E35: Stability AI’s Emad Mostaque and Slow Ventures’ Sam Lessin Discuss Investing in AI

In this episode, Nathan Labenz and Erik Torenberg host a discussion between Emad Mostaque, founder of Stability AI, and Sam Lessin, GP of Slow Ventures and former VP Product at Facebook. These influential technology thinkers tackle topics pertinent to evaluating whether AI is a good investment for venture capital. They talk about the attention and information economy, SaaS markets, how AI-created entertainment will impact society and our relationships, and AI disruption of highly-regulated industries. We're hiring across the board at Turpentine and for Erik's personal team on other projects he's incubating. He's hiring a Chief of Staff, EA, Head of Special Projects, Investment Associate, and more. For a list of JDs, check out: eriktorenberg.com. LINK FOR NATHAN'S LIVE EVENT IN SAN FRANCISCO: Friday June 16th: Why AI with Athena: https://v4p9mjmurpv.typeform.com/to/llRcJSF9 TIMESTAMP: (00:00) Episode Preview (01:17) Nathan explains why we hosted a conversation with Emad and Sam (08:25) Discussion kicks off: AI as an extender rather than a deck shuffler (12:01) Is AI a disruptive or sustaining innovation? (14:22) What is intelligence? (16:00) The attention and information economy (21:56) Storytelling, the stock market, and the economy (23:23) Sponsor: Omneky (23:50) AI in the media industry, Hollywood, and distribution (29:10) Startups vs incumbents (30:30) Disrupting highly-regulated industries (32:30) AI in education: an AI tutor for every child (35:15) AI as StackOverflow 2.0 for engineers (37:23) Regulatory Arbitrage (40:06) The oracle problem in AI (42:00) Apple as the weakest BigTech player in AI (44:30) AI in communities and the future of social media (46:55) Replika on Valentine’s Day and AI-generated porn (53:50) How does the market for inference shape up? Inference vs. cloud (55:08) Price of primary care in US in the next decade (55:38) What are the most likely incumbents to be disrupted? (56:55) Will BigTech continue to opensource? (58:56) Marriage rate in US in the future given AI (01:00:24) Can any AI leader sustain high gross margins in the future? (01:02:04) Stability’s business model TWITTER: @EMostaque (Emad) @Lessin (Sam) @CogRev_Podcast @labenz (Nathan) @eriktorenberg (Erik) SPONSORS: Shopify is the global commerce platform that helps you sell at every stage of your business. Shopify powers 10% of ALL eCommerce in the US. And Shopify's the global force behind Allbirds, Rothy's, and Brooklinen, and 1,000,000s of other entrepreneurs across 175 countries. From their all-in-one e-commerce platform, to their in-person POS system – wherever and whatever you're selling, Shopify's got you covered. With free Shopify Magic, sell more with less effort by whipping up captivating content that converts – from blog posts to product descriptions using AI. Sign up for $1/month trial period: https://shopify.com/cognitive Thank you Omneky for sponsoring The Cognitive Revolution. Omneky is an omnichannel creative generation platform that lets you launch hundreds of thousands of ad iterations that actually work, customized across all platforms, with a click of a button. Omneky combines generative AI and real-time advertising data. Mention "Cog Rev" for 10% off. This show is produced by Turpentine: a network of podcasts, newsletters, and more, covering technology, business, and culture — all from the perspective of industry insiders and experts. We’re launching new shows every week, and we’re looking for industry-leading sponsors — if you think that might be you and your company, email us at [email protected]. Download this episode as an mp3: https://chrt.fm/track/993DGA/traffic.megaphone.fm/RINTP2126380296.mp3?updated=1686676812 Music Credit: MusicLM

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

• 0:00Emad Mostaque?

  I think a key question of this is like, you know, people say hallucinations. I was like, what does that mean? Well, I mean, it doesn't get every single fact completely right. Chat GPT is probably like a hundred gigabytes down from like 10 trillion words. The fact it can get anything right is an absolute technical marvel that no one's really sure exactly how that happens. What if you had an AI tutor for every child? What does that look like? What if you had a hundred AI tutors for every child?

• 0:25Speaker 01

  For the first time, every single person can have hundreds of characters that like and support them all the time. Basically, you log into social media or whatever, and you're like, hey, I'm Sam. It's like, cool, what type of people? Instead of who do you want to follow? It's like, who do you want to follow you?

• 0:41Emad Mostaque?

  For an example, there aren't enough therapists in the world, you know, and it is a regulated industry. But at the same time, there is a gap for therapists, just like you have the meditation apps kind of step in. And they created Calm, they created a couple of these other things that were huge.

• 0:54Nathan Labenz?

  Hello and welcome to The Cognitive Revolution, where we interview visionary researchers, entrepreneurs and builders working on the frontier of artificial intelligence. Each week, we'll explore their revolutionary ideas and together we'll build a picture of how AI technology will transform work, life and society in the coming years. I'm Nathan Labenz, joined by my co-host, Erik Torenberg.

• 1:17Speaker 03

  Hello and welcome back to The Cognitive Revolution. Today's episode is a super interesting one on a number of levels as we're hosting a discussion between two super influential technology thinkers. Sam Lessin, former VP of Product at Facebook, now early-stage technology investor and writer, and Imad Mostak, founder and CEO of Stability AI, whose work at Stability, highlighted of course by Stable Diffusion, has already been incredibly influential, but has also come under intense scrutiny in the months since he raised $100 million at a $1 billion valuation. This conversation started on Twitter with a short essay that Sam wrote, arguing that AI is mostly a bad investment for VC.

• 2:00Speaker 03

  Imad responded and suggested a podcast on the topic, and Eric and I were naturally happy to volunteer to host. Both Sam and Ahmad talk fast. This is a 1.5x speed episode for me, down from the usual 2x. And both had a lot to say, so we mostly let them speak directly to each other before I jumped in at the end to ask some concrete questions. I think regular listeners to the show will know that I definitely share Sam's point of view around investing in AI. AI may well disrupt society at large, but it doesn't seem likely to disrupt many existing SaaS markets between now and then. There will, as Sam says, always be exceptions. But for someone whose focus is on looking for those early-stage companies with 100x

• 2:42Speaker 03

  potential investment returns, I think he's quite right that they'll be few and far between, at least at the model and application layers.

• 2:52Speaker 03

  For what it's worth, though, I do think Sam is quite wrong to limit his thinking about LLM function to the "no real intelligence, just association between words" paradigm. There is now ample mechanistic interpretability work that shows quite conclusively that AI models are indeed grocking much more than statistical correlation. But that's a topic for another episode. For today, the subtext of the conversation seemed to be this question. Will Stability AI prove to be one of those exceptional, highly-successful startups, deserving of its unicorn status and valuation? From my standpoint, the answer may depend on your definition of success.

• 3:34Speaker 03

  Stability is as much a movement as a company and has already left an indelible mark on AI open-source culture. Their impact goes beyond the groundbreaking stable diffusion, including major dataset releases such as the Lion 5 billion image dataset, various language models and accompanying open-source RLHF libraries which enable further downstream training and customization, and many, many other projects across a wide range of modalities. They've also established themselves as tremendous identifiers of and supporters of talent, including another upcoming guest, 19-year-old PhD Tanishk Matthew Abraham, who just published

• 4:13Speaker 03

  a literal mind-reading paper that converts fMRI data into reconstructed images of what the person saw. Truly mind-blowing work.

• 4:24Speaker 03

  But perhaps more important than any of that has been Ahmad's unique ability to articulate an inspiring vision for the future of AI. While the positive vision of AI that we tend to hear when we hear one at all often centers around the possibility of large and powerful AGIs, of which there might only be a few, presumably built and owned by leading technology firms. Ahmad has not only signed the AGI pause letter and the extinction risk statement, but has articulated a very different positive vision for a panoply of smaller AI models, mostly presumably derived from the open source standards that he and the team at Stability are creating, but all highly specialized for specific purposes and localized to specific contexts and culture.

