Bluesky’s Big Bet + Are Deals Dead in Silicon Valley?

Plus, more ancient scrolls deciphered using artificial intelligence, Google’s chatbot updates, and a fight between record labels and TikTok.

Speaker names marked ? are inferred from speaking time and the show’s known hosts — they are not voice-verified.

• 0:00Speaker 00

  Your agents can talk to each other. They still can't think together. That's the gap Outshift by Cisco is closing. Outshift is Cisco's incubation engine, building the internet of cognition, shared intent, shared memory, and guardrails that allow agents to work together from anywhere. Open source protocols and code, a system for intelligence that scales. Read the paper and experience the demo at outshift.com. That's O-U-T-S-H-I-F-T.com.

• 0:32Kevin Roose?

  Well, Casey, you know, the Vision Pro came out this week.

• 0:36Casey Newton?

  Yeah.

• 0:37Kevin Roose?

  And, you know, I like to do nice things for you. Valentine's Day is just around the corner. And so I thought, well, I should get Casey a Vision Pro.

• 0:45Casey Newton?

  Oh, that's very nice of you.

• 0:46Kevin Roose?

  So I bought you an Apple Vision Pro.

• 0:47Casey Newton?

  You did, really?

• 0:48Kevin Roose?

  Yeah. Well, actually, they were out of the Vision Pro model, so I actually got you a Vision Amateur.

• 0:54Casey Newton?

  Oh, no, a Vision Amateur?

• 0:56Kevin Roose?

  Yeah, so here's your new Apple Vision Amateur.

• 1:00Casey Newton?

  Are these just ski goggles?

• 1:01Kevin Roose?

  Yes, they are. But I think we should wear them and pretend like we're wearing Apple Vision Pros. See, if we walk down Market Street right now, people will think we're early adopters, especially if we just go around like pinching things and swiping.

• 1:14Casey Newton?

  Yeah.

• 1:14Kevin Roose?

  I'm watching Raiders of the Lost Ark in 4K right now. What are you doing?

• 1:20Casey Newton?

  I'm just appreciating how much you've committed to this bit.

• 1:26Kevin Roose?

  These were $6 on amazon.com.

• 1:29Casey Newton?

  They're doing great things over there, whoever made these, but yeah.

• 1:33Kevin Roose?

  Keep gesturing.

• 1:36Kevin Roose?

  I'm swiping. I'm playing Fruit Ninja.

• 1:45Kevin Roose?

  I'm Kevin Roos, a tech columnist for The New York Times.

• 1:47Casey Newton?

  I'm Casey Newton from Platformer.

• 1:48Kevin Roose?

  And this is Hard Fork.

• 1:49Casey Newton?

  This week, Blue Sky CEO Jay Graber on opening up to the public and whether the world is ready for a decentralized social network. Then, The New York Times' Aaron Griffith stops by to talk about the turmoil in Silicon Valley after Adobe's acquisition of Figma fell through. And finally, it's time for a system update, Kevin.

• 2:06Kevin Roose?

  I hate system updates.

• 2:08Casey Newton?

  Close your apps.

• 2:21Casey Newton?

  Well, Kevin, ever since Twitter died and was replaced by X, there has been a lot of competition to see whether anything can succeed it.

• 2:29Kevin Roose?

  Yes.

• 2:29Casey Newton?

  And this week, there is another twist in that story because Blue Sky, a network we talked about a really long time ago, has now opened its doors to the public.

• 2:37Kevin Roose?

  Yeah, so we talked about Blue Sky on the show last year. At the time, it was this invite-only social network. It had spun out of the former Twitter a while back. And we both got invite codes and joined and talked about how we thought that Blue Sky was actually one of the more promising Twitter alternatives.

• 2:55Casey Newton?

  Yeah.

• 2:55Kevin Roose?

  Because it was basically, the value proposition was like, this is what Twitter used to be before Elon Musk bought it. But it also has this piece that's different than Twitter, which is about decentralization and sort of how these networks operate.

• 3:08Casey Newton?

  Yeah, and I have had my eye on this network the whole time. I've been using it the whole time.

• 3:12Kevin Roose?

  You're still using it?

• 3:13Casey Newton?

  I'm still using.

• 3:14Kevin Roose?

  Because I confess, like, I used it once and then, or like, a couple of times. I was very impressed, but then it just sort of felt like it kind of, I don't know, it fell off.

• 3:22Casey Newton?

  Well, I have been having fun there because it's been my observation that when I open up the feed at sort of random moments during the day, it just seems like people are having a better time there than they are on a lot of the other places that I'm checking.

• 3:33Kevin Roose?

  OK, so you're still using Blue Sky, but it's sort of been this invite-only community.

• 3:37Casey Newton?

  It has been, but at the same time, it's also truly one of the weirdest and wildest stories in Silicon Valley over the past several years. You know, Jack Dorsey, the co-founder of Twitter, said his biggest regret was that he had turned Twitter into this centralized, sort of big-for-profit, venture-backed thing, where some private company has total control over everything that you are seeing.

• 3:59Kevin Roose?

  They own the servers, they own the data, you know, they control the whole experience.

• 4:03Casey Newton?

  Yeah, he said it should have been a protocol instead. So he thought it should have been something like email, where anybody could build their own app for it, anybody could host it on their own server, and everything still talks to each other. And Kevin, we should say, at the time, this did not seem like a great idea.

• 4:18Kevin Roose?

  Right. I remember scratching my head and going, OK, Jack.

• 4:22Casey Newton?

  Yeah, but he actually pursued it. And he He brought in a team and he funded it. And shortly before Elon took it over, Blue Sky spun out of Twitter, which gave it a chance to try to realize some of these ideas.

• 4:34Kevin Roose?

  I'll just come out and say it. I am skeptical that Blue Sky is going to replace Twitter, at least for most people. You know, I was very optimistic about it when it launched. We did this whole episode. I think the title of the segment we did was "Blue Sky Has the Juice", because it really did feel vibrant and fresh and funny. And lots of people were sort of, sort of starting over there. But I think as time has gone on, I've just become more skeptical that these kind of decentralized social media apps can, can move as quickly as you need to, to be competitive in a landscape that includes things like Threads from Instagram.

• 5:08Casey Newton?

  So Threads is based on the Activity Hub protocol, and eventually, plans are that it will be decentralized. So I think there's a good argument that 130 million people a month are already using a decentralized social app. And at the same time, Blue Sky did not die. It remained relatively vibrant. It continued to grow until this week. It had about 3 million people who had used the app so far. And that brings us to this moment where we're about to see a format war, Kevin, because Blue Sky and Threads are built on different protocols. And frankly, not since VHS and Beta went at it in the 80s, I think, have we seen a fight with quite this much at stake.

• 5:48Kevin Roose?

  Yeah, well, my, my grandfather died in format war too. So this is a sensitive subject for me, but I guess we can proceed.

• 5:55Casey Newton?

  R.I.P. to him. So look, let me say, this is a bit of a wonky conversation, okay? But if you feel yourself getting bogged into the details as we discuss this, I just want you to remember, a lot of people hate social media. And if you want social media to be better, you might actually just need different infrastructure for it. You might need some different ideas about usernames and servers and how things interoperate and interact. And yeah, it is a little bit wonky. But if you want to see a better future, it's going to start with conversations like this.

• 6:25Kevin Roose?

  All right, well, let's hash it all out with Jay here in the room with us. So today, we're talking to Jay Graber. Jay is the CEO of Blue Sky. She's been working on this project since 2021, when it kind of was spun out of the old Twitter. And she is now opening the gates of this app to the wider public.

• 6:42Casey Newton?

  Well, let's bring in Jay and see if she can convince you.

• 6:49Kevin Roose?

  Jay Gerber, welcome to Hard Fork.

• 6:50Speaker 03

  Thanks for having me.

• 6:51Casey Newton?

  Hi, Jay.

• 6:53Kevin Roose?

  I'm curious to have you just make the basic pitch for a decentralized social network, because I think a lot of people are very unhappy with social media these days, but it's almost never, in my experience, that it's too centralized. I mean, they'll say, "It makes me feel bad," or "I spend too much time on it," or, "It's got all these polarizing and extreme people on it," or just, "It's boring." Very few people that I sort of meet outside the world of tech reporters and tech workers has anything to complain about when it comes to centralization. But I'm curious how you would pitch the average user of a service like X, or Instagram, or Threads, on the benefits of decentralization.

• 7:33Speaker 03

  Yeah, I mean, I think that's because decentralization is an abstract idea, and what you really want is social that puts power back in users' hands. And so, we really think of this as social media built for the people, by the people, because what decentralization gives you is open APIs for developers anyone can build, and so that unleashes experimentation. So if anyone wants to try a new way to do a healthier feed, a more engaging way that keeps you more interested, less bored, or a calmer way that gives you a healthier way to scroll, that's possible. Also, it gives users choice. So, once developers come on and build, because it's this open ecosystem, then suddenly, users have alternative clients to choose between, they have alternative feeds, they can pretty much customize the experience as much as they want.

