important thing.
12:38 Speaker 03 Right. Okay, so some of these downsides are obvious, but I know, Adam, we were talking and you think there is an upside. You said that you think there are sort of some, some new products that you can build around sharing and messaging that, you know, maybe take advantage of some of these features. Like what are you getting excited to build at Instagram?
12:54 Speaker 04 So if you look at the range of things people do on Instagram, we talked a bit before about feed, about stories and about messaging. All the growth right now, really, in most of the world is in stories and it is in messaging. And we think it's just a sign that there is this paradigm shift, which is as important as the shift to mobile, towards more private forms of communication, for a whole bunch of different reasons. And as Boz alluded to, the motivations range depending on where you are in the world. But we're seeing more and more that it's sort of a demand. And you like stories, for instance, we talk a lot about effemerality. People always think about stories, they say, oh, they're only around for 24 hours. It's also a people-first model. You decide who you want to look at. So before you actually start to watch stories, which allows people to not worry so much about bothering their friends.
13:35 Speaker 04 But even more important, the actual conversations that come from stories are all private because they're messages themselves. You're not arguing about or looking at how many likes you have or having a comment argument out in public. And so that is just a more private conversation by nature. And the messaging, obviously, is the most private. And what Instagram messaging, I think, is great at, we use the jobs to be done framework a lot. We're not hired usually for utility messaging. You and I trying to coordinate maybe going out for a dinner or whatever it might be. It's about conversation starters. It's about having an excuse to talk to someone because maybe you just switched high schools or maybe you're single and you're interested in someone. These, you start conversations from feed or from stories. And then you have talked about everything else. You talk about life.
14:18 Speaker 04 The vast majority of messages on Instagram aren't story replies or reshares from feed, but they do make a majority of conversation starters. And so there's a lot we think we can do in that space. There's a lot we think we can, particularly for young people.
14:32 Speaker 03 You said if you're single, does that mean that Instagram's maybe thinking about like Instagram dating?
14:36 Speaker 04 No.
14:37 Speaker 03 Okay. Some people say it's already like one of the better dating apps out there.
14:41 Speaker 04 Well, here's the thing. I think we get hired for jobs like dating a lot without actually building products necessarily for them. There's, I mean, one thing is, oh, interesting. We see sometimes high schools will use Instagram for events. So you'll create a Finsta or a second account. You'll create it for an event. It'll be private, so you get to approve every follower. And basically you send out a bunch of invites and people actually ask. And then you basically create an invite list. That is a pretty hacky way of doing what Facebook does pretty well. But it works. It works for certain groups of people, usually teenagers. And that's great. We don't think there's like an issue with that. If people are finding different ways to hack the experience to give them what they need, we're all for it.
15:22 Speaker 03 Right. So something I still am struggling to understand is if there is this big shift into private messaging, what that means for the newsfeed, which is, you know, remains Facebook's biggest moneymaker, used to run the newsfeed. What's the case for the newsfeed still being vital in like three years?
15:36 Speaker 04 So stories and messaging are better if the conversation is more fleeting, if you don't want it to be around forever, you don't necessarily want to tell everybody in the world about it. Feed is the opposite. It's great if you want it to be around forever. It's great if it's something you want to stand up and yell to 150 people. Maybe you just switched jobs, or maybe you had a kid, or maybe you just graduated, whatever it might be. Maybe you just got in a relationship. Maybe you just ate breakfast. Yeah, exactly. And so I think that's going to continue to be an important part of Instagram. People come to Instagram to be with their close friends, whether that is in Feed or Stories and Messaging. They stay to be inspired by the world around them. And Feed, I think, is going to be for the highlights on both. It's going to be the highlights, the best of the best from your friends and also from
16:20 Speaker 04 your interests, be them music or entertainment, food, fashion, travel, whatever it might be.
16:25 Speaker 03 Right, but your expectation is people will maybe look at it less over time because they're going to be spending more time in messages?