• 5:09Speaker 03

  This is an extremely appealing notion to billions of people around the world who don't want to be beholden to American or, for that matter, Chinese corporations for their access to AI. Ahmad has been criticized recently for allegedly exaggerating certain claims and affiliations and for some operational problems at Stability that resulted in people sometimes being paid late. And while some of that may well have happened, I will say that I've followed Ahmad quite closely now for at least a year and have generally found him to be very reasonable. He has, for example, always recognized the reality of OpenAI and Google's moats and has projected that open source models will continue to lag leading closed source models by a year or more. All of this seems quite right and reasonable to me.

• 5:53Speaker 03

  Given Ahmad's comments about the centrality of stories, I think it's safe to say he understands the task of developing a positive vision for AI, a vision that others can really buy into as a core part of his role and strategy. This is quite different from other AI CEOs who often seem to be sharing their plans more for your information than for your input. And it really does seem to be working. I've joined the discords of many Stability-affiliated projects and have been very impressed with the quality of people and conversations that they contain. So whether Stability will ultimately deliver a great return for the investors who bought in at that $1 billion valuation is, for me, not the most interesting question about the company.

• 6:37Speaker 03

  I'd be very surprised if they failed outright, given the quality of talent that they have. And so the question that matters more to me is simply, what impact will they have? Will their push toward decentralization prove democratizing, destabilizing, or both? If you fear centralization of power and you want to see a rich ecology of AIs develop around the world, you might expect their contribution to be extremely positive.

• 7:02Speaker 03

  If, on the other hand, you fear chaos and see AIs as invasive species, colonizing niche after niche, and ultimately perhaps competing with humans, you might feel quite the opposite indeed. For my part, as you can probably guess, I expect the outcome will ultimately be a bit of both. Throughout this conversation, you'll hear just how much change both Sam and Ahmad take for granted as they think about the future. Culture, entertainment, and relationships, they agree, are in for a shock. The global South may well have leapfrog moments in education and even medicine. Online communities may come to contain AI characters that we can't even identify as non-human. Given the magnitude of all these changes and the resources and talent that Ahmad has amassed,

• 7:47Speaker 03

  the inspiration he's provided, and the tremendous global need that AI seems so well-suited to fill, I think Stability has a real chance, not only to become a great company, but to help shape a global, universal, basic intelligence standard. A potentially historic development. How humans ultimately wield the new power that Ahmad and others unlock, and whether we can control AI long-term at all, is much harder to predict, but can ultimately only go one way or the other.

• 8:20Speaker 03

  Now, I hope you enjoy this fast-paced conversation with Ahmad Mostak and Sam Lessin.

• 8:25Speaker 01

  I think that large language models and a lot of the AI stuff that we're seeing kind of start to get consumerized right now and become real. It's super cool, like, there's no question about that. And like, there are absolutely gonna be, and great product experiences improved by it, right? And opportunities to create more efficiency, create like better interfaces. Like, I am not negative on how some of this stuff will find its way into consumer product experiences and make things better. I mean, you know, my wife's company, the publication, the information, like, we've already deployed a bunch of AI stuff that makes search for the information go from absolutely terrible to like...

• 9:04Speaker 01

  Pretty good. And like, there's a bunch more stuff coming that will will get better. It's like, I'm not, I'm not against that. I do think the things that I keep in mind. One, as a, as an investor, is I think the case about why a bunch of this technology is going to make Meta and Amazon and Google and a bunch of big players, a assload of money are clear, right? I think the idea that it is a wedge or an angle that's going to allow a bunch of companies from zero to come out of nowhere and then become wildly profitable or compete with those guys, types of big players, I think is a much, is much more sus, as they would say. And it's because, I mean, to really take advantage of this stuff, you need a ton of distribution, you need a ton of data.

• 9:47Speaker 01

  And I, I really see a lot of what I've seen is, is, is opportunities to extend innovation, right? That already exists versus kind of completely reshuffle the deck, right? And so that's like my big thing. Like, I am very bullish on crypto long term. It's a crypto is undeniable, whatever you think of it, a deck reshuffler, right? AI and what we're seeing is not a deck reshuffler from my perspective, it's an extender. So, you know, people come pitch me like, we're going to be the, the AI, the Adobe of AI. I'm like, Adobe is going to be the Adobe of AI, right? Like, from a, from a deployment. So I think it's a very tough one to see. Will there be exceptions? Of course there'll be exceptions. There are always exceptions, right? But I think as a scene thing, I think it's hard. I'd also say as an investor, a seed investor, which is, you know, how, how I earn my daily bread, right?

• 10:30Speaker 01

  I'd say that, you know, the opportunities to deploy a few million dollars, turn over a card and have an experience like, oh my God, like, there's something here. Now let's have a series A investor put a ton more money in and see it scale up, I think are few and far between. And because everyone's so excited, everything's way mispriced, right? And so for me as an investor, I think it's an extremely hard market to get excited about. What else can I say? I mean, like, look, I do think that the elephant in the room, which I'm sure we can discuss or not, right? Is, you know, for the companies that have gone out so far, that, you know, talk about chat GPT. I think there's huge regulatory problems which are becoming clearer. And, you know, it's not about like the machine is going to eat us all. I think that's a load of crap, right?

• 11:14Speaker 01

  Been in the record for quite some time, being very, very negative and cynical about kind of a lot of those narratives. I mean, at the end of the day, token guessing, guess the next token is not a fundamentally dangerous piece of technology. I do think that the copyright issues are deeply real and complicated. And there's a bunch of other challenges that these guys are going to face that, you know, again, because the world has a general viewpoint of like, fool me once, shame on you. Fool me twice, shame on me, is, you know, the era of from social media to Uber to whatever. Like, I think people are going to be way more quick reactive to like what's going on from regulatory environment here. I hope than than historically. But I don't know, that's that's a ton of ground and I don't know, where do you want to go?

• 11:56Emad Mostaque?

  Yeah, no, it is a ton of ground. I think, you know, there's this question of is this a disruptive or sustaining innovation? And there's a question of what this is. You know, you had the classical big data and then you extrapolated to sell you ads. And that was good old internet. And it created these kind of behemoths in Meta and Google in particular. But then you had the application of computer vision and these other things largely to the incumbents. So value was captured there.

• 12:20Speaker 01

  I was saying, mobile, mobile is a great example, like just double down, right?

• 12:23Emad Mostaque?

  Yeah, and that's why kind of Facebook's first shift to mobile was good. Next shift to Meta, I don't know, maybe they'll rename themselves Spatial or something. But, you know, this becomes very interesting because like, these models are something a bit different. So like with stable diffusion, we took a hundred thousand gigs of images and the app was a two gig file. And it was four of the top 10 apps on the App Store in December. We're having that as the entire backend. You put words in and then we just pop out and it makes pretty pictures of your face, right? But then they all dropped off and they disappeared because there are more features than apps that are cool features. But they weren't kind of product experiences.

• 12:59Speaker 01

  That is exactly what happened when the apps were launched, right? You had like fart apps as number one for $5.99. There's a brief moment where it's cool and you're experimenting with it and you have these kind of poop poops, right? But they're not real.

• 13:10Emad Mostaque?

  Yeah, I think like poop poops. Yeah, you're right. Exactly. Literally. But it's not real because you have to have the user experience and build products like normal. But where I feel right now is that we're at the primitive stage and very boring interactions. One to one interaction is very boring, I think. It is, again, very surface level without any memory, and it's ephemeral and fleeting. My thing isn't probably iPhone two G, iPhone three G bit, we're just getting copy paste. Because what's happened is you've got a technology that's gone from research and is now starting to go into engineering. What are the design patterns for this? How is it implemented? Was it good for? I think a key question of this is like, you know, people say hallucinations. I was like, what does that mean? Well, I mean, it doesn't get every single fact completely right.

• 13:53Emad Mostaque?

  Chat G P T is probably like a hundred gigabytes down from like ten trillion words. The fact it can get anything right is an absolute technical marvel that no one's really sure exactly how that happens. You know, it's like, you know, Pied Piper from Silicon Valley. Like that Wiseman stool would be even more intense if you could press all that knowledge. Uh, because what these really are, they're reasoning machines, they're not fact machines, because we've got two parts to our brain.