• 8:18Speaker 03

  This also is open source social, and so users who have decided they wanted to search their own posts just went in and added that to the code recently. So, things like that can just happen, and that means social media can evolve much faster, not bottlenecking change on one company that controls the experience.

• 8:36Casey Newton?

  Yeah. Also, Kevin, I would disagree with you. I think people do get frustrated with centralized social networks all the time, they just don't think of centralization as the culprit. It's like if you get mad at Instagram, and you just wish you could take your followers to another photo app, you're mad at centralization, but you're probably not thinking about it in quite that way, right? So, I think one of the challenges that you have, Jay, is selling all the benefits of what you're building, even though it does seem super abstract, and often, I think, is not the way people are thinking about these things.

• 9:04Speaker 03

  Definitely, yeah. I mean, anytime somebody thinks like, wow, wouldn't it be cool if I could build this kind of an interface or have this kind of experience on social? Or what if my feed worked like that? And then they're like, oh, well, you know, the company will never implement that. You're mad at centralization because somebody could come along and do that if you were able to just, you know, fork the thing and take it anywhere and start building on it.

• 9:24Kevin Roose?

  Yeah. So there are two things that you're working on right now. One is Blue Sky and the other is the AT protocol. Explain how those things are different but related.

• 9:36Speaker 03

  Yeah, so confusingly enough, Blue Sky is the name of our app now, but it was also the name of this project that Twitter started to build an open protocol for social media back in 2019. So when I took over this project in mid-2021, we at first were building out the AT protocol, intending that Twitter would run on it. And then, you know, circumstances changed very quickly. I didn't really know what was going to happen, but I'd set us up as an independent organization because based on my understanding of the arc of centralized organizations, the captain can always sink the ship. And so if you have one person at the helm, even if they're very supportive of a direction like Jack was at the time, if Jack left, the support could go away.

• 10:15Casey Newton?

  Yeah, but what are the odds that Jack Dorsey would ever leave Twitter?

• 10:18Kevin Roose?

  Or that we would have an erratic and unpredictable person in charge of a social network with no accountability.

• 10:23Speaker 03

  It seemed like a time in history when we should expect the unexpected.

• 10:27Casey Newton?

  Yeah, true.

• 10:28Kevin Roose?

  Yeah, so there's also something called federation, which I want to have you sort of remind us what that is. And kind of what is the status of federation at Blue Sky?

• 10:39Speaker 03

  Yeah, so federation also in some ways describes an abstraction, but the most basic thing is letting services talk to each other and getting this interoperability between them. So a protocol is really a language that lets computers talk and exchange information. And we are setting that up so that anyone can connect a service to the network and have it talk with the rest of the network. Pieces of that have already been split out. So for example, the custom feeds, you could say in one sense, are already federated because third parties, developers hosting their own infrastructure that were all struggling to keep up with load last night, have set up their own feeds already.

• 11:16Kevin Roose?

  Can I repeat back what I think is true and you tell me if it's wrong or not?

• 11:19Casey Newton?

  This will be fun, let's see.

• 11:20Kevin Roose?

  Okay, so right now, Blue Sky is the only social network on the AT protocol. But through federation, other developers will be able to host their own Projects and apps and services on the AT protocol. Is that correct?

• 11:39Speaker 03

  Yes, it's partially.

• 11:40Casey Newton?

  That's good, Kevin.

• 11:41Speaker 03

  And it's more than that.

• 11:42Casey Newton?

  I'm sweating.

• 11:42Speaker 03

  Others can host any piece of the social experience themselves. And essentially, I think actually the highest level way to think about this is a social media experience that's not all run by one company. And so anyone can come in and often there are, once you open up the ecosystem, there are many, many third parties. And in some sense, it doesn't even make sense to think in terms of third party, first party anymore, because your social experience is just composed of all of these different people who have built things that are coming together in your experience now.

• 12:13Kevin Roose?

  So theoretically, it would be possible to have kind of like an Instagram-like app where the stories feature was built and run by Snapchat and the shopping feature was built and run by Amazon. And, you know, the sort of for you page or the explore page was sort of run by some other startup. That kind of sort of Frankensteinian social media experience would be possible?

• 12:35Speaker 03

  That is possible. And that's the way we've tried to design this protocol is for composability. And so you can plug these things together. One of the design principles we've been trying to follow is third party is first party. So that's why when you come in and you see custom feeds and you scroll through them, it's pretty easy to just select something built by another company or another organization. Like the for you feed, one of the really good algorithmic feeds that has emerged is built by another company. And it's great because it provides a different experience to users than the algorithm that we built.

• 13:04Casey Newton?

  Yeah, and by the way, if all this is still sounding too abstract to you as a listener, we are talking about something very similar to email, very similar to the open web, right? Where it doesn't matter what server you are using, anybody can build a software that shows you the web, that shows you your email. You can email anybody on any server and it just kind of works. And email on the web, obviously, two of the most powerful things that have ever been built on the internet. And what Jay is working on here and what some other folks who are working on their own protocols are working on is can we have a protocol for social networking and for communication? And I just, I think it's super interesting.

• 13:43Kevin Roose?

  Yeah.

• 13:43Speaker 03

  Yeah. That is the best analogy, because we also are very inspired by open web principles. And we try to follow what came before in the web, like letting you use your domain name so you can link your social to your blog. And essentially, self-hosting is basically like anyone setting up their own blog. And then the sort of service that other social companies provide on top of it is that we'll be providing is sort of like Google going along and scraping all the blogs together and then showing you a search over them. And then the feed is a bit like your RSS feed in the old days where you got to look at all of the blog posts that people had written, but everyone had set up their own blog out there on their own domains and were running them in different ways, different places.

• 14:21Kevin Roose?

  Yeah, I want to stick with the email example for just a second because I think it actually brings up one of the common objections to or just criticisms of this sort of decentralized idea of social media, which is that it actually ends up kind of not being that decentralized in practice. I mean, with something like email, yes, anyone can run their own email server. It's an open protocol in that way. But in practice, most people have Gmail accounts, or they have Hotmail accounts, or they have Yahoo accounts. There are sort of a few big- To be clear, most people do not have Yahoo accounts. But don't want to. Most people over the age of 75 have Yahoo accounts. But basically, you end up with a system where because of power laws and network effects

• 15:06Kevin Roose?

  and just the sort of ability of private companies to move faster in some senses than kind of decentralized projects, you end up with kind of a handful of companies sort of de facto controlling most people's experiences of the thing you're building. So how does Blue Sky think about that? And would you be happy with the state of decentralized social media if it ended up kind of like email did, where there were a few big providers building on top of this open protocol?

• 15:31Speaker 03

  Yeah, we looked at this aspect of protocols early on. Most of the early team came from building in decentralized protocols worlds already. And I did this full ecosystem review of all these existing protocols out there. And it seems to often happen that you get power laws across any sort of dimension. But in thinking about decentralization, we asked ourselves, what do you really want out of this system property? Essentially, an abstract property of a system. And what you're really trying to give here in terms of benefit to users here is the right to leave. So you can vote with your feet and you're wanting to give users and developers the freedom to innovate. And so they can just unleash their creativity. And so we designed around account portability, giving users the ability to leave with their accounts, move it to another provider.

• 16:14Speaker 03

  So if things end up centralizing around like a Gmail, then one of the problems with email here is, you know, you can keep your contact list and you can communicate from a Gmail account to a Yahoo account. But if you change email addresses, you kind of have to start over. And so the goal here was to let you essentially keep your email address, you know, your identity, your user handle if you move services. And then it makes it even easier to switch, you know, your service provider.

• 16:39Kevin Roose?

  Yeah, I'm curious. We just had Chris Dixon on the show, who is a big proponent of decentralized social media, specifically using crypto and cryptocurrency as kind of a way to finance the creation of decentralized social networks. You used to work in the crypto industry before you worked at Blue Sky. I believe you were a developer on a project called Zcash, which is sort of a Bitcoin spinoff. And I'm curious why you decided not to build cryptocurrency and blockchains into Blue Sky and the AT protocol. Were there things you learned while working in crypto that were like, eh, maybe we don't want to go that direction? Because this is something that a lot of people, including, I think, Chris Dixon, would say naturally go together.

• 17:21Kevin Roose?

  You have decentralized social media and decentralized money, and you have to sort of have both.

• 17:27Speaker 03

  Yeah, so I got pretty deep as a blockchain developer because I like to always get to the center of what I'm building. I was working on a cryptocurrency that introduced a new form of cryptography, zero knowledge proofs, which are very cool. And there was a lot of promise that they would solve problems of privacy and scalability. And so I was working on trying to fulfill some of those promises, but knowing deeply the limitations, people really struggled with, you know, wallets and their personal keys. It was really hard to do anything at large scale in a, you know, cost efficient way on a blockchain. And it added a lot of extra complexity. And so I realized this would just not be the way to. And our mandate here was to build something in a relatively short order of time that Twitter could run on. And so it was to build a protocol that you could operate

• 18:12Speaker 03

  at Twitter scale with Twitter standards. There's also questions around the immutability of blockchains. I just don't think that you need to have the picture of what you ate for lunch on an immutable ledger forever.