16:29 Speaker 04 Certainly as a percentage of the overall time people spend on Instagram. We're seeing that the growth is being driven elsewhere.
16:35 Speaker 03 All right, so Facebook has proposed this privacy-focused future. Others have proposed a future where Facebook is broken up into its constituent parts. So, yeah, I know. Wild idea. You know, some of the reasoning people give, you know, the argument is that it could encourage competition, if there were sort of more little laboratories of ideas to try things out, that, you know, maybe frankly, some of their companies would just be better at handling some of the challenges that y'all have had. And then still, other people just say that you would make a really great CEO of a public company. So why don't we just break up Facebook and see how it works?
17:14 Speaker 04 Just as an experiment?
17:16 Speaker 03 Yeah, I mean...
17:17 Speaker 04 Just like run it for a month?
17:18 Speaker 03 Right, exactly.
17:19 Speaker 04 And then decide if you want to pull it back?
17:20 Speaker 03 Do an A/B test.
17:23 Speaker 05 That's not how they work.
17:24 Speaker 04 That's how that works.
17:25 Speaker 03 All right.
17:26 Speaker 04 I will answer a few questions.
17:27 Speaker 03 It's okay for keeping it together. Why is it good if it's big and giant?
17:31 Speaker 04 Personally, if we split it up, it might make a lot of my life easier and it would probably be very beneficial for me as an individual. But I just think it's a terrible idea. I think it depends on what problem you're trying to solve. If you're trying to solve election integrity, if you're trying to approach content issues like hate speech and you split us up, it would just be making exponentially more difficult, particularly for us on Instagram, to keep people safe. Right now, there are more people who work on integrity and safety issues at Facebook than anybody who works on Instagram. So when I joined Instagram, it was May of last year. As the head of product, I promised everybody at Instagram that I was not going to advocate for any major changes. I was just going to be a sponge for a number of months and try to learn the ropes.
18:15 Speaker 04 And the one place where I broke that promise was on safety and integrity. Because when I dug into the details, in far too many areas, we were rolling our own solutions as opposed to leveraging those from the much larger teams at Facebook. And so I broke that promise. It actually upset a lot of people, but it was by far the most responsible way to address our responsibility. So if you just split us up, you would cut that off. It would make those problems way more difficult. So I just think the thing is, you're talking about the split up. The question is, there are problems. There are many different problems. It's not like we don't have room to improve. But I think you have to be really clear about what problem you're trying to solve and why this will actually help.
18:48 Speaker 03 To me, I just sort of hear a circular logic in like, we are now so big that only we can solve our own problems.
18:53 Speaker 05 Well, I don't think it's that. I think one of the things that's interesting is when you start a new network, you know, at first, maybe there is some period of time where the content is all reviewable. Like, the content can be reviewed.
19:06 Speaker 03 When Facebook is just in the Harvard.
19:08 Speaker 05 At some point, it's like, okay. And by the way, it raises major privacy concerns. Like, should we be hiring a human to review all content? What if content actually is more encrypted and less available? But let's set aside for a second. At some point, you cross that threshold and now like your content is not fundamentally reviewable. And the bigger you get, the more attractive you are as a target to people who would abuse it, but also the more resources you have to fight those targets. Now, when you were workshopping this question on Twitter earlier, I saw, you know, Josh had a good answer, which was, yeah, you take Instagram and Facebook apart, you have the same attack surfaces, they just now aren't able to share and combine data. So this isn't circular logic, this is an economy of scale. This is about the ability for a company to invest really massively in a space. And I really do think we're behind on this. We are behind on this.
19:53 Speaker 05 The last year and a half has been exactly as humbling as it should have been here way sooner. But we believe these are solvable problems. They're hard problems, they're solvable problems. We're on a path to doing that, both internally in terms of the investment and also working with regulators to try to put regulation in place around things like content and misinformation and election integrity and data portability and privacy. And I think, so I believe these are solvable problems. You certainly don't get yourself any closer to solving them by splitting up the teams and giving each team proportionally fewer resources to deal with.