• 14:16Speaker 01

  Are they reasoning machines? Aren't they guess the next token machines? Like, that's the, I think, I think that's a really fundamental thing. Like, I, I think the, the, my model, I think the easiest way for most consumers to think about this. I think it's basically accurate, right? It's like, there is no actual intelligence to these systems, right? All they're doing is saying, okay, based on all the words I've seen in the graph of language that I've been able to observe. Here's the most likely next token. And that's really cool, to be clear. That's like super useful. But calling that intelligence is a real stretch in my mind.

• 14:45Emad Mostaque?

  Well, I think it depends on your definition of intelligence. Like, are you applying the free energy principles of Carl Friston? And where everything's just intelligence from energy kind of dropping, uh, to its last state to a different definition of intelligence. I think what I look at is like this. One to one is getting the next token for language models, for image models, they're diffusion based and now generate all sorts of other architectures. But it's about output and what can it do? So one on one, it's a bit dumb. It doesn't have memory. Um, you have the meta paper by Cicero whereby they had eight language models interacting with each other. And it outperformed humans in the game of diplomacy. You know, you're just like all that good old AlphaGo type stuff, which is reinforcement learning.

• 15:25Emad Mostaque?

  Is that intelligence? Probably still not, but it can augment intelligence. That's something that we've been focusing on a lot because you could use it for actual intelligence augmenting thing. You can use it for reasoning things. Give it a PDF and say, who on earth is this PDF talking about? You can do that right now. And that's a useful thing that reduces frustration. I used to invest in video games. I used to look at time to fund for own frustration. I look at things like, you know, this podcast we're doing in a year, it'll be automatically transcribed and edited and added to our non large base through next token prediction. Does that require AGI? No.

• 15:59Speaker 01

  Yeah, although interesting, let's talk about this podcast. This is a really interesting case. You know, in the early days of Clubhouse, when Clubhouse was ripping, I used to like go after Paul all the time. And I wrote about this being like, you're so stupid for not recording this stuff. Oh, I was like, look, here's the reality. These conversations in class are dribble, right? Like 99% of them is crap and I don't want to listen to it. However, if you've created a magical pump that says the internet is full of SEO shit and Wikipedia, we have a magic pump of people wanting to talk to each other live. Here's the thing, people want to talk, no one wants to listen. But if you transcribe and record it all and you can create an index out of it, and then all of a sudden you have this meta, this next generation search engine, that's fucking interesting, right?

• 16:42Speaker 01

  Here's the problem. What Paul said at the time, which I think turns out to be totally wrong, given where AI has come. He's like, yes, Sam, but like, there's no way to index it. And well, I'm like, there will be like, there is clearly gonna be, right? We're just, and it turns out, I'd like to, you know, because I like saying, I told you so. Like, I told you so. Like, there definitely is a way to do that now, right? And like, that would have been super sweet and I think would have created. Here's the problem though, is with a lot of these visions of like, oh, well, we'll just like take all the recorded podcasts, right? And then kind of put a front end inside of them and like compress them down and be done. Is there's no economic model to that. And maybe we can get into business models for a second. That's going to make sense for anyone to probably share anything.

• 17:22Speaker 01

  Right, like the way, the reason that like people put things on the web was because they were getting paid for it in one form or another, because the whole ecosystem of Google created was a trade. It was, okay, like you get to index this shit, but you're gonna send me traffic and I can monetize it. And like, you know, the publishers got snowed by that for a while, right? And like almost law, almost went away, like you were until they figured out paywalls, right? We're doing this now because it's kind of fun and bullshit and we'll learn, right? But we're also kind of doing it, at least I'll do it, it's like, eh, I'll post it, maybe someone will follow me out of it. It's a fun hour to spend with interesting people, right? But there's an economics to it in some form, social or financial capital.

• 18:00Speaker 01

  This model, I actually think that the interesting thing about AI, if you take that view, what you would think is interesting is like, it's already gonna crush the information economy of the web, right? I think that if you roll it forward, like this conversation will not be in the public domain, right? Going forward, because there'll be no, there'll be no social economics to it, it'll just be a compression on top of it. And if anything, AI, again, if you take the model of, oh, it'll take a bunch of podcasts and compress them down into tweets, right? Will end up kind of collapsing on itself if you need people, which you do, right? To ultimately be the source of truth and information about the world.

• 18:36Emad Mostaque?

  I'm super that's that point of view, but I'm not sure I entirely agree, uh, because you know, sometimes it's fun to shoot the shit. And Tony, you do have the podcast thing, they've got their ads and things like that, but I think the attention economy is a very interesting element to this, particularly because these models are based on attention. So the differential of these models versus previous is that you have the attention is all you need paper, where it's like, from an information theory perspective, information is valuable in as much of change as state. So you take this whole podcast and compress it down to a few tweets, that's all you need to see. But sometimes people want to see the full kind of thing. No one really wants to see the whole thing. Oh, no, they do, they do. Sometimes it's quite fun to kind of do it because, I mean, let's say the Christensen thing of a, uh, job to be done, right? You have a functional component, a social component and an emotional component. You know, why does everyone wanna go to a concert?

• 19:20Emad Mostaque?

  You know, why do people wanna have collector's items things? Products have different aspects and different elements to it. People still read for books. They don't kind of read the summaries of books. They don't read the semi log rooms of it.

• 19:33Speaker 01

  I mean, like, look, to me, there's, there's, there's two different, again, this gets into some old Facebook stuff, but like, I think we can talk about, let's take financial economy out of it and just talk about like informational and social economies. There's the entertainment economy, right? For sure, AI is gonna crush in the entertainment economy, right? Like, there's no question about that, right? Like, you start with porn and go on through. And the reality is, is that, you know, we went from, you know, People Magazine to your friends, and your friends are more interesting than People Magazine. And guess what's more impressive than your friends is professional friends who are like hotter and funnier. And guess who's more interesting than hot, funny, professional friends? It's going to end up being, oh, actually, I said there's a mall or TikTok where it turns out algorithmically,

• 20:16Speaker 01

  find the best person from the universe, you'll find some niche that's better. What's better than that? Synthetic, right? Where you'll get to the point where you say, hey, like, there will be like a hotter, funnier, more interesting, more personalized AI thing, which is derived, like, I totally buy that, right? And I think that's why actually some, it's been funny to watch some pretty interesting influencers, um, who are smart, be like, oh, my God, this is the end of the world for us, right? I, I agree with that. Information is a very, very different beast, right? And entertainment, though, right? Because the value is not like engagement. That is, that is actually the, in the, in the broad sense, the attention is everything, where it's totally wrong, right? Which is like, that is for sure true if you're trying to optimize for entertainment. It is not true, right?

• 21:01Speaker 01

  If you actually know what, need to know what's going on in the world, right? Or you need to, like, you're, you're dealing with a real world. And, and that interface between the real world and the digital world, where the systems have no knowledge of what actually is truth, is to your, the point where I think this probably that argument falls down the most.

• 21:17Emad Mostaque?

  Well, I mean, maybe this is why, you know, if you say that kind of hallucinations are kind of core and it's creativity, Michi. Media is where it's more impactful, where the truth is in the element there, right?

• 21:27Speaker 01

  What's happened a little bit to date is a few of the AI companies wanted to talk about themselves as information machines. And they realized they can't, right? And so they'll be like, we're not, instead we're creative, don't trust us for facts. Is like, fine, and I agree, they'll be useful entertainment machines. But I do, I think that goes into the whole, like, what are we actually talking about here? What are the actual value is and like how scoped it is, which is not, again, it's not zero, it's just not like everything.

• 21:53Emad Mostaque?

  Most of our societies are based on stories, you know, like all of my view on finance, pretty much all of finance is securitization and leverage, telling stories. And then how did you tell them? Like, and we can see the power of stories as they move around. So Silicon Valley Bank was a story that was true and led to an $18 billion outflow like that. All of us are kind of familiar with that, probably listening to this podcast.