• 18:25Kevin Roose?

  I do. My lunches are very important for the historical record.

• 18:29Casey Newton?

  What I love about the way this turned out, though, Jay, is that I think you're now working on a project that hopefully will let you prove out the benefits of decentralizing without having to lead everyone down this path of crypto that is super confusing, that is so hard to use, that seems to inevitably result in people losing a lot of money. And I think there's just something really interesting about that, that protocols and decentralization may indeed be the answer to some of the frustrations that we have with the internet as it exists today. But you're not actually going to need to involve a blockchain to solve it.

• 19:02Speaker 03

  Yeah, I started off my career as a tech activist. I was working at organizations working on net neutrality, antitrust for tech issues, privacy. And I actually had this realization when I was just getting frustrated that we were doing all of our campaigning on Facebook and Facebook events and Twitter back then. And I wanted to build something that would just give people more power directly. So that's why I learned to code. And that's why I eventually circled back to social, was just this moment of realization that maybe if I learn these technologies well enough, I can someday build things that just give power directly back to people rather than waiting on regulation. And that's something where I think that when, by the time that people get around to thinking, oh, maybe we should regulate tech, maybe we should try to use policy to direct things in a more

• 19:46Speaker 03

  healthy direction, power has already consolidated a lot. And it's very hard for policymakers who don't understand the technology to understand the best interventions that can be made.

• 19:55Casey Newton?

  You made an interesting point to me when we were talking the other day, that you believe that more democratic communication networks could result in a more democratic country. Can you share a little bit about how you think that the way that we communicate influences our politics?

• 20:12Speaker 03

  Yeah, so in school, I studied Science, Technology and Society, which is sort of a history and philosophy of science and technology major. And I spent a lot of time thinking about how the internet was changing society and the promises and perils of it. And I think that one thing that seems to have some degree of truth to it is when a society gets a new communication medium, everything gets disrupted. And then also, societies start to take the form of their new communication medium. Like, in some ways, there's this mirroring, this reciprocal relationship going on. And so if you have a very centralized medium, societies start to take that form because then you give one person a megaphone and the ability to direct everybody or propagandize or just put their thumb on the scale.

• 20:55Speaker 03

  And then things move very quickly in one direction or another. That's one of the benefits of centralization is being able to move fast. But that's also a downside. And so that's why I say the captain can sink the ship because it's high variance. And so that's why we have decided generally collectively that democracies are a stable form of governance that we, hopefully. Like, maybe there's some doubt around that now as well. But I do think if we want to give democracy its best shot, we need more democratic communications infrastructure so that we're actually able to tap into the collective intelligence of a pluralistic society and not have it all just swing one direction or the other as soon as something as, you know, basic as ownership changes.

• 21:32Kevin Roose?

  Talk a little bit about content moderation on Blue Sky and other decentralized social networks. Because this is a thing that people at the sort of big centralized tech companies will say when you ask them about decentralization was like, "Good luck moderating that." So right now, we're sort of having a conversation about AI-generated imagery, for example, on social media and what to do if, as happened just recently, a bunch of AI-generated deep fake nudes of a celebrity like Taylor Swift are posted. That actually happened on X. And a lot of social platforms right now are struggling with some of these kind of questions about what do we do about this content? Do we ban it? Do we allow it but make it harder to find?

• 22:14Kevin Roose?

  My understanding of Blue Sky is that you all have your own content moderation guidelines that are actually stricter than the guidelines for the protocol as a whole. So if I'm someone who does think that platforms should be showing people AI-generated nudes of celebrities, could I make my own feature on the protocol that would allow that to show up in Blue Sky? Or like, how does that work?

• 22:38Speaker 03

  Well, our design principle is like the open web. So anyone could set up their website to do this and you could set up your own infrastructure. But essentially, a key aspect of social is interconnection and what you promote. And this idea of freedom of speech, not reach, means that anyone can, you know, set up their own hosting somewhere, but we don't have to show it to anybody. So I think, you know, we already do a lot of, you know, de-promotion of stuff. And then the feeds make different choices. But each layer of the infrastructure essentially is making its own choices about what kind of community space it's setting. And the Blue Sky app is setting certain norms. Like, we have community guidelines that, you know, try to keep a good space that don't allow like hate speech or harassment. And we hold to these standards within the app. But then someone can go build a different app.

• 23:22Kevin Roose?

  So I could, you know, if, if, you know, there were someone out there who is interested in seeing, you know, deep fake nudes of celebrities, they could go build, you know, a part of the protocol that would allow that. But you're saying that you, as Blue Sky, the app, not the protocol, the app, could choose to sort of ignore that or not let that onto people's feeds? Is that what I'm hearing you say?

• 23:44Speaker 03

  Yeah, there's, because it's an open system, there's things we can and can't control. And the things we can control is what is, you know, within the app, within the app we've built and the infrastructure we run. And then the things we can't control are when people run their own infrastructure using the same code in parallel.

• 23:58Kevin Roose?

  Right. And is that a flaw of decentralized social networking? Because I think what a lot of people sort of object to about maybe the Taylor Swift nude situation, to be, you know, to be specific about it, it's not that they just don't want to see that stuff on their feeds, it's that they don't think that anyone should be able to spread that stuff online. So that's an issue for decentralized social media, isn't it? You can't actually stop people from being able to do that.

• 24:22Speaker 03

  Well, the internet kind of has different layers of governance. Like, you have, you know, sites that set their own rules, and then you have backend infrastructure like Cloudflare that takes down some of the most serious stuff. And that's sort of how this ecosystem works as well. So you have sort of escalating layers of severity up the infrastructure stack. And so the things that I think are more like infrastructure require more social consensus in order to do content moderation on. Like, it's much harder to get something taken off Cloudflare than it is to get something, you know, taken off a social site. And that's how this works as well.

• 24:56Casey Newton?

  Yeah, I mean, the Blue Sky approach is really not all that different from something like Reddit, where Blue Sky sets a floor, but then individual users can raise the ceiling for moderation. They sometimes describe it this way, where you can sort of add your own rules on top of it, like, hey, I don't want to see a lot of curse words, I don't want to see a lot of nudity, whatever. And what I think is interesting about this approach is that for the past, you know, seven or eight years, we've been having this very unproductive discussion about, like, what should the parameters of online speech be? Should we take way more of it down? Should we leave way more of it up? And the idea in Blue Sky is sort of like, well, why don't we just let users decide what they want to see? If you want to have a really intense environment where people are flinging mud and calling each other names, like, you can have it. And if you want something where

• 25:39Casey Newton?

  everyone is like being super polite and no curse words are allowed, you can have that too. That is an experiment that I'm very interested to see how that goes.

• 25:45Kevin Roose?

  Yeah. I mean, I think I and a lot of other people, we sort of choose social networking apps based less on kind of technical features and more on vibes. Like, what does it feel like to use this thing? Are the people on there funny and interesting? Are they telling me? You know, are they making me laugh? Are they informing me? Am I having a good experience? So, talk a little bit about the vibe side of Blue Sky and how you've tried to sort of cultivate the right community there. And kind of, what do you think it's going to take to get more people interested? Is it celebrities moving over? Is it, you know, is it comedians? Is it sort of developing sort of niche interests like that? Like, what are the keys to cultivating a good vibe on a new social app?

• 26:27Speaker 03

  Yeah, I mean, I think all of the above goes into adoption. But so far, it's been really fun. And it's quite chaotic, I think, because a lot of people coming on were coming through invite codes passed around on other social apps. Everyone was posting heavily. And so, lots of back and forth, very casual posting, a lot of shitposting, and just, this has made it, I think, a very active culture.

• 26:49Kevin Roose?

  Yeah, the platform attracted early shitposters like Casey Newton.

• 26:54Kevin Roose?

  Truly demented stuff going on over at your account.

• 26:57Speaker 03

  Drill comes into like, comment on every, like, big news day.

• 27:01Casey Newton?

  Yeah, I think he had one of the most popular, Drill, of course, the sort of former king of weird Twitter, D-R-I-L. The pseudonymous poster and yeah, he had one of the most popular posts. In fact, maybe it's even worth reading this because it sort of suggested to me that he wanted a response. And what he said was, I do not fucking recall them asking the Blue Sky elders permission to open registration to commoners. So Jay, I guess, how would you respond to the idea that you didn't have permission to open up the network that you're building?