20:25 Speaker 03 Right. I wonder, like, what to you is the best evidence that Facebook, at its current size, at this very large size, is a net positive for the world? Right? Like, when I write about these platforms, to me, I feel like I'm always just writing about unintended consequences, just stuff you didn't see coming. And the reason, to my mind, that you don't see it coming is it's too big for you to even understand everything that's going on on your platforms. So what is the counterbalance there?
20:48 Speaker 04 I think that what gets written about are the mistakes. And that makes sense, because essentially what we're dealing with is creating a sense of, or creating real accountability. It's not fun for us to be criticized, you know, out in public, because we go home, we're people, we've, our family asks us questions and sometimes they're very clear on what they're upset about and why. But it's fundamentally a healthy thing. We're going through, through that sort of, I don't know, accountability process. But that isn't most of what happens on any of these platforms. My brother lives in LA, he's a musician, he's also a film scorer. My sister's a furniture designer, she lives in Berlin. I use a couple of our services to keep in touch with them on a regular basis. Small businesses use us to reach customers and they can hire more people because of it. People use us to learn about the world around them.
21:30 Speaker 04 Maybe that's the world of news, but maybe that's travel or cooking or something else. We create an immense amount of, I really believe we create an immense amount of value in the world. But I also understand that technology isn't good or bad. You asked, the way I would actually answer your question is, I think the mistake that we made was more about Not focusing enough on the unintended negative consequences of connecting so many people at such a large scale in the very early years. We were very focused on the good. I still 100% believe in that. We were not sufficiently focused in the bad. And then social media specifically is a great amplifier. It can just raise awareness of a good issue or a bad issue. So we need to do more to nurture and grow the good, but also more effectively address the bad.
22:13 Speaker 03 Boz, what's your case?
22:14 Speaker 05 Yeah, no, this is it. I mean, I think people use this product every day. They're not using it because habit. They're using it because it's creating real value. People have stories like Adam's about who they're connecting with, being close to their community. And I'm always a little surprised. I feel like if you go back to mid 90s, before any of this was a thing, and you ask people like, hey, like, what do you think is more valuable and more important in the world? Getting your goods delivered in two days, like having access to all the information, or like being closer with the people you care about? Like, that third one feels like a fundamentally human good thing that we care about. So I believe in the value, and I believe people with their preferences and how they exhibit and how they use the systems are exhibiting that this is a valuable thing for them. None of that takes away from the very fair and valid criticism of our company about
22:58 Speaker 05 being Pollyannaish coming into last year. We are, you can't emphasize how dramatically we've shifted internally over the last year to try to get ahead of the issues and be more transparent, more open about those issues. We're not there yet. We're trying to have a conversation about it now. But I do believe they're solvable problems, even if they're hard problems.
23:17 Speaker 03 What do you make of the argument, which gets talked about a lot, that companies that have advertising-based business models in this space are just sort of doomed to create products that have bad incentives? And that Facebook would be better if we paid it a monthly subscription fee?
23:32 Speaker 05 Yeah, this is one of the pieces. I spent a long time at Facebook working on the ads business, and it costs us. If you want to be really raw capitalist about it, which is not, of course, how we try to approach our work or our business, but if you really wanted to be, it costs us way more to have any of this marginal content, any of this, you know, marginal behavior on the platform in terms of the investment that we're making, tens of thousands of people that we're hiring to review content, to put in place. That's just raw cost, and it far exceeds any, like, top-line benefit. So, you know, if you really were a ruthless capitalist, which we're not, you would actually have a much smaller... There's a laughter there. I'm gonna dig into that with y'all in a second. Hold on to that thought. If you actually, if you actually really being ruthless, you would get rid of all speech that was even remotely objectionable.