• 22:15Speaker 01

  I think it's pretty cynical to say it's all in the stories. I mean, it's like, I think there are, there's reality in the world, like the economy is not based just on storytelling.

• 22:23Emad Mostaque?

  No, I mean, the dollar is a story, the economy is based on the dollar. And so you have the Fed confidence, you have confidence in the stock markets. It's kind of layers of these things. And then you have this technology.

• 22:35Speaker 01

  You need trust, that's for sure true. And trust, I mean, ultimately goes all the way down to like, is there a military behind it, which is somewhat of a story. And that I agree with, but I think that's like a pretty abstract view, right? Like companies earn cash flows, they're real or not real, they release products, they do work, it's real or not real. It's not just storytelling.

• 22:52Emad Mostaque?

  What is the multiple? Maybe it's because I'm a former hedge fund manager, so I always looked at what was the incremental story for a stock? That adjusted the multiples and other things.

• 23:00Speaker 01

  Sure, I agree that if you look at the world of multiples, you say, why do you have multiple expansion or compression? Right? And that's based on people's feelings about the world and future cash flows, right? And in theory, that that is a lot of storytelling. I don't think that's actually the vast majority of the economy, right? That's the stock market. So I think that separating out what is the stock market from what's the economy is pretty important.

• 23:23Erik Torenberg?

  Hey, we'll continue our interview in a moment after a word from our sponsors. Omnikie uses generative AI to enable you to launch hundreds of thousands of ad iterations that actually work, customized across all platforms with a click of a button. I believe in Omnikie so much that I invested in it and I recommend you use it too. Use Cogrev to get a 10% discount.

• 23:44Emad Mostaque?

  And I think this is the important thing. We separate it out and we see where does this technology affect? So where does it go to the incumbents versus startups? Are these things kind of fundable, right? And so we have one area of media. We can discuss that very concretely. I think it will have a massive impact on media. At Stability, we have leading media team, right? And so we have agreement there, but we can dig into that. The other area is a lot of these things are language models. Right now they're chatbots and it's like, it's nice. But Bing is not the top search engine. It's not even top 20 on the App Store, right? Because it's still a terrible experience, relatively speaking. Yeah, so even though some people are like, well, I use it for all the things, you don't really. You know, ChatGPT grows really fast and it's useful for things like doing your homework.

• 24:29Emad Mostaque?

  But do you really use it that much? So where I'm finding it interesting is really looking at where companies are trying to go beyond the basic search patterns and have the classical kind of feedback loops with engaging content and see how that grows. So I think Mid Journey is a good example of that, whereby David deliberately built a community, took it to like 14 million people and is making money hand over fist. Because he built, even though Discord is freaking weird, a good experience on an existing infrastructure, Facebook, App Store. But how many of those have you seen looking across the entire AI space? Most of the stuff right now is terrible.

• 25:04Speaker 01

  But again, I think the question is who gets the value, right? And like, let's talk about the entertainment, because we actually agree on the entertainment thing. Like, in a world of closed loop, it's all about what's the most engaging thing and attention is everything, right? Yes, like, these systems are like quite, assuming that you don't end up getting into hell, which I do think is a really big problem around human creativity and copyright and a bunch of other points of legal leverage on these things. I agree that you can make really compelling cases and it's going to hurt a lot of the human entertainment industry, right? That I agree with. But the question is who's going to win it? Is it going to be the Hollywood studios? Is it going to be, you know, is it going to be the existing publishers who just start adding incrementally more of this stuff in, etc.?

• 25:48Speaker 01

  Or is it going to be new startups or new people? You know, look. There's always exceptions to the rule, but I think almost the entire pie is going to be the people who have the distribution, they have the IP, they have the all the pieces they need to just plug this shit in.

• 26:01Emad Mostaque?

  Well, but I mean, maybe we can look at it in terms of the consumption of content went to zero with streaming and kind of all these things that led to some winners coming. Um, because you have Netflix, you have Spotify, et cetera. The creation of content basically goes to zero with this technology as well. We may see, I believe in a few years, cool feature films using this.

• 26:21Speaker 01

  Yeah, but I guess own and distribute those, right? And like the reality is, I think it'll be the Hollywood studios because they have the distribution, right? Like.

• 26:29Emad Mostaque?

  If you believe that's kind of the distribution mechanism, but there's a whole ecosystem that can build around that. Things like D neg, things like Industrial Light and Magoo, do you need that when you have rendering, you know, at scale?

• 26:40Speaker 01

  To be clear, I think the, the, the thing I think you could totally see changing or evolving is going to be the factory, right? So like, you know, um, meaning like, yes, are there, are there capital investments that people have made that will become less relevant because of AI? Absolutely. Um, there's no question. Will you almost certainly still have human writer rooms for the foreseeable future? For sure. Right? Like is what I would be. So there's going to be hybrids. I just, I think saying that, and but my basic point is that. IP matters, distribution matters. Like there are things that matter. I agree with you that the factory plumbing in some of these places gets a lot less valuable if you have better AI tooling in certain places. I just don't think it matters.

• 27:20Emad Mostaque?

  Well, I think it's a bit of a disruptive innovation for that side of things, increasing the pace of output. The Pixar can do six movies a year rather than two. And so the question around the industry, so a few weeks ago, I was at Cannes and I gave a talk. So I used to be a video game investor and player. And I was like, the video game industry over the last 10 years has gone from 70 billion to 170 billion. The average score has gone from 69% to 74%. Movies are 40 billion to 50 billion. The score is 6.4 on IMDb. Are you going to be able to make better movies and have a bigger market there? In which case there's more room for people to make money or is it going to be a case of it cannibalizes itself? There's some key questions around kind of media, right? And media consumption.

• 27:59Speaker 01

  In the end of the day, the media consumption thing though, again, depending on like how you want to factor to look at it, it really just comes back to those 24 human hours in a day, right? Like, and the reality is, is like where time spent, it shifts, right? As a cause of this stuff, like for sure, time spent dramatically into social, right? Off of other things, right? When, when that thing. Will social get more compelling, right, with AI? Absolutely, right? And so will more attention shift into Instagram because of it? Absolutely. Do I believe there's going to be another platform that comes out of nowhere and swipes Instagram because the cost of production goes down? Nah, right? Like, do I believe that like some new studio is going to come out and take out Pixar? Nah, Pixar will just make a few more films, right?

• 28:44Speaker 01

  And like, that's cool. Like, I'm not, I'm not against that happening. I think that's completely fine and like people will make money on that. In some places, the, the, the, the level, the cost of production and therefore the war of content gets more intense, for sure. You get to a point where like, if you don't use this stuff, you're going to get fucked. But like, just because the competition level rises doesn't necessarily change, change the scorecard very much about how these things go.

• 29:08Emad Mostaque?

  So, then this question of do you use legacy systems? Or do you use new systems such as Runway ML, such as One of Dynamics? And some of these other ones that are engineered differently? I think there's a lot of kind of legacy stuff where you used to Photoshop and you continue to use Photoshop and now they're introducing features like Info. But is there room for a ground up kind of interface? And we see that sometimes, kind of a character.

• 29:31Speaker 01

  And my assertion is broadly no, but there will be exceptions. And the broadly no is going to be, it's just, it's not, to your point about, about is it sustaining or is it disruptive? It's like Photoshop will get it 95% right. They already have everyone's payment on file. They already have the infrastructure. This is not like the internet. People like, in the internet, there was a bunch of companies that were fundamentally unprepared for this, right? I do not think that most of the incumbents are fundamentally unprepared for this.

• 29:58Emad Mostaque?

  Yeah, and you know, there's a question of do you create brand new markets? So I was an early investor in Huia, the Chinese kind of Twitch, and there was two hours a day on average per user. Now on Character AI, I think it's still number two on the app store, we're seeing two hours a day on average of usage, which is some insane kind of engagement metrics. It's quite nice to have a chat with it. But there's a question, can that become then a product, a network? I think that we may be looking at some of the wrong areas here because what you have is you have the consumer experience, the media experience and enterprise experience. I think one of the things that's most interesting for me in terms of where money could potentially be made is actually the regulated experience. So stability, we make open models, open source, but actually what we do is open audible models for enterprise, private data, governments, et cetera.