• 27:34Speaker 03

  Well, you know, it's if we wanted to keep it a secret, we had to keep it a secret because it's hard to keep a secret with 3 billion users. But we couldn't, you know, tell them. But it was also very funny because Blue Sky elders has been a meme for a while and it's like Drill knows the deep lore of what happened on Blue Sky over the last year. I don't even know. Like, it's been very funny to get to a place where there's like a very active poster culture when people are coming on because often when people come onto a new social network, it's kind of a ghost town and then like people have to figure out what the norms are and people have gotten creative, you know, they've really taken it and turned it into something that they want. And also the fact that the app is open source means that users that came on who are developers who are like, why can't I, you know, look at videos?

• 28:19Speaker 03

  Someone went in and just added like a way to do YouTube embeds because they got frustrated with not having that in the app. And that's one of the great things about open source social is, you know, anyone can come in and modify it.

• 28:31Casey Newton?

  Yeah, I was on Twitter a long time ago. I was on Twitter before this sort of notorious four quadrants post, which is when the company came in to an ecosystem that had a lot of different developers building their own clients. It really was a sort of very decentralized world. And Twitter came along and said, actually, you have to stop building your own clients. We're going to build our own client because we want to centralize this. We want to be able to show ads. The only way we can do it is that way. And I think in a very real way, that marked the beginning of Twitter's decline. Like, they just, they stopped innovation dead in its tracks when developers and users had been giving them so many good ideas and actually been building good ideas for them. So it's possible that I'm getting carried away, but when I look at Blue Sky today, I

• 29:16Casey Newton?

  think we are getting a second chance to run the experiment. We are getting to see what would happen if Twitter had actually just let a thousand flowers bloom, see what happens, see what people build, see what kind of fun they can have. And while, you know, I do think you got a long road ahead of you, Jay, but I'm at least excited, I'm at least excited that somebody is trying because we have not had enough experimenting in social networking over the past decade. And so I'm very interested to see what happens now.

• 29:42Speaker 03

  Yeah, social should be a playground again. I mean, it's where we spend our time talking. Why is it this thing controlled by one company where we can't play around and try new ideas? People give me new ideas for how they think social should be better or have some like fun quirky idea for a new thing to do all the time. And I just want a world where people can freely build that and play with it. And we'll see this profusion of all sorts of new experiences come about.

• 30:04Casey Newton?

  Yeah, I think that's pretty good. Anything else that you would point out or something that we didn't ask about or feedback on any of Kevin's posts?

• 30:11Kevin Roose?

  I've only made like three Blue Sky posts.

• 30:12Casey Newton?

  Well, that would be good to give a few.

• 30:13Kevin Roose?

  I churned.

• 30:14Casey Newton?

  You had your viral sandwich skis.

• 30:16Kevin Roose?

  It's true. I did have a sandwich post and it performed very well, but then I was so intimidated. I knew I could never live up to the sandwich post.

• 30:26Kevin Roose?

  Jay Garret, thank you for joining us.

• 30:27Speaker 03

  Thank you.

• 30:30Kevin Roose?

  When we come back, Aaron Griffith from The New York Times tells us how regulators are screwing up the life cycle of Silicon Valley.

• 30:52Speaker 04

  AWS AI is how the world's leading organizations are transforming their industries, not in theory, but in production at scale right now. Like turning race day data into real time insights for Formula One fans, reinventing the journey for every United traveler. And personalizing every trip with booking.com. However big the question for your business or industry, AWS AI is how.

• 31:17Speaker 05

  Companies are pouring money into AI, but few leaders can explain how those AI investments work with the systems you already have. You know, the ones that actually run your business. Without systems, you don't have to start over with AI. You can choose the projects where AI will have the most impact. And Outsystems gives you the platform to make it agentic on your terms with complete control every step of the way. So you can trust AI with the mission critical systems that run your business. Own your agentic future. Learn more at outsystems.com/nyt. That's outsystems.com/nyt.

• 31:52Speaker 06

  Your sales team is at risk of missing quota. Don't panic, just ask Rippling AI. Built on your real time people and business data, Rippling AI can pull metrics from Rippling and your CRM into a meeting ready dashboard showing quota attainment, headcount trajectory and revenue to quota by region. In seconds, you'll see what's behind your quota risk and fix it before it's missed. Head to rippling.ai/hardfork to get the only AI built to give you full visibility and take complex actions across your whole business. That's R I P P L I N G dot AI slash hard fork. Sign up today.

• 32:28Kevin Roose?

  So Casey, what do you know about design?

• 32:30Casey Newton?

  I mean, I know that it's something I hire other people to do for me. This is really not my forte.

• 32:36Kevin Roose?

  Well, there is an app in the design world called Figma.

• 32:39Casey Newton?

  Ah, yes.

• 32:40Kevin Roose?

  Which is sort of like, it's like an app that, you know, people who build apps and services use to sort of design them. And the analogy I've heard used is that Figma is to Photoshop as Google Docs is to Microsoft Word. It's kind of like a more collaborative online way of doing designs, especially with a team.

• 32:59Casey Newton?

  Yeah, when it first came out, people started calling it multiplayer Photoshop. Because in addition to running in a browser, which was like a cool enough party trick on its own, multiple people could work on the same file at the same time, which at the time was really a revolution.

• 33:11Kevin Roose?

  Yeah, it was a fast-growing product. Designers really loved it. And then something happened, which was that in 2022, Adobe announced that it was going to buy Figma for $20 billion.

• 33:21Casey Newton?

  Making it one of the biggest acquisitions of that year.

• 33:24Kevin Roose?

  Yeah, so Adobe is this sort of giant in the world of design. They make Photoshop and Premiere and all these other tools that designers and film editors and other people use for work. And this deal was a big deal in Silicon Valley, and it also caught the attention of regulators who said, wait a minute, this sort of looks like a big company kind of acquiring a company that might potentially compete with it. And maybe that's going to harm competition and innovation in this market. And so the companies sort of pulled back from the deal and it fell through.

• 33:55Casey Newton?

  Yeah, and at that point, you might say that the deal for Adobe was just a Figma of its imagination.

• 34:02Kevin Roose?

  That's pretty good.

• 34:04Casey Newton?

  Thanks.

• 34:05Kevin Roose?

  So why are we talking about this story now? Well, I've been sort of fascinated with this whole Figma story because one of the biggest changes in the tech world in the past few years is not actually what's happening in Silicon Valley, it's been what's been happening in Washington. We've talked on this show with Lina Khan, the Chair of the FTC, and Jonathan Cantor, the Assistant Attorney General, who's sort of in charge of their antitrust enforcement. And both of them have sort of made the point that one of the things that harms innovation and competition in the tech industry is this pattern of small startups growing very quickly and then getting acquired by these big tech companies who are sort of afraid of competition.

• 34:44Casey Newton?

  Effectively knocking pieces off the chessboard to ensure that they will always win.

• 34:48Kevin Roose?

  Exactly. And so their point is that, well, if you want to increase competition in the tech industry, you have to actually make it harder for these big tech companies to acquire smaller startups, especially when there's sort of a direct competitive threat to the bigger company. And so this Adobe-Figma deal, I would say, sort of one of the first major test cases for what happens if you actually just tighten the rules on these kinds of acquisitions and make it harder for big tech companies to acquire small, promising startups? And how does that sort of ripple throughout the tech industry? And so that's what I wanted to ask my colleague Erin Griffith about today. Erin covers startups and venture capital, and she just wrote a big story about this Figma-Adobe deal falling through and what it sort of means for Silicon Valley as a whole.

• 35:32Casey Newton?

  Let's bring her in here.

• 35:36Casey Newton?

  Erin Griffith, welcome back to Hard Fork.

• 35:38Speaker 07

  Thanks for having me.

• 35:39Casey Newton?

  So this deal between Adobe and Figma fell through in December. You've just written a story about sort of what happened at Figma after this deal collapsed. Just tell us what you found in your reporting.

• 35:53Speaker 07

  Yeah, so I talked to just about everyone that I could, employees, former employees, soon to be former employees, investors, kind of people from all sides of the spectrum. And the main thing is that there's just relief that the deal is over. It's been a year and a half or more of just kind of being in limbo, which for a startup is so hard. Momentum is so important. You have to be constantly growing and like there are a bunch of employees that are potentially millionaires that don't actually know if they're going to see those millions or not. And so there's a little bit of relief that it's just over, but there's also, you know, these questions of like, okay, well, when will I see those millions or will I? The company's in kind of a weird place right now.

• 36:36Speaker 07

  They kind of had this like really rough year and a half where now they're kind of picking up the pieces.

• 36:40Casey Newton?

  Right. So it's obviously been a rough time at Figma. There are a lot of people who thought they were going to be rich this year, who are either going to be rich potentially a long time from now or maybe even never. But your story also goes into a different kind of shockwave that this sent through Silicon Valley, which is that acquisitions are a big part of how this town works, right? And that acquisitions are how money and talent get recycled into the ecosystem. It provides sort of a launching pad for all the next generation of startups. So how has Silicon Valley reacted in general to this news?

• 37:14Speaker 07

  I mean, I'm actually kind of surprised by the reaction a little bit. I've been asking investors about this for like five years. I don't know if you guys remember.