24:17 Speaker 05 Because that's just pure downside. That's what you would call red revenue. And likewise, if you got rid of it, we've got good models of this WhatsApp. You know, we still have important investments that we're making in WhatsApp around misinformation. There's no advertising model driving that. And so I think it's a bit of a red herring. I think there's issues with advertising business models. Then we should deal with those directly for what those issues are. But it has to be done with a total accounting. Not just, hey, what are the benefits and time spent? What are the costs in terms of investment that you're outlaying? And I haven't really seen that full throated analysis done.
24:46 Speaker 04 We obviously believe in the value that we create. Some people don't. That's reasonable. We can always argue about that. But assuming that we create some value, it is, I think, something that we should be proud of, that we give that value out for free, right? Because you can, you can use our service, whether or not you use a $1,200 phone here in the States, or you live in Ecuador, or you live in Japan, doesn't matter. And we actually can afford to provide that service for everyone that wants to use the service, because it's an advertising business model. Which, by the way, is mostly paid for by people in developed markets, you know, who can afford to. It's easy to make the argument that if there was a subscription fee, that the incentives might be better. But then all of a sudden, you're cutting off access to a large percentage of the world's population, which I think we too often forget.
25:30 Speaker 05 The legalization of charging people money is so surprising to me. We're like, that's fundamentally regressive. It has to be. Like, if you're charging people money, it's going to be regressive. We, yeah, we like are actually building a service that people value for free.
25:43 Speaker 03 So you talked about these enormous investments that you're making in safety and security. And this relies heavily on a contract-based workforce. I'm particularly interested in content moderation, the people who do this work. And at this point, I've talked with dozens of moderators, you know, many of whom are struggling with post-traumatic stress disorder, having, you know, spent months or even years looking at some of the content on Facebook and Instagram. So, you know, I wanted to ask you, do you think these are good jobs? Are they safe jobs? And, and why aren't they full-time Facebook jobs?
26:14 Speaker 04 So a couple of different things. One, I think that the people who do manual review of stuff are actually an incredibly important part of the systems. And one of the things that we don't talk about enough is how all that we do to address safety problems on our platform need to be a collaboration between people and technology. Too often, I think, these things are put in opposition. I think that's a false dichotomy. People are great at certain things, particularly nuance. Technology is great at other things, particularly scale. And so what we try to do is engage with both as best we can to address the issues that we have to address. I actually really enjoyed your article. I think highlighting the issues for the people in these roles is super important. We take it seriously. There's obviously room to improve. We care a lot about the experience. We care a lot about the environment, et cetera.
26:57 Speaker 04 I mean, the working environment in addition to the actual environment, especially given that last talk. And one of the things that we've tried to do, like more recently, is we raised the minimum wage here in the US to $20 an hour. By the way, we were at $15 an hour for a few years now, which is not the minimum wage in the US, even though that keeps getting talked about.
27:13 Speaker 03 It's higher, yeah.
27:14 Speaker 04 It's significantly higher. And we've changed how we evaluate people in these roles, focusing more on accuracy and less on volume. Again, steps to just, I think, examples to demonstrate that we do care, but there's certainly room to improve.
27:26 Speaker 03 Yeah, we're almost out of time. One more thing I wanted to ask you about is you are currently undertaking an experiment to hide likes.
27:34 Speaker 04 Yes.
27:35 Speaker 03 Why? And given that people can still like content and presumably that data will still be informing how the platform works and feels like, what is the point of hiding likes? And what are you learning?
27:46 Speaker 04 Well, we don't want Instagram to be a pressurized environment. We want people to spend their time and their energy connecting with the people they care about and the interests that they care about. And likes can create that competitive dynamic. So what we're experimenting with, if you don't know, is just to make the accounts private. If you want to see how many likes are on your own piece of content, you can tap through and go see that. If you want to see how many likes are on someone else's piece of content, you can tap through and manually count it up if you have the time. We can't stop you from doing that. But the idea is that the small changes might actually really change the tenor of the experience. Now, will it work or not? I don't know. We're launching the test right now. Early data's been really positive so far. I think one of the interesting challenges is how can we actually measure the effect on sentiment? Is it changing how people feel about the environment or Instagram as an experience?