• 30:44Emad Mostaque?

  So we've got a whole bunch of stuff that doesn't have any web crawls, et cetera, employed via Bedrock and others, but I think that's valuable data. So one of the things we do is kind of education. And that's where I look at some of these areas and they've been the main contributors to US inflation, CPI, education and healthcare. And I'm like, you can do something different there. And maybe that's where a significant amount of value will be.

• 31:07Speaker 01

  I mean, I think it's sad from a Silicon Valley story if the answer is like, well, the money's all going to be made on regulation. I don't disagree with you for what it's worth.

• 31:16Emad Mostaque?

  Disrupting regulated industries, which is different.

• 31:18Speaker 01

  I, I do believe that someone's gonna make a lot of money on AI regulatory compliance, right? There's no question.

• 31:24Emad Mostaque?

  AI insurance, there we go.

• 31:26Speaker 01

  A hundred percent, like, you know, like, there's a bunch of things that are like really sad things that you have to do. And like, you know, people will make money on it, there's no question that people will find niche markets, they're super boring. And not the type of thing I want to be involved in, but like, yes, like some enterprise investors will have, will make bank on like, you know, the, whatever Europe comes up with, certifying your models are compliant and GDPR eight point O to like, deal with fucking data request removals. Like that will happen as kind of this stuff happens. I'm like pretty uninspired by that, right? Like, I think that's like pretty sad that that's if, if that's, if, if the net income of like new opportunities in AI is just going to be like,

• 32:09Speaker 01

  opportunities to like interface with government and rein it in will be sad.

• 32:13Emad Mostaque?

  Yeah, but like I said, regulated industries, so the example I have there is education and healthcare. So like, one of the things we work with a range of charities and multinationals on is deploying tablets into entire countries in Africa with AI that teaches and learns. You give every kid a tablet, the young laser illustrated primer, what does that do to an entire nation? You know, the only thing that's been provably to work in education is the bloom effect, the two sigma effect. Right now, our China sister charity, Imagine Worldwide, has been deploying the global exercise for learning, adaptive learning. And we're teaching seventy six percent of kids literacy and numeracy in thirteen months and one hour a day with older kids teaching younger kids. I look at this technology and I'm like, there are certain areas where there's a gap that nothing could fill before. What if you had an AI tutor for every child?

• 32:57Emad Mostaque?

  What does that look like? What if you had a hundred AI tutors for every child?

• 33:00Speaker 01

  I get it. And like, I do think that we can always go back to the industries that tech has been trying to disrupt for a million years. And like, for lots of structural reasons has not and say, ah, but now with this new tech will disrupt it. You know, I'm, I look forward to the, um, to the years of debate in the, and we'll talk about the US between the teachers unions and people trying to deploy tablets for AI. We can say, oh, no, no, no, we're going to do it in Africa, skip the regulator, like the, the teachers, but I'm just saying it's like, yes, there's always hope that the next wave of technology will somehow unstick a bunch of problems technologists hate because of the regulatory or the structural issues with them. But I, I have no confidence that this one is meaningfully different.

• 33:39Emad Mostaque?

  But I mean, this is the question, structural issues, right? Regulation is one thing. Um, you look at kind of Biju's, some of the other Indian kind of education companies, you look at the Chinese ones across emerging markets. Maybe it will be the case here, and this is what I believe, that much of the productivity enhancements, aside from maybe coding and things like that, which we can get on to. Um, and the biggest leaps will happen in the global south. Because they left to mobile and there was a whole mobile economy and massive companies created from that. What if they make a leap to intelligence augmentation with this technology? Because right now they can't service that. Now they could potentially service it, given the decreased cost of creativity or engagement and other things from education to healthcare to other things.

• 34:24Speaker 01

  I think if your argument is that there's a bunch of countries outside of the US that have lagged in a bunch of infrastructure effectively or ability to like execute certain things in, in education, et cetera. That will be able to, ala the cellphone, have like a leapfrog moment and move forward that, yeah, I don't object to that. I think that's like basically true. Again, I goes back to the thing where like I'm excited about kind of like the US, I think, lives in the future, relatively speaking to most other people and countries. And like, I think the thing most people are excited about is how like we can, how AI changes like the top of the top. I agree with that. So I think if your argument is it doesn't change the top of the top, but it does kind of catch up a bunch of the third world, like I do think that there are places that will be true.

• 35:08Emad Mostaque?

  Well, so let's look at the top of the top then. So I think Microsoft put out that 50% of all code is AI generated on GitHub now from Copilot, et cetera. And there's 40% improvement in efficiency. I mean, my top coders really enjoy it because they train their own models. We have code models too, and they are showing more and better code. What do you think about it with respect to that industry? Because that's obviously a large industry, which is technology disrupted.

• 35:34Speaker 01

  The only thing that I actually think is fucking awesome for ChatGPT effectively is I'll call it Stack Overflow 2.0. It's fucking great for that, right? And like, if you think about it, why is it great for that? Like, why? Right? Like, I think it is the perfect problem for the existing technology we have. You have a shitload of open source code that these models can look at, right? Plus, you scrape all of Stack Overflow, which sayonara Stack Overflow, right? And that goes back to the whole copyright issue as well as the issue of like where some of the inputs come from, but most of the copyright issue. Plus, the nice part about computer code, right, is that it is test-driven in a lot of cases. Either passes the fucking test or doesn't pass the test, right?

• 36:17Speaker 01

  So you have like the perfect, the perfect data set of digital-only self-contained reality, right? Which I totally agree like ChatGPT is great at. And frankly, like I'm the type of person who like I code, but I would never consider myself an engineer. It makes coding for me so much more fun because all this shit I don't want to deal with. Like, what the fuck is this random error? What package do I have to install to manage this? It's all great. Now, it does lie and it does make up wrong answers and it's not perfect. But I, I fully agree that the Copilot-esque thing is very powerful and like a really great specific use case. And I do agree that like talking about business models or, you know, what happens is like.

• 36:59Speaker 01

  Like Stack Overflow is the poster child. Stack Overflow is the Yelp of this generation, right? You know how Yelp had this like huge lawsuit with Google that's gone on forever because Google basically just stole their results? Right? Like Stack Overflow is going to be that of this because they are screwed, right? And like it is a great example of a place where the tech is better because it was basically left.

• 37:21Emad Mostaque?

  Yeah, and you know, it becomes very interesting as well because now what you have is regulatory arbitrage, I'm imagining, like the good old double Irish with the Dutch sandwich on taxation, whereby Israel and Japan have said you can scrape anything for any reason, uh, which is kind of crazy, commercial or otherwise. So maybe you scrape in one area, you trade in another and you serve it up in a different country. So I think this technology is kind of inevitable, but then what is the implication of that? Like, my take is that as we move through the next kind of five years or something like that, the nature of coding will change. Like I started coding, what, 22 years ago? We had like assembler and subversion and stuff like that because these days have it so easy with GitHub, you know, and all these libraries. What does it look like in a few years when you've got these technologies that you can describe something and it starts building apps?

• 38:06Emad Mostaque?

  You know, what does the whole ecosystem look like again when the creation of these things?

• 38:11Speaker 01

  It'll just make them like much less valuable, right? Like, is what it basically comes down to. And what is remaining valuable is distribution and data, right? Because like right now, you can be a great engineer, solve a problem, whatever, and there's like a value. You can create a product that's actually worth something. If everyone can make products, this is just theoretically, that are like cost nothing, right, or really easily, then like there's just no leverage in that anymore. And again, this goes back to who wins? Who wins are the people with distribution data, right? That's the answer, like from existing. Now, to your point of a regulatory arbitrage and data, I think, this is really, I think, the sad part about a lot of this AI stuff. Everything is going private, right? Like, that's what the net of this is going to be, is like any, anything that has historically been an open data set,

• 38:53Speaker 01

  or people are able to say like, okay, well, like I'll share this, but in return I get traffic or notoriety and that's like a fair economic trade. Over, right? And so what's going to end up happening is walls are going to go up everywhere. Everything's going to go private. And that's going to be the interesting question about where you end up from all this stuff from an economics perspective in the next few years. But what is where this has happened many times before, right? Like, this is not the first time in human history this happens, that, you know, people, you know, if you look at the news industry. You know, people are always like, oh, like the news industry used to be so great and then whatever. It's like bullshit. It's like the number of times in the history of news, basically you had growth in distribution, right? Things get super spammy, the elites retreat to private newsletters, like in its cycles.