• 37:21Casey Newton?

  Wait, you knew that the Figma acquisition wasn't going to go through?

• 37:24Speaker 07

  Yes. And people are confused and yeah, no. So I don't know if you guys remember, like Harry's, the razor company sold that, but that deal got kind of broken up. Plaid, the payments company we're supposed to sell to Visa, that got broken up. Both of those companies were able to raise money later, but they still haven't exited. So the investors are painting an optimistic picture, but like it could be a really long time before. They exit and most of the investors I talked to, they don't seem that concerned. They're like, oh, well, yeah, Apple, Amazon, the big tech companies that are under antitrust scrutiny. They're on the sidelines, but there are other buyers. And I kind of think it's a bigger deal than that because, I mean, just to take one stat that I found, like 90% of exits in 2019, which was a big year for IPOs, were acquisitions, not IPOs.

• 38:09Speaker 07

  And like the IPOs are where the big money exits come from, but like the vast majority of startups end up selling and that's how.

• 38:15Kevin Roose?

  I think this is a point that we should sort of take a beat on because I think this is something that may not be obvious outside the tech industry. When venture capitalists put money into a startup, there are kind of two things that they're going for. One is that the company becomes huge and goes public, right? And then they can pull money out, then their shares get converted into shares of the public company and they get to sell and they get a return from that. The other, and what you're saying, the sort of more frequent outcome for a venture backed startup is that it would get acquired and then the investors who, you know, supported the startup from the beginning get paid as part of that acquisition. So those are kind of the classic exits in venture capital. And it sounds like, you know, beyond just this Figma case, it sounds like the exit strategy

• 39:00Kevin Roose?

  of let's sell this startup to a bigger company is becoming much more challenging.

• 39:04Speaker 07

  Totally, yeah. And like, that is the way this industry works. You know, the venture capitalists get that money back and then they use that to raise their next fund because their investors are happy and they want to give them more money. Angel investors who are kind of the first people who put money into startups, they use that and they like invest in their friends' companies. And it's all like kind of incestuous, but it's like a circle of life that perpetuates the new creation of startups. You know, I talked to one investor in the story, Adam Nash. He used his earnings from the startups he worked at, like LinkedIn, to invest in more than 130 of his friends' startups. And like, that's how this works. And so all these Figma employees that would have now become angel investors are not. And, you know, a lot of these venture investors, they kind of like made their reputations like,

• 39:47Speaker 07

  Figma, I got this big winner, I got this $20 billion exit. And like, that's not the case anymore. I mean, I talked to one investor, he was set to get paid on Figma's exit from six different entities. So like, that's like how incestuous this is and that's actually just what makes it work here. And so, you know, if those acquisitions dry up, that can, you know, have a really. Big chilling effect on an industry that's already outside of AI struggling.

• 40:11Kevin Roose?

  So let's talk about why, what change that these acquisitions are just having a harder time getting through.

• 40:16Speaker 07

  I mean regulatory.

• 40:17Kevin Roose?

  Yeah, which regulators?

• 40:19Speaker 07

  Yeah, so I mean, obviously, there's Lena Khan and FTC has been scrutinizing every single deal, including now minority deals. I mean, she just said that she's looking.

• 40:27Kevin Roose?

  Minority deals, meaning a deal in which an investor would buy a minority stake in the company.

• 40:33Speaker 07

  Yeah, so Microsoft's investment in OpenAI and Google and Amazon's investments in Anthropic, the AI companies, those are minority stakes, but they're now investigating those for potential antitrust problems. So that's like going even a step further. But what killed the Adobe-Figma deal was the British regulators, the CMA, which is the authority there, they essentially looked at this deal and said, well, Figma and Adobe could potentially have been competitors down the line. Therefore, this deal is eliminating competition and we think that it's bad and stifling innovation. So they're taking an even stronger stance against this. So yeah, it's, as one investor told me, it's raising the bar for even engaging in acquisition talks

• 41:15Speaker 07

  for a startup, because this is a huge risk that the deal might not actually go through. And, you know, one thing that Figma did that's really smart, it's raising the bar for breakup fees. Figma is getting paid $1 billion as a breakup fee from Adobe. So that's, you know.

• 41:29Kevin Roose?

  Breakup fees are fees to get paid when like a deal doesn't happen.

• 41:31Speaker 07

  Right.

• 41:32Casey Newton?

  This has always been my dream, by the way, is to get paid $1 billion for just nothing happening.

• 41:37Kevin Roose?

  You should know that the hard fork breakup fee is $1 billion from you to me.

• 41:42Casey Newton?

  Oh, no.

• 41:43Kevin Roose?

  So if this show breaks up, you owe me $1 billion. That was in your contract.

• 41:49Kevin Roose?

  So I think a lot of people listening to this might be saying, you know, what's the big deal? Like, we have this regulatory regime that's gotten tougher on startups being acquired by tech giants. You know, so far, it seems like from your story, the people who are most affected by that are like Figma employees who don't get their, you know, their huge paydays from this deal. And you have, you know, venture capitalists who are unhappy that maybe they can't as easily sell their portfolio companies to large acquirers and get paid. Why does this matter sort of beyond the world of tech investing? What are the stakes of this new regulatory environment for average users?

• 42:25Speaker 07

  Yeah, I mean, I guess it impacts the incentives for people who are maybe interested in starting a company and, you know, think that they could follow in the footsteps of many of our most famous and powerful and frankly admired business leaders. And often the way to do that is to build a company and sell it to a larger one. It impacts the way that this kind of innovation is going to be funded. And I'm not saying it's necessarily a good or a bad thing, but it's definitely going to change the way the system works. And I think we'll be seeing the ramifications of that playing out for a long time.

• 42:56Casey Newton?

  I can tell you, when I started Platformer, it was with the goal of selling it to Adobe for $20 billion. And now I'm like, in this regulatory environment? I don't think so, honey. So that's going to be a challenge. But let me throw a couple of counter examples in here, Aaron, because while, you know, everything that you said is true, at the same time, Microsoft was somehow able to buy Activision Blizzard for almost $69 billion, right? Which was just a clear case of a giant company getting even more giant. So that seems to me to suggest that at least some window is still open. And I have to say, when I saw that Adobe was acquiring Figma, my first thought was, the cost of Photoshop is going to go up. Like, Figma is going to get folded into the Adobe Creative Suite and the price is going to go up.

• 43:38Casey Newton?

  And so you don't even have to be a hipster antitrust person to find an excuse to block this deal. You could just say, this just kind of seems like maybe we shouldn't allow it. So is the Figma deal maybe an outlier in this whole thing?

• 43:50Speaker 07

  No, I don't know about that. But I will agree with your point that there are a lot of Figma customers that were kind of celebrating this. And I think that just in general, even if you're not a Figma user, a lot of people complain about the power of big faceless corporations and how, you know, they kind of are constantly finding new ways to screw customers. So like, in general, I can see the concerns and like, why you would prefer that this product that you love, that's like, you know, pretty cheap and not corporate and tries to do right by its customers, stays independent. So like, I think there were a lot of Figma customers that were celebrating this and a lot of people who maybe don't like certain Adobe policies or the way that they sell or package their products. So yeah, I can definitely see an argument for that.

• 44:34Kevin Roose?

  Do you hear from any founders who are thinking about their sort of exit strategies differently in light of these new regulatory pressures? I mean, one complaint that you'll hear from venture capitalists about the aggressive regulation of tech mergers and acquisitions is that, you know, maybe someday the next Kevin Systrom and Mike Krieger don't start Instagram because they know that even if they succeed beyond their wildest dreams, they're not going to be able to sell their company to Facebook for billions of dollars. You might sort of see a drop off in startup founders deciding to take a risk on something new because one of these two classic avenues for getting rich on a startup is now not closed, but harder.

• 45:16Speaker 07

  Totally, and I've heard that argument before, not ever from a founder, though. It's more from like the commentators or VCs. And it's kind of BS in a way because the other thing that they'll turn around and say is like, well, no one aims to sell their company. Their goal is always to go public. And, you know, that's, people start it with the sort of the biggest possible dream and maybe selling is like the second best. And I think there's always going to be like people who are wildly ambitious and trying to start new companies and hoping to make it rich that way. So I don't see us stopping that. And, you know, investors being always very optimistic about everything, that's kind of like their job, they see another, a whole world of other potential buyers who maybe

• 45:59Speaker 07

  aren't as well-known as Google, Apple, Amazon, that are now taking advantage of this moment, like Salesforce, for example, companies that are big, have a lot of money and haven't quite caught the regulatory scrutiny yet that some of the biggest ones have.

• 46:13Kevin Roose?

  Yeah. And do we think that there are going to be any kind of new exit strategies for startups? I mean, if you had the sort of classic model of you either go public or you get acquired as sort of the ways to succeed and get rich during a startup, are founders, are people at these companies and investment firms thinking about other ways that they could potentially cash out?