28:31 Speaker 04 That will take time and that is difficult to measure, but I'm still really bullish on it, and so I'm hoping that we can make it work.
28:37 Nilay Patel? All right, I'm going to step in, interrupt Casey, Missary, and Boz here. We've got to take a quick break for a minute.
28:47 Speaker 06 Your heart can tell you a lot about your health. Apple Watch Series 12 measures your heart rate every five seconds with the most accurate heart rate sensing in wearable. So your vitals app, now with heart rate variability, can tell you when something is off. And your readiness score can let you know when to rest and when to push. Hear the story in every heartbeat with Apple Watch Series 12. The features described are for wellness purposes only and not for medical use. iPhone 11 or later required. Based on Apple conducted study of heart rate accuracy, August 2026. Visit apple.com slash apple watch series 12.
29:17 Speaker 07 Purchase new carpet from The Home Depot and get it installed for free right now. Shop the stylish choices and simple pricing of carpet from Home Decorator's Collection, Traffic Master, or life-proof with pet-proof technology. And see for yourself why nobody beats our prices. Enjoy new carpet with free installation right now at The Home Depot. Offer valid October 15 through November 1, 2026. Exclusions apply. For licenses, see home depot.com slash license numbers.
29:47 Nilay Patel? That's enough advertising. Back to the Code Conference with Casey Newton, Adam Mosseri, and Andrew Brosworth.
29:51 Speaker 03 All right, let's take some questions from the audience. When you get up to that microphone, if you could please identify yourself, that would be very helpful. Let's go right here.
30:01 Speaker 08 Great. Hi, my name is David Samuel. I'm with Freestyle. Adam, this is a question for you. I have six kids, ages 10 through 16. All have been on Instagram since about nine or 10. I live in the Bay Area. Instagram is a primary messaging platform for my kids. One question for the audience is how many people in the audience have kids younger than 13 on Instagram? But all of these kids have registered with false ages, as they can't register for Instagram until they're 13. And I appreciate some of that direction is coming from Washington DC. So Instagram thinks my kids are 20 or 30 years old. My problem, my 10-year-old is now seeing ads kind of related to e-cigarettes, Juul, et cetera. It's really tragic. I pay YouTube for premium, so my kids don't see Google Ads.
30:44 Speaker 08 The question is, is how can my kids, and most of the kids who are on Instagram, experience the product without getting false or incorrect advertising to them? Because there are many young kids under 13 on Instagram. That's my question.
30:58 Speaker 04 So I think it's a great question. I have a three-year-old and a one-year-old, so they're thankfully not old enough to even try. So I haven't been tested on this yet, but we do have under 13 years on the platform, and that is not something that we want.
31:09 Speaker 03 Right.
31:09 Speaker 04 It's red revenue, to use an expression that Boz used before. It's something we need to do better at. How to do better at that is challenging, but essentially what we need to do is get better at identifying that you're... And we are better in certain languages and in certain countries and others to identify someone's problem as being underage. And also, you can report someone as underage, which we then go and look at a bunch of... What we can look at from a privacy-sensitive point of view to try and identify whether or not they're actually under 13. And if they are, we take them off the platform. We are much better in English than we are in most of the world. But you also brought up ads, which I think is an adjacent issue that I also want to speak to real quick. Which is, you're not seeing ads for e-cigarettes. If you're seeing ads... Actually, do we even do... We don't do ads for e-cigarettes.
31:49 Speaker 04 But you won't see an ad for something that is age-gated or age-problematic because you are of that age. You're gonna see that because there's some other reason why we think you're interested in that content. Which is, so make sure we don't conflate age with ads targeting, which are both important issues, but not necessarily one and the same. Yes, they can come together in problematic ways, but there are other reasons why some might be actually seeing the ads of your kids that they're seeing.