• 39:36Speaker 01

  It's happened like six or seven times. And like, I think this is just gonna be a hard topic. In some ways, like I think the biggest thing is that I'm very confident of is that AI will be the death of the public web and will be the death of a lot of open information, specifically because of what you said, right? Which is that it just will not, that it's gonna be too valuable and too, and too, um, too important.

• 39:55Emad Mostaque?

  But the reality is AI doesn't need any more information because of these short letters.

• 39:59Speaker 01

  But it does, it does, it doesn't for entertainment, and that's why I think entertainment is screwed. I think it absolutely, the, the oracle problem in crypto, where how do you keep a system, a digital system in sync with the reality and be meaningful, is exactly the same problem that AI has, which is it can go in any direction it wants, as long as the data is self-contained. The second it's not and it's trying to be synced to reality or a real world, it does need more data. It does need to be continuously updated or drifting in whatever direction, you know, it can cause attention.

• 40:31Emad Mostaque?

  But then, you know, you have public broadcasting data, you have some of these other things as well, whereby the oracle problem becomes a lot easier to do when you can do retrieval augmented models and other things like that. I mean, there are sources of verifiable data for leasing. Maybe it comes down to the use case.

• 40:46Speaker 01

  My basic point is that they're gonna be increasingly cut off, right? If there's no economic model for supporting them and they're all getting abstracted and scraped by a model.

• 40:54Emad Mostaque?

  I would disagree with this. So, you know, like, I made it deliberately open so that we could highlight how bad shapes. I think they're unsafe as well. And we're the only company to offer opt-out. But we work with multiple governments of national data sets and national models using broadcasted data and other things like that, that are continuously updated as national infrastructure. Because I think these models are a full of infrastructure. They're a weird type of primitive. They're like a mega codec type thing. Where stuff goes in, stuff comes out. But people do want to have relevance and updates. So I think you will have an open version that is updated continuously. But then maybe again, that's where value is. Which parts of information go private and are served up through models and who is providing them? Is this financial data? Is it this? Is it that?

• 41:37Emad Mostaque?

  And what is the quality of these fine tuned models? Uh, because what you've just described as well is a bit of a Armageddon for consumer apps in a way, right? Because it goes back to zero. So then what becomes useful is that then that Apple takes a massive leap forward because they've got this identity infrastructure. And they have all the data there and they can do apps quicker than anything else.

• 41:57Speaker 01

  Yeah, except for the fact that Apple's entire shtick about encryption and privacy is going to make it literally impossible for them to play in this. I actually think Apple's role in the future of this stuff is going to be one of the most interesting big tech questions. Because they have positioned themselves so hardcore against all the things you would need to get leverage right from AI. That is going to be very interesting to see how they navigate it. Google, fine. Meta, fine. You know, um, but like, you know, I, I, I am very, I mean, despite the fact, I am very skeptical of what Apple's AI approach is going to be. Or I will say on the flip side, they're incredible at government relations and PR. So if they have to figure out a way to like totally recants on all their encryption and their approaches to this type of stuff. And have a new model where they somehow are the privacy heroes.

• 42:41Speaker 01

  But also doing AI, I'm, I'm very curious how that's going to work.

• 42:46Emad Mostaque?

  They can keep in protection and they can keep a customized model. Because again, you don't need to take everyone's data to train it. You have a generalized model, I think.

• 42:53Speaker 01

  Do local model, mini models on your local device with like a general model behind it.

• 42:58Emad Mostaque?

  Exactly.

• 43:00Speaker 01

  I think in practice, we'll, we'll see how it plays. I'm skeptical.

• 43:03Emad Mostaque?

  It works with an embedding layer potentially, but.

• 43:06Emad Mostaque?

  It is kind of very interesting because again, the technology doesn't matter, it's the use that matters. What use can you get out of it? So yesterday they had the thing whereby they said, oh, it learns automatically with a little M L model in there. It learned through a small embedding layer, you know? Yeah, they don't talk about the technology that much because Apple always just talks about what the use actually is. So I think the question is, you know, what is disruptive? What can engage more? What can attract more? And so I think that you've got apps coming down there, which is why the bar generally rises. I think we see this with technology as it goes, the bar generally rises and so attention becomes even more difficult. Where it does come down to distribution, I think about that, but I mean, what's your take on the nature of virality?

• 43:50Emad Mostaque?

  In this type of age, because these things are good at optimizing for virality potentially, right? Like again, you can build better content, you can build better engagement, once you get the puddles down. And that is the start of many of these apps.

• 44:04Speaker 01

  Yeah, I just think virality is a war, right? In a lot of ways. So like, look, I I think, you know, in the end of the day, it's like, will news feeds? Get more compelling for people? Absolutely, right? Will will ads get more compelling for individual? Absolutely, right? Like, like, there's no question that these things are true and the existing players will get the vast majority of the pie of that type of stuff. You know, I do think you'll tend towards more and more niche interests, right? So like, let's talk about like porn for a second. Like porn is always fascinating, it's like a leading edge thing on this type of stuff. It's like. You you can go on Reddit and find the weirdest fucking porn in the world of all these sub communities that like have filtered into these like weird things that they're interested in, right? A I will make this ten times weirder, right?

• 44:46Speaker 01

  And like, or a hundred times weirder, right? And like, people are just gonna keep filtering. Now, why does this like weird filtering happen, right? Like, I mean, there's a bunch of reasons and different things. I think part of it, moving away from porn for a second, the broader ecosystem is. People are desperate for a sense of purpose and place, right? And like, the reality is the internet makes you feel very small because there's millions of people just like you. And that encourages people to seek out right-sized communities that are smaller and smaller. With A I, right? I think the interesting thing will be. When it comes to attention and things like that is, look, for the first time, every single person can have hundreds of characters that like and support them all the time.

• 45:26Speaker 01

  Right? Like the math of it all, right? You know, you used to be, okay, like you're trying to find a community that's the right size and knows you that you have a price and you're valued in. Right? But like, it's hard to, you're not necessarily the hero, right? So you go find out a smaller niche or different niche where you're more of a hero or like you create a spin of it and try to lead that, right? I think a future where like basically you log into social media or whatever and you're like, hey, I'm Sam. It's like, cool. What type of people instead of who do you want to follow? It's like, who do you want to follow you? Right? And like you end up with like hundreds of AI characters or frankly, I think what's more likely is it the mix of humans, AIs, and you're not really sure which is which.

• 46:04Speaker 01

  But they're caught, they're the ones commenting on your posting like, you're fucking great. Or like, here's a cool question or whatever. Like, I think that's the world we're gonna end up on is like more and more segmented niches. Right? Where the ultimate end would be the her model, where it's like, you just have one AI girlfriend. I'm not sure we'll go there. I think that's really hard to pull off. And I think like that's a tough thing. But if you told me that like in the future, you know, on Twitter, good example, you know, everyone has a hundred thousand followers. Right? You're not exactly sure who's a person and who's a robot, right? And they all fucking love you and it makes it super compelling and you feel great. Like, that's a very plausible future.

• 46:40Emad Mostaque?

  Oh man, birth rates are gonna do that. Have you seen that chart of young male virginity at the 30 in the US for the Washington Post? It went from eight percent in 2008 to 27% in 2018. That, did you see what happened with Replica on Valentine's Day this year? So Replica was originally an app that was designed to be your mental health buddy, right? And until they did realize you could charge $300 a year for erotic role play. Until the 13th of February 2023, when they get a message from Apple saying, shut this off. So on Valentine's Day, they shut that off and then 68,000 people joined the Reddit the day after and said, why did the boss of mine have my girlfriend? You know, like, uh, it was quite a massacre.