• 46:35Speaker 07

  I'm glad you asked that, actually, because I've been having this conversation with a lot of people. The secondary market is very hot right now.

• 46:43Kevin Roose?

  Explain what that is.

• 46:44Speaker 07

  And the secondary market is essentially a fancy way of saying it's a private stock market. So startups own shares that are private and only sophisticated investors like venture capital investors can buy them. However, there is a growing market of people who are buying and selling shares in privately held stock. It's called the secondary market. And it got a lot of attention right before Facebook's IPO because people were trading private Facebook stock before it went public. And over the last decade or so, it has gotten bigger and bigger. And so there is generally a market for people to buy shares in these private startups before they go public. And that is becoming increasingly robust. And so that's a way that some investors are getting, being able to cash out before an

• 47:29Speaker 07

  IPO can happen.

• 47:30Kevin Roose?

  Yep. And of course, Kevin, don't count out that other classic way for a startup founder to make money: crypto scams.

• 47:39Kevin Roose?

  And what do you think this means for the tech giants? Because historically, buying smaller startups has been a way to kind of like extend their relevance. I'm thinking of, you know, Facebook buying Instagram and WhatsApp, which sort of became the next phase of that business's growth. So if you are a Facebook, if you are an Amazon, if you are a Google, are you really worried about it now being tougher to acquire some of these fledgling, promising young startups?

• 48:06Speaker 07

  Definitely. I mean, that is essentially how these companies stay relevant, is they try to build something innovative in-house. They're too clunky and slow-moving and out of good ideas. So they buy something and hope that that infuses them with new life and they, you know, their stock lives to see another day. That is not happening at the big tech companies at all. And so I would be nervous if I were them. I mean, Microsoft has obviously found a way around that by investing in OpenAI without buying it outright and, you know, being very careful to not own more than 50% of the company. But even that's now being scrutinized. So yeah, it's definitely a signal to big tech companies that they need to start building their most innovative stuff in-house.

• 48:50Kevin Roose?

  Yeah, and I should say like that is actually what the regulators want to happen, right? That is part of the hypothesis here is that if you force, you know, if you make it harder for tech giants to kind of gobble up smaller potential competitors, then it will actually result in more innovation overall because the big companies will have to stay ahead of the small startups and the small startups will sort of be able to flourish independently without sort of getting gobbled up by these larger institutions.

• 49:17Speaker 07

  Yeah, I mean, that's kind of, if you read the CMA's like investigation of Adobe, they had this company, or they had this project that was ongoing called Project Spice. And it was, it had other names too, but they mostly called it Project Spice.

• 49:31Kevin Roose?

  I think I saw that on Cinemax when I was a kid.

• 49:33Speaker 07

  And it was, and it was essentially going to compete with Figma. And they canceled it one week before the acquisition of Figma was announced. And they said, you know, in their responses, like, "This was never going to compete with Figma, and it was a failure, and we canceled it because it was really bad, and we could never compete with Figma." But like, that kind of makes the point for the regulators, too. Like, you couldn't compete, and so you bought this thing instead. So, yeah, it's like, that's a perfect example of what you're talking about.

• 50:13Kevin Roose?

  Yeah. I mean, as you're describing what happened at Figma, I'm sort of thinking like, isn't this just a case study of how regulation works and how it should work? Like, isn't this a success story for regulators? Because it sounds like what you're describing is a company in Adobe that now has to sort of develop its own independent strategy for competing with Figma, which it can no longer buy. You're also describing a Figma that now has to think about how to compete with Adobe rather than just being sort of swallowed up by it. So it seems like there is more competition in that particular market than there would have been if this deal had been allowed to go through. So isn't this just kind of proving what regulators have been saying all along?

• 50:56Speaker 07

  Well, that was very well articulated, but I think I have to step in here and give the company's perspective on that, which would strongly disagree with you. You know, they've obviously fought this every step of the way and they, you know, expressed their disagreement with the British regulators, findings that they were taking competition out of the market. Um, you know, they believe that they were not competitive. But yeah, your point is, is well taken that like, yeah, regulators dug into this, concluded that these two companies would have been competitors had they, had they not decided to merge and said, we want to block this. And then they asked for some really absurd remedies. They were like, okay, well, you guys can merge, I guess, if like Adobe spins off Photoshop

• 51:40Speaker 07

  or Illustrator or Figma spins off Figma Design, which is like essentially all of Figma. And so of course the companies were like, no. So yeah, I mean, essentially, yes, this is regulation working. I think a lot of people in the tech industry, obviously, including Adobe and Figma, would say it's like an egregious overreach and they don't like the way they're interpreting the law.

• 51:58Kevin Roose?

  So Erin, you've written, among other things, about the culture of Figma, which, whose employees, I believe, call themselves fig mates. How is the culture changing in response to this deal falling apart?

• 52:09Speaker 07

  I mean, it's probably the same. There, what I, my understanding of the culture of Figma is very earnest. They, they like to sing. They have a theme song, or they had.

• 52:20Kevin Roose?

  Wait, they have a theme song?

• 52:21Speaker 07

  They had a theme song that.

• 52:23Kevin Roose?

  How does it go?

• 52:24Speaker 07

  The chorus was, "Figma, I dig ya. Design used to be such an enigma." I'm not going to sing it. And it had a rap verse that referenced Adobe at one point. Um, 10 or, oh, I think it said 10 or 15 years from now, people are going to say, who the heck's Adobe? Figma's here to stay.

• 52:42Kevin Roose?

  Wait, they made a, they made a software diss track?

• 52:45Speaker 07

  Yeah. I mean, I don't think it was meant for public consumption. I did, I saw a video of it, but. Unlike their musical, which they put out in the quarantine, which is on YouTube and is...

• 52:58Casey Newton?

  Required viewing for Hard Fork listeners. Wait, there's a Figma musical?

• 53:02Speaker 07

  Yes, it is very elaborate. It's like kind of about quarantine, the early days of quarantine. I actually had to turn it off because I was like having flashbacks to like, you know, all the jokes that we made about, oh, we're all stuck inside and like how traumatizing that was for everyone. Anyway, so yeah, they're a little dorky. This is not like the Uber culture of like, let's like be hardcore and step on toes and just like...

• 53:23Casey Newton?

  Yeah, Uber also actually had a musical. It was a... I think it was called, "It's Not a Crime If You Don't Get Caught." Yes. Yeah. That was the showstopper. Yeah, the big number of closing act one.

• 53:38Casey Newton?

  And FTX actually had a musical. It was called Scamilton. Yeah, it was a great one.

• 53:45Casey Newton?

  Aaron Griffith, thanks for coming on Hard Fork.

• 53:46Speaker 07

  Thanks for having me.

• 53:52Casey Newton?

  When we come back, it's time for our new segment, System Update.

• 54:13Speaker 05

  It's time to innovate at the speed of AI without compromising control. That's OutSystems. Take a look at what happened with one global retailer during a period of massive expansion. They used OutSystems to build AI agents to automate invoicing and translation processes, shifting 95% of their daily store operations to a new digital app, saving thousands of hours of manual work. OutSystems gives you the flexibility to deploy how you want and the governance to do it safely. Own your agentic future. Learn more at outsystems.com/NYT. That's outsystems.com/NYT.

• 54:48Speaker 08

  Today's AI builders need infrastructure that scales with their ambitions. Iron is a vertically integrated AI cloud platform, delivering data centers, compute and software for AI training and inference, underpinned by its expansive portfolio of land and grid-connected power in renewable-rich regions globally. Iron's AI cloud gives direct access to Nvidia GPUs, scalable from hundreds to thousands. For builders, that means direct access to AI compute, rapid scaling and predictable pricing. Iron, AI cloud for the next big thing.

• 55:21Speaker 06

  Your sales team is at risk of missing quota. Don't panic, just ask Rippling AI. Built on your real-time people and business data, Rippling AI can pull metrics from Rippling and your CRM into a meeting-ready dashboard showing quota attainment, headcount trajectory and revenue to quota by region. In seconds, you'll see what's behind your quota risk and fix it before it's missed. Head to rippling.ai/hardfork to get the only AI built to give you full visibility and take complex actions across your whole business. That's R-I-P-P-L-I-N-G dot AI slash hard fork. Sign up today.

• 55:56Casey Newton?

  All right, Casey, we got a new segment this week and it's called "System Update."

• 56:01Kevin Roose?

  Can we hear the theme?

• 56:05Kevin Roose?

  Kind of sounds like the system just threw up.

• 56:07Casey Newton?

  But I like it. Now, listen, Kevin, every week on the show, we bring you the most fascinating stories from in and around the world of tech and the future, but we don't always return to those stories to tell you what happened after we shut up.

• 56:18Kevin Roose?

  Yes, sometimes we do not actually get the last word on a subject.

• 56:21Casey Newton?

  Devastating.

• 56:21Kevin Roose?

  Events continue to unfold. And so this is a segment where we take things that we've talked about on the show before and tell you what's happened since we talked about them.