32:10 Speaker 03 All right, we're gonna move on to another one. Let's keep the questions fast.
32:13 Speaker 09 Hi, I'm Saher Zavari, co-founder and CEO of King Children. We're actually part of the Gift Lounge. So we make 3D-printed, custom-fit eyewear using AR. And my question is really about how you guys think about using AR and VR in kind of your traditional business lines, which are more social and gamified, versus using AR and VR to solve some more tangible problems that could not have been solved without using technologies like AR and VR.
32:37 Speaker 05 Yeah, we think both uses are important. I mean, certainly for augmented reality, you imagine a world where almost every screen or interface that we're interacting with today, things that we're paying hundreds or thousands of dollars for every couple of years could be replaced with software. And it makes it much more accessible to a much broader range of people for the kind of the price of entry. Those are a little farther away. Those use cases that are really blending the physical and digital together are a little bit farther away, just from a technological standpoint. In the more middle term, there's a tremendous opportunity, I think, to connect people to their work, connect people to each other in professional contexts with VR technology in particular in the near term, but also AR over the long term. There's a real... Unfortunate trend in the United States in particular.
33:19 Speaker 05 If you've followed Raj Chetty's work at Stanford, where social mobility is way down. When mobility goes down, access to jobs that are a good fit go way down. We're not only not taking advantage of the workforce that we have because they're geographically distributed, and that's a loss not just for those people, but also for us in terms of being employers. So I think there's a real opportunity to take augmented reality and virtual reality in the long term, this really kind of cool blended physical digital world. In the very near term, quite tangible, taking jobs that we have today, making them better through what is a real step forward in technology from an interactive platform.
33:53 Speaker 10 Dylan Byers, NBC News. Questions for both of you, which is just about commerce and retail, which is I'd love to know sort of what your ideal vision is for how you integrate retail and commerce into both the Instagram experience feed and story, and then also into the AR experience. And then also how you do that without sort of so cluttering the user experience that you end up sort of ruining the product.
34:21 Speaker 04 A few different things. We are very focused on commerce right now and Instagram. We think there's a lot of organic activity happening on the platform. There's obviously businesses that sell on Instagram and brands or people who are interested in that content. And there are creators and the economic engine behind the creator ecosystem is branded content. But we think there's a lot more room to improve. We think if we can connect, thoughtfully connect the dots between all those involved, we can unlock a lot more value. We need to rank content better for people. By the way, Instagram is personalized, feed is personalized. So if you're interested in shopping, you should get shopping content in your feed. Should you follow shopping related accounts? And if Casey isn't, then he shouldn't. That's how we should try to address that, address the color issue. We also need to get creators better insights into what content is doing well and what's not.
35:06 Speaker 04 These people, this is their livelihood. They use Instagram to make a living. They don't want to bother people with content that they don't feel is authentic to their own personal brands. And then we need to get advertisers better measurement on the return on investment because right now it's very, very little. So what I worry about is that there's some distortion in the market. I don't know exactly what it looks like, but we don't have healthy channels in each direction. But what I'd like to do is if we can do all of that, we can work towards a world where people can, who are interested can discover products on Instagram. And through the lens of people they look up to. Maybe you think someone dresses well. Maybe they have a similar skin tone. So you know that makeup that looks good on them will probably look good on you. We think that can be the future or is a big part of the future of shopping.
35:45 Speaker 03 No, they're gonna kill me. But I want to give one last question to Eli Patel. That was a very comprehensive answer.
35:49 Speaker 04 All right.
35:50 Nilay Patel? Casey has to pick me because I'm his boss.
35:52 Speaker 04 I'm sorry.
35:52 Speaker 03 There's also that.
35:54 Nilay Patel? I would characterize your relationship with Apple as tense.
35:57 Speaker 03 Are you saying you would?