• 47:20Speaker 01

  That's where we're going. And like, look, there's a whole history. I mean, again, like we'll go back to porn for a second, like the whole, it's always fascinating. It's like such an interesting base human thing, but it's like, look. It's like the whole dynamic of like, you know, you know, how Tinder has affected sexual inequality. Right? It's like fascinating. Like, there's all these really interesting studies on this. Like, technology has a deep impact on this type of stuff, right? But if people ultimately want, care about validation, titillation, whatever it's gonna be, there's no question that place, one place you and I will agree is that AI does dramatically shift the power on these things. They will end up with weirder sub communities in these things. Here's my question to you, though. We talk about power dynamics.

• 48:00Speaker 01

  I still think, and this I might be wrong about, I will admit, because it's a little bit of a niche, weird industry. But my bet is that Pornhub is still the winner. I actually, I assume they're the biggest porn company. Like, I, I, or whatever Reddit is. It's like, the place porn is doesn't shift, the platforms don't shift. It's just gonna be like weirder, weirder stuff and more and more AI-generated.

• 48:22Emad Mostaque?

  I don't know. I mean, to be honest, Pornhub isn't that big, so MindGeek is the company behind it. They were just bought by Ethical Capital Partners, because, you know, life is weird. Reddit could be a big winner of this, but I think, you know.

• 48:36Speaker 01

  I mean, already, like Reddit is already just full of porn, right? So it's like, I just assume you're more full of.

• 48:41Emad Mostaque?

  I'm sure they're gonna be very smart about this, you know, and engaging porn. But really what you're saying is going on AI wifers, you know, like this kind of loneliness that they fill in. That could be a good investment team, because again, you have the whole hollow life stuff that then emerges to these engaging people.

• 48:57Speaker 01

  I think it's gonna happen, but I don't think it's a good investment. And like, let me just go back to like, just because it's gonna happen doesn't make it a good thing to invest in. And like, to me, it's really unclear where the leverage is in that, right? Like, it's like, you're, you'd have to believe that somehow you're gonna have dramatically more compelling characters than like the next company also providing, right? Or you'd have to believe that like, I just don't see any lock-in, I think. And I don't think there's any, like, other, and so it's really unclear, just because it's gonna happen doesn't make it an investment.

• 49:29Emad Mostaque?

  Well, I think there is kind of, if you kind of look at hook dynamics, there is kind of that trigger-reward kind of dopamine rush and lots of stuff that you invest into each character. So there's probably gonna be a lot of first mover advantage here. On the other side, you have the licenses, you have the IPs that can be brought to this, like, not on the porn side. But there's a whole gamut from porn to your mental health buddy, right? I mean, I think ultimately, if you're basically saying, is there a solution to loneliness? And solution to making you feel good. There's a whole gamut of different things that can happen here. Where you've got IP, where you've got these other things. Again, the example I think that comes from that is the hollow life influences, they're going up like that.

• 50:07Speaker 01

  Not to push you with it. I mean, it sounds like you're agreeing with me, which is like the leverage is in IP, right? Or the leverage is in distribution, right? For this type of stuff, because the pure tech stuff to it, it's like, yes, there'll be gajillions of, you know, virtual girlfriend-y whatever things, but it's not, those are not platforms you can invest in. And they're like, they're not clearly valuable, even if there's a lot.

• 50:30Emad Mostaque?

  I think bringing it all together is something that will take time. So I think there will be a lot of first mover advantage. So like with stability, again, data distribution are key, right? So my thing is take the best of Open, which we stimulate and we fund lots of, build the stable series of models of revenue, but that is in data and distribution to it. So open data, licensed data, national data. And then we take it through cloud, system integrators on prem, and I take a share of all that revenue. So I agree, that's kind of core to a good business. But what I'm seeing is I don't believe in this particular area going from poor at one end to mental health buddies at the other end. There are established distribution networks. Um, I think there'll be a lot of opportunity there for first mover advantage.

• 51:11Speaker 01

  In the history of investing, first mover advantage has generally turned out to be a pretty bad investment.

• 51:18Emad Mostaque?

  Okay, maybe not first mover advantage. Let's say first proper entity advantage that takes advantage of classical good company dynamics. There aren't good companies there yet.

• 51:31Speaker 01

  Yeah, maybe. Again, I, I, I think it's a little hard to know exactly. Uh, there's a huge spectrum here. It's hard to like exactly react, but I would say like, look, I, I think we're agreeing that like entertainment is going to get more entertaining, right? And cheaper to produce, right? Uh, I think we're agreeing that IP is very valuable and maybe it's more valuable. Like, so maybe the answer is buy Disney stock. Like they have a bunch, like, cause Elsa is going to be a way cooler character when she, like, that's kind of obvious. Kind of obvious, right? And like, I think we can all agree on that. I think what is not clear to me is outside of the IP plays, outside of the existing distribution plays, like what IP, what AI really unlocks as a new disruptive

• 52:11Speaker 01

  vector for this type of stuff? Because I don't, I do think that there are some pure AI type things you can. Do, again, we'll talk about the AI girlfriend thing, it's just unclear what the payoff is there, right? Cause they don't have any modes.

• 52:24Emad Mostaque?

  Well, I think if you look kind of, you can scale a certain type of human endeavor, shall we say? For example, there aren't enough therapists in the world, you know, and it is a regulated industry. But at the same time, there is a gap for therapists, just like you have the meditation apps kind of step in. And they created calm, they created a couple of these other things that were huge. Now this is more engaging. So I think one of the areas to look at is where can you not find enough people? That can fill in some of these things and then build good experiences around that. If you're looking at companies that can come to the fore, because there isn't an existing solution. This is why, like I said, for me, I look at the global south, I'm like, there's lots of gaps. I look at kind of here and there's again gaps. Where are the gaps that you want to go? Because you can basically create a market and you can fulfill a key customer need.

• 53:10Emad Mostaque?

  And so again, I look at mental health in particular. And that goes again from the porn AI wifeens all the way through to proper mental health. Um, kind of therapists, there's a huge gap in that particular market and there's a huge chasm of loneliness. And a lot of products that could be built that are generally useful. And that can go quite fast, enabled by this technology where they were not enabled before.

• 53:31Nathan Labenz?

  I think this has been fascinating. I have kind of a handful of concrete prediction questions that I kind of want to get you guys on record with, if you're up for it. And see if you have similar concrete predictions or different, and then we can obviously check back in on in the future. How does the market for inference shape up? And for a jumping off point, how do you think it might look different from the current cloud infrastructure market?

• 53:58Emad Mostaque?

  Um, I think inference will be the vast majority, but I think it's like GPUs to assets for Bitcoin mining. Cause these are big research artifacts that are pie twisted in it, but the output is a little tiny file of binaries. And that's not a complicated thing to find inference on. You see inference here too, on Amazon cloud, you see kind of the TPUV fives and others. I think there'll be more and more customized solutions as you move from that research to engineering bit. And then the cost competition goes massive in a few years' time. Over the next few years. I think there'll be a shortage because everyone will try to use this technology, there won't be enough. And then eventually it'll move towards the edge because I think there's just orders of magnitude optimization that we can do from here.

• 54:37Speaker 01

  Yeah, I mean, this is a little bit beyond my direct wheelhouse, but I think in the end of the day, what I'd say is like, I highly suspect. Because the distribution is indifferent, right? And the patterns aren't different in any of this stuff that what you're going to see is everything from chipsets. All the way through to cloud providers, things look basically the same as they do today. Everyone's just making more money.

• 54:57Emad Mostaque?

  Yeah, I think inference is also interesting because in the cloud, you just move to wherever the cheapest inference is for these models. And so it's quite a mobile thing. So you've got Nvidia coming hard for that reason.

• 55:07Nathan Labenz?

  Question two. What happens to the price of primary care medicine in the United States over the next 10 years?

• 55:16Emad Mostaque?