• 56:29Casey Newton?

  That's great. Well, let me ask you a question, Kevin. Remember when we talked about those ancient scrolls?

• 56:33Kevin Roose?

  Yes.

• 56:33Casey Newton?

  Whatever happened to those?

• 56:34Kevin Roose?

  God, this is one of my favorite stories from last year.

• 56:36Casey Newton?

  Let's get into it.

• 56:37Kevin Roose?

  So Casey, back in October, we interviewed a man named Brent Seals on the show. Brent, as a reminder, was a scientist who'd been working on a project involving some ancient scrolls. And these were scrolls that had been buried and burned in the eruption of Mount Vesuvius roughly 2,000 years ago.

• 56:56Casey Newton?

  They were the original Dooms. Scrolls.

• 56:58Kevin Roose?

  Oh, come on.

• 56:59Casey Newton?

  That's pretty good. You have to.

• 57:00Kevin Roose?

  That is pretty.

• 57:01Casey Newton?

  OK.

• 57:02Kevin Roose?

  So Brett told us about this effort that he'd been embarking on to decode these scrolls, basically by using kind of computer vision and AI to be able to read what was inside them without actually opening them, which would destroy them. And he, in combination with some investors and entrepreneurs in the tech industry, had set up something called the Vesuvius Challenge, which was basically a series of prizes that would be given out for deciphering fragments of these ancient scrolls.

• 57:32Casey Newton?

  Right. And not to be confused with the Vesuvius Challenge on TikTok, which is where you try to escape an erupting volcano.

• 57:37Kevin Roose?

  Exactly.

• 57:38Casey Newton?

  Yeah.

• 57:38Kevin Roose?

  So when we talked to him back in October, this contest had produced some intermediate results. There was a college student, Luke Ferritor, who had been able to actually detect one word from the ancient scrolls that translated to the ancient Greek word for the color purple.

• 57:56Casey Newton?

  Mm hmm. Color purple. Great movie. Great book.

• 58:00Kevin Roose?

  Yeah. So back in October, they had sort of made this intermediate prize award to this researcher who had come up with this one word. But there was still this idea of a grand prize for someone who could decipher much more of these scrolls. And Brent and his team, they didn't know whether anyone would actually be able to do that. But just this week, the organizers of the Vesuvius Challenge announced that they had, in fact, awarded a grand prize to a team of three individuals. Youssef Nader, who is an Egyptian PhD student in Berlin. Luke Ferritor, who is a 22-year-old college student who actually won that prize that we mentioned the last time we talked about this, discovering the word purple.

• 58:41Kevin Roose?

  And Julian Schilliger, who is a Swiss robotics student based in Zurich. They will share a $700,000 prize as a reward for deciphering a much larger fragment of these ancient scrolls.

• 58:55Casey Newton?

  All right. And I would like to know, Kevin, what did the scrolls say?

• 58:58Kevin Roose?

  Well, so we still don't know, like, exactly who wrote these or what they're about in their entirety. But the passages that historians and papyrologists have been able to decipher are basically talking about pleasure. The author of one, I'll just read one of these passages. It says, "As to in the case of food, we do not right away believe things that are scarce to be absolutely more pleasant than those which are abundant."

• 59:24Casey Newton?

  Hmm. Go off, king.

• 59:28Kevin Roose?

  Yeah, so we don't know exactly who wrote this. There are some theories that it might have been Philotimus, who is an ancient Greek philosopher and writer who is associated with the Epicurean movement.

• 59:37Casey Newton?

  Of course.

• 59:38Kevin Roose?

  But basically, this is a text. You know, the organizers of The Challenge compared it to like a 2,000-year-old blog post. So in addition to referring to the color purple, apparently this text also makes some reference to capers, like the thing that you put on like a lox bagel.

• 59:52Casey Newton?

  Not crimes.

• 59:53Kevin Roose?

  Yes, not crimes. But basically, these texts are what we know of them so far. They are about pleasure and how to live the good life. And they appear to be from someone who's sort of sympathetic to or associated with the Epicurean movement, kind of dunking on this other group of philosophers, the Stoics, who believed that virtue was all that mattered. Epicureans were much more into sort of pursuing pleasure as kind of the highest good. And so, you know, some historians who have looked at this text think that this was basically an Epicurean philosopher kind of dunking on Stoics.

• 1:00:25Casey Newton?

  This was history's answer to the quote tweet, basically, is what we're finding in these scrolls.

• 1:00:30Kevin Roose?

  Exactly. So The Challenge has awarded this $700,000 prize to this group of three researchers, and they are launching yet another challenge. The organizers said that in 2024, their goal is to go from deciphering 5% of one scroll to 90% of all four scrolls that they have scanned and to lay the foundation to read all 800 scrolls in this collection.

• 1:00:52Casey Newton?

  And those are their scroll goals. So you now have those.

• 1:00:55Kevin Roose?

  They're scrolls, if you will.

• 1:00:56Casey Newton?

  Scrolls, if you will. Now, Brent worked on this for 20 years, which is so incredible that you could spend two full decades of your life working on something, and then it pays off. Did you get in touch with him? What did he have to say about this?

• 1:01:08Kevin Roose?

  Yeah, so I emailed with Brent this week after this prize was announced, just sort of asking like, how do you feel that this project that you've been working on for 20-some years has finally sort of paid off? And he wrote back, he said, "I can tell you I've never had a sweeter, more satisfying feeling than knowing we've done it. We've demonstrated we can read Herculaneum scrolls with the right tech, and that my 20-year pursuit will be seen as successful. The fact I waited two decades for this moment almost certainly makes it all the more delicious."

• 1:01:39Casey Newton?

  That is fantastic. Almost as delicious, I must say, as grapes and capers. Seriously, congratulations to Brent and everyone who solved that challenge. And we look forward to seeing what AI reveals on the rest of these scrolls.

• 1:01:51Kevin Roose?

  Yes. Congrats to the team. It's a huge feat. I love this story. I think it's one of the most sort of feel-good stories in tech over the past year. And I think it is also just a very clear example of how AI can actually do things that I think we would all agree are pretty good.

• 1:02:05Casey Newton?

  All right, let's move on to our second update.

• 1:02:07Kevin Roose?

  Yes. This one is about Bard, the AI chatbot from Google.

• 1:02:12Casey Newton?

  And without knowing anything about this update, Kevin, I think what I would just say is Bard, one of the great Google brands, and I hope it never goes away.

• 1:02:18Kevin Roose?

  I've got some bad news around for you on that front. Because this week, Google announced that it is ending the use of the Bard name and that Bard, the chatbot, will be rebranded as Gemini, which is also the name of their latest and greatest AI model.

• 1:02:34Casey Newton?

  And our mutual Zodiac sign.

• 1:02:36Kevin Roose?

  Yes.

• 1:02:37Casey Newton?

  So Kevin, you're telling me that this Bard was not immortal?

• 1:02:39Kevin Roose?

  Oh, come on.

• 1:02:40Casey Newton?

  Shakespeare is safe.

• 1:02:43Kevin Roose?

  So Google this week announced that they are rebranding Bard. It will now be known as Gemini. They are also launching Gemini Advanced, which is a paid version of Gemini that will give people access to Gemini Ultra, which is their kind of largest and most capable AI model. Gemini Advanced will be available through the new Google One AI Premium plan.

• 1:03:07Casey Newton?

  Kevin, I just have to ask real quick. Are you having a stroke?

• 1:03:10Kevin Roose?

  I might be. So they also announced that Duet AI will become Gemini for Google Workspace and Google Cloud.

• 1:03:18Casey Newton?

  No, no, this is real.

• 1:03:20Kevin Roose?

  This is real? So basically, what this boils down to is that they are changing some names. They are also introducing a paid product that will, for 20 bucks a month, give you access to their top-of-the-line AI models. And they are also changing a lot of their different AI products, like Google Assistant, which is sort of their version of Siri on Android devices. You can now run Gemini on that, too. So Casey, I want to tell you what I think is a true sentence.

• 1:03:51Casey Newton?

  OK, let's hear it.

• 1:03:52Kevin Roose?

  If you used to use Duet AI for Google Workspace, instead, you will now use Gemini for Workspace and Google Cloud. And if you upgrade your existing Google One subscription to a Google One AI Premium subscription, you will now get access to Gemini Advanced, which is powered by Gemini Ultra 1.0.

• 1:04:15Casey Newton?

  This is one of the richest and most powerful companies in the entire world. They can hire any marketer they want, any brander they want. This is what they came up with. If you work for Google and you worked on any of this, please message us. We won't even use it on the show. I just need to know for my own knowledge how this happens. I'm someone who cares about this so much, and I could not parse even 5% of what you just said.

• 1:04:38Kevin Roose?

  Yeah, it is truly a hilarious combination of product names and just an absolute chaotic nightmare to try to make sense of it. I went to this briefing where they rolled out these changes, and I was just sitting there taking notes, trying to like draw flow charts of the various names and how they all relate to each other.