35:58 Nilay Patel? It's complicated.
35:59 Speaker 04 You would?
35:59 Nilay Patel? Yeah, I think you would too.
36:00 Speaker 04 He said he would characterize it as tense.
36:02 Nilay Patel? Uh, it's complicated.
36:05 Nilay Patel? They just took a pretty direct shot at you by launching Sign In with Apple and mandating that their developers use it if they have a Sign In with Facebook button. Do you perceive that as direct of a challenge as everyone else does?
36:16 Speaker 05 I think Facebook has a good relationship with developers. And a lot of times when developers are trying to engage with us, whether it be for marketing or for the platform, they're engaging with us directly. I think, you know, we launched something significantly similar to that three or four years ago at F8, and it wasn't popular with the developers. And the reason it wasn't popular with developers is because developers use email address not just for the login, but also for a bunch of different parts of their workflows. So if you ever use an app and then you go and see a bunch of ads for an app that you've already installed, that developer probably doesn't have your email address. If they had your email address, they could save a little money and you could save yourself an ad that isn't actually relevant to you because you're already a consumer. And so the real question, I think, is what developers do and how that affects consumers. My sense is that for developers, they get a lot of value
36:58 Speaker 05 in their entire chain of acquiring consumers and re-engaging them by virtue of having access to that email address. They were not interested in a product like this three or four years ago. Maybe things have changed. Maybe they have. Maybe the convenience and efficiency will be good enough finally that it's actually worth it for them. Or alternatively, maybe they'll just roll their own and we'll have a bigger security, privacy kind of sprawl. That's a possibility as well. So I don't know. I think for us, we feel pretty confident about the relationship we have with developers. And I think it'll be interesting to see how it plays out for Apple as it rolls out.
37:28 Speaker 03 All right. Thanks very much. Please give a round of applause to Adam and Buzz.
37:32 Speaker 04 Thank you, everyone.
37:34 Nilay Patel? Okay, that was Casey Newton onstage at the Code Conference with Instagram's Adam Misery, Facebook's Andrew Bosworth. Thanks to Casey for doing that interview. It was great. I was in the audience. It was a lot of fun in the room. If you want to listen to more Verge podcasts, why don't you push that button? It's coming out with the second episode of their new mini-series called Death Online. It's coming out on Wednesday. This week, they're talking all about people mourning their dead robot, the Jibo robot. Caitlin and Ashley get into the relationship between people, their social robots, their cloud services, and explore the grief, and it's like real grief, of people that no longer get to use their Jibo robot. You should listen to it. It's something else. You can also subscribe to Vergecast for free in your favorite podcast app. Tap the link in the show notes to get new episodes. Please leave a rating and review on Apple Podcasts.
38:17 Nilay Patel? The producers always tell me to say Apple Podcasts, so do it there. You can talk to me on Twitter. I'm at Reckless. I'd love to know who you want to hear from on the show. The feedback has been great. I'd love to get more of it, so hit me up. We'll be back on Friday with Dieter and Paul for the chat show.
38:42 Speaker 07 Purchase new carpet from The Home Depot and get it installed for free right now. Shop the stylish choices and simple pricing of carpet from Home Decorator's Collection, Traffic Master, or LifeProof with PetProof technology, and see for yourself why nobody beats our prices. Enjoy new carpet with free installation right now at The Home Depot. Offer valid October 15 through November 1, 2026. Exclusions apply. For licenses, see homedepot.com/license numbers.
39:12 Speaker 11 Everyone's got a hard question about AI: your job, your kids, where it's going. Anthropic was built to surface those questions and share what it finds along the way, because there's hope in hard questions. Ask yours at claud.ai/spotify and keep thinking. That's claud.ai/spotify.
Previous Introducing Nice Try! Utopian Next Death, fear, and anxiety at Facebook’s worst-performing content moderation site in North America Page updated October 16, 2026 · transcript and summary are machine-generated