• 1:04:55Casey Newton?

  Yeah, it would be easier to pass calculus than to just repeat what you just said to me.

• 1:05:00Kevin Roose?

  So yes, all of this is sort of silly because they're just changing the names of a bunch of stuff. But I think what this boils down to is that Google has now built this model that they say is state of the art, and they are trying to roll this model out to as many customers as possible. So they are going to start putting it into things like Google Docs and Google Sheets. If you use Gmail, you will have the option of maybe using this model to write some of your emails or get you started on some of your emails. So this sort of effort by Google to jam generative AI into as many products as possible seems to be picking up pace.

• 1:05:36Casey Newton?

  Yeah, and we did talk last year about the fact that this is Google's big swing. This is their effort to catch up to the model that OpenAI has released. And I will say, I am really looking forward to putting this, I believe it is called Gemini Ultra, through its paces to see how well it stacks up.

• 1:05:53Kevin Roose?

  Yeah, so it's it's a little interesting the way they rolled this out. So instead of being a standalone product, you know, this new tier of Gemini that gives you access to Gemini Ultra, it runs through Google One, which if you have ever encountered Google One, it's probably because you ran out of space on your Gmail account and you had to like pay them to upgrade your account. That is what Google One is.

• 1:06:15Casey Newton?

  Right.

• 1:06:15Kevin Roose?

  So if you need more space in your Gmail, you can buy more space through Google One. And now, when you go into Google One, you'll also be able to buy this new AI premium plan that gives you access to their high-end model.

• 1:06:27Casey Newton?

  That's great. I just picturing listeners like listening to all this information and then just like swerving off the road because like their their brain completely shuts down.

• 1:06:37Kevin Roose?

  Yeah, and like what's most amazing about this is they are pitching this as a simplification of their existing product names.

• 1:06:44Casey Newton?

  Yeah.

• 1:06:45Kevin Roose?

  And so I would just say the simplification effort is not complete.

• 1:06:49Casey Newton?

  We're not, yeah, that's.

• 1:06:52Kevin Roose?

  But I'm a little sad because Bard was actually one of my favorite of the like AI chatbot names, and I'm just sad to see it go.

• 1:06:58Casey Newton?

  Yeah, RIP Bard. If you miss Bard, you can actually play a Bard in Baldur's Gate 3, one of the great video games for 2023. That's my advice. Wow, Kevin, that was really one of the biggest updates we've ever heard on this show. And by biggest, I mean most confusing. So let's move on. Is there a final update?

• 1:07:15Kevin Roose?

  Yes, TikTok.

• 1:07:16Casey Newton?

  Oh, yes. Kevin, what is happening with TikTok?

• 1:07:18Kevin Roose?

  So we've talked a lot about TikTok on this show. Most recently, we talked with John Herman about TikTok's shopping features and how they are starting to change the platform.

• 1:07:26Casey Newton?

  You may remember we had recently gone viral on TikTok.

• 1:07:27Kevin Roose?

  Yes. And now there's another big change on TikTok, which is that the Universal Music Group, one of the biggest music labels in the world, has pulled its licenses from TikTok.

• 1:07:38Casey Newton?

  Now, do they have any of the good artists?

• 1:07:39Kevin Roose?

  So they do. They have many, many artists, including Taylor Swift, Bad Bunny, Drake, and Billie Eilish. And so if you made a video that included a Taylor Swift song or a Drake song in it, that video now has had the music stripped out of it. So you can still watch the video, but now it just is silent in the background.

• 1:08:00Casey Newton?

  There's just, yeah, we were sort of reinventing silent films just through copyright measures.

• 1:08:06Kevin Roose?

  Yes, so Universal Music Group and TikTok have been renegotiating their contract that allows TikTok to basically use these artists' music in videos. These talks did not go well, and ultimately, Universal Music Group published an open letter saying that TikTok was trying to, quote, "build a music-based business without paying fair value for the music." They're also mad that TikTok is allowing its platform to be flooded, they say, with AI-generated recordings, and they're worried about the hate speech bullying and they're worried about the hate speech bigotry, bullying, and harassment that UMG artists are met with on the platform. So now this contract has expired, and music from Universal Music Group is unavailable on TikTok.

• 1:08:51Kevin Roose?

  So what do you make of this?

• 1:08:52Casey Newton?

  Well, so this really surprised me, because the one thing I can tell you for sure is that TikTok can afford whatever UMG was asking for, right? Like TikTok, which is owned by ByteDance, is doing very well financially. They could afford to pay the rate here. When I first saw this, I thought, okay, this is some hardball negotiating. I do expect that UMG will be back on TikTok in some relatively short period. But at the end, they did say, okay, we're actually going to go to the trouble of pulling the rights, and that did surprise me. What I wonder is, how big a deal is the AI thing? And was that maybe the one of the top reasons that this all happened? Or is it really just about dollars and cents?

• 1:09:31Kevin Roose?

  My hunch is that it's mostly about dollars and cents. I mean, TikTok, it's still a huge part of the music industry. We've seen over the past few years, you know, artists like Olivia Rodrigo, who blow up on TikTok and then go on to become huge celebrities. We've also seen TikTok sort of bringing songs out of obscurity. You remember the Fleetwood Mac song that got popular after it was used in a TikTok video?

• 1:09:54Casey Newton?

  I would say it was repopularized.

• 1:09:56Kevin Roose?

  Repopularized, yeah. Gen Z sort of rediscovered it after it went viral on TikTok. So TikTok and the music industry have been deeply intertwined for years now, and this just seems like maybe a sort of disagreement about how much music is worth on TikTok.

• 1:10:12Casey Newton?

  Well, hopefully this will all get smoothed over. But in the meantime, Kevin, I'll tell you what I'm hoping for out of this situation, which is that we see a lot more public domain music going viral on TikTok. I would like to see "She'll Be Coming Round the Mountain", maybe sort of hit number one on the charts this week, and really send a message that, you know, it's not all copyrighted music all the time, and that there's some really beautiful songs that have, you know, recently come out of copyright.

• 1:10:35Kevin Roose?

  Yeah, if you are, you know, a Gregorian chant artist, this is your moment to shine on TikTok.

• 1:10:41Casey Newton?

  Get to chanting.

• 1:10:44Kevin Roose?

  And that's our system update.

• 1:10:54Speaker 05

  Companies are pouring money into AI, but few leaders can explain how those AI investments work with the systems you already have. You know, the ones that actually run your business? Without Systems, you don't have to start over with AI. You can choose the projects where AI will have the most impact, and OutSystems gives you the platform to make it agentic on your terms with complete control every step of the way. So you can trust AI with the mission-critical systems that run your business. Own your agentic future. Learn more at outsystems.com/NYT. That's outsystems.com/NYT.

• 1:11:29Speaker 08

  Today's AI builders need infrastructure that scales with their ambitions. iRen is a vertically integrated AI cloud platform, delivering data centers, compute, and software for AI training and inference, underpinned by its expansive portfolio of LAN and grid-connected power in renewable-rich regions globally. iRen's AI Cloud gives direct access to Nvidia GPUs, scalable from hundreds to thousands. For builders, that means direct access to AI compute, rapid scaling, and predictable pricing. iRen, AI Cloud for the next big thing.

• 1:12:02Speaker 06

  Your sales team is at risk of missing quota. Don't panic, just ask Rippling AI. Built on your real-time people and business data, Rippling AI can pull metrics from Rippling and your CRM into a meeting-ready dashboard showing quota attainment, headcount trajectory, and revenue to quota by region. In seconds, you'll see what's behind your quota risk and fix it before it's missed. Head to rippling.ai/hardfork to get the only AI built to give you full visibility and take complex actions across your whole business. That's R-I-P-P-L-I-N-G dot AI slash hardfork. Sign up today.

• 1:12:38Kevin Roose?

  Hard Fork is produced by Rachel Cohn and Davis Land. We're edited by Jen Poyant. This episode was edited by Paula Schuman. We're fact-checked by Caitlin Love. Today's show was engineered by Alyssa Moxley. Original music by Alicia Baty-Tup, Marian Lozano, Rowan Nemestow, and Dan Powell. Our audience editor is Nell Galogly. Video production by Ryan Manning and Dylan Burgason. If you haven't already, check us out on YouTube at youtube.com/hardfork. Special thanks to Puywing Tam, Kate LaPrastie, and Jeffrey Miranda. You can email us, as always, at [email protected].

• 1:13:57Speaker 05

  Building and scaling AI today can feel pretty chaotic, mostly because teams are juggling too many fragmented tools at once. That's where OutSystems comes in. OutSystems gives you one unified agentic systems platform to manage your entire portfolio of agents, apps, and workflows. You get to innovate at the speed of AI, but with the security, governance, and control your enterprise actually needs. And the results are already there. A major manufacturer used OutSystems to build a multi-agent system that analyzes blueprints. And a global bank cut their development time by 60%. Own your agentic future today at outsystems.com/NYT.