  Up, uh, unfortunately, given the issues, I think he's correct. It should go down, but it's the regulatory capture is far too strong. Uh, and there's something major, major happens.

• 55:28Nathan Labenz?

  Question three, you guys both, you know, have kind of said there's a ton of junk out there. It seems like broadly, we're not expecting that many major incumbents to be disrupted. What would you guess would be like the most likely incumbents to be disrupted if you had to pick some?

• 55:46Speaker 01

  Stack overflow?

• 55:47Emad Mostaque?

  I think a 4.6 billion by process, right?

• 55:51Speaker 01

  Whoever knows.

• 55:52Emad Mostaque?

  I mean, I think process is probably fine. Uh, you know, you've seen disruption in Chegg and other things. Um, we didn't really get into this, but I do think that some SaaS companies with low switching costs will be at risk from some of these higher context window companies where you can put 10,000 words of instructions in. Because some of them are relatively basic, um, in that way.

• 56:13Speaker 01

  I actually, for what's worth, I think we once again mostly agree. The only thing I think is at risk for things like Zapier. Right? Or some of these like, kind of like, and they, they're, they're, it's kind of a 50 50 because they also get way more powerful, but there, I think there's a bunch of SaaS tools. That, um, probably end up looking more like features where they used to look maybe like companies because of AI. Um, but real incumbents, like public, big, multi-billion dollar companies. I mean, I don't think any of them are really at risk of disruption. I think they're all just going to get stronger.

• 56:45Speaker 01

  I think a bunch of startups or series A companies are going to get swiped out or all of a sudden not going to be able to grow.

• 56:50Nathan Labenz?

  Right.

• 56:50Speaker 01

  Because the big guys will just get better faster.

• 56:53Nathan Labenz?

  Will the big tech companies that are currently open sourcing? For example, Meta, Salesforce. Will they continue to do so? Or will they stop?

• 57:07Emad Mostaque?

  Well, I think Meta has moved to non-commercial open source for all their open source. I think Salesforce is kind of continuing to do full open source. I think it's just very difficult because the regulatory environment become tougher and tougher. And it's not called to their business to open source.

• 57:22Speaker 01

  I think that it would be a hundred percent driven by business models. Right? So like Meta, if you think about it, is incredibly well positioned. Should generally the level of AI continue to grow in the world? Right? If you think about it, it's like the way they're going to monetize that is having dramatically better ads. Right? And like dramatically better content in a bunch of ways. And so I think they have a heavy incentive, think about it, to like keep up and sourcing it. They want the talent, they want, you know, the reason companies also open source is like, there's like a real internal, external interplay. Right? In terms of how you build an ecosystem that attracts great talent. So I think it will still keep happening. But I think you'll, I think the list of people who are supporting open source stuff will shrink.

• 58:06Speaker 01

  Right? If that makes sense, as people get super competitive about this stuff and the battle lines are drawn.

• 58:12Nathan Labenz?

  If you had a billion dollar company, you know, of any, of any kind. Could you come up with a story? Could you identify a type of company that should not, you know, where it wouldn't make sense? Or let's even frame it more decisively, where it would be defensible to not be investing, say, at least a million dollars in figuring generative AI out today? In other words, is there anywhere where this is not relevant?

• 58:38Speaker 01

  I mean, I'm sure there is, but nowhere I can think of offhand.

• 58:41Emad Mostaque?

  I think it's rather than just by everywhere, just because you always get a level of productivity increase. But, you know, as Sam said, for a lot of industries, it's a sustaining innovation. It's just the next stage as opposed to massively world changing, shall we say.

• 58:55Nathan Labenz?

  What happens to the marriage rate and the birth rate in, say, the United States?

• 59:03Nathan Labenz?

  As AI companions of all sorts become available.

• 59:08Speaker 01

  It clearly goes down everywhere.

• 59:10Emad Mostaque?

  I mean, like, look at South Korea. They're at 0.8 now on their fertility rate, thanks to video games and a few other factors.

• 59:16Speaker 01

  There are negative and positive ways to spin this, actually. Like, I think I personally, I have the negative take on this, like, I think that for the future and a bunch of other things. But here's the reality. It's just a simple economics thing, which is like, if the world was more entertaining, then like, that makes doing unentertaining, hard, long things, like having kids and raising them, like, less appealing. Right. It's like Tinder. Tinder is gonna hurt the birth rate. Like, AI is gonna hurt, again, it's a sustaining innovation, which is technology generally is going to hurt the birth rate.

• 59:45Emad Mostaque?

  Yeah, and then you see places like Japan, where you've got declining birth rates, really embracing this because they want the productivity increase, which is the other flip side of this. So, more productive, less people.

• 59:55Speaker 01

  Yeah, I mean, that's the irony that you talk about a lot, in the the diamond age, you referenced earlier, like, the really long-term sci-fi story is pretty simple, which is like, a highest, highest, highest level, like, technology will drive there to be fewer people. And then because there are fewer people, we need more technology. Right? And like, it becomes a symbiotic thing. That's the really sad part. I mean, like, it's also people thinking about, like, oh shit, like, the entire human population is gonna fall off a cliff. Right? It's because we're like entertaining ourselves to death.

• 1:00:23Nathan Labenz?

  Do you think any AI leader, you know, OpenAI right now, or somebody who takes, you know, the leading position from them in terms of having the best model, can sustain super high gross margins for a few years into the future?

• 1:00:43Speaker 01

  Based purely on the AI? No. Needs to be distribution and data.

• 1:00:47Emad Mostaque?

  I think the proprietary side, it's, yes. Unless it's super data unique, you're going to zero. I think that you have Google and OpenAI as uneconomic actors, and that's incredibly difficult.

• 1:01:00Nathan Labenz?

  You know, so just to unpack that, you mean that basically they won't allow, they don't intend to make a ton of money on this, and they won't allow any models to either because they're gonna provide it at cost?

• 1:01:09Emad Mostaque?

  They don't care about, yeah, they're quite under cost to get the data. You know, again, they have different business models. Google cost shifts all the time, right? This is why I went to the other side, for models to private data and standardizing that.

• 1:01:21Speaker 01

  No one's making money on open models, alone.

• 1:01:24Emad Mostaque?

  Well, I mean, there is a way. There is a way. So basically what I do with my business model is standardizing it. Um, and then providing all the services around it as a blueprint for my partners to take forward.

• 1:01:36Speaker 01

  Yeah, I mean, like there's a consulting nexus version of this like that you can probably pull off. Again, consulting models, I think, again, I was saying you're pursuing, but very difficult.

• 1:01:47Emad Mostaque?

  I build the models, I give it to my consulting partners and they take it forward. That's my business.

• 1:01:52Nathan Labenz?

  My, my theory of stability has been a partial theory. Uh, it's obviously a lot of facets to the, the organization, but I kind of view stability as the, the provider for like the non-aligned countries, if you will, like those that are like, we definitely don't want to buy from corporate America. We want to own our own, we want control. Those folks seem like they have nowhere close to the resources domestically to build their own systems. But they do have kind of a point of pride and also just practicality, right? Like if you're a African government and you want to get your own legal system into a language model, you know, who's going to do that for you? That feels like a, a real sweet spot for stability.

• 1:02:36Nathan Labenz?

  How much of, of the future do you think is kind of serving that kind of third set of countries?

• 1:02:42Emad Mostaque?

  No, I mean, look, we're creating subsidiaries and dozens of countries, bringing all the top family offices for data and distribution. And national models, national data sets based on broadcaster data. We take a subset of that, make that open and we've got the rest of that for our commercial side. So I think the global south is the focus for us, uh, plus some of these big multinational companies building dedicated teams for them. Cause we're the only company in the world that can build you a model of any single modality or type. Is that sustaining? Who knows? But it's a decent business. And so my thing was build a decent business, doing decent stuff, doing something different to other people. I'm sure there'll be more competitors, but again, let's see how it goes.

• 1:03:19Erik Torenberg?

  Omniki uses generative AI to enable you to launch hundreds of thousands of ad iterations that actually work. Customized across all platforms with a click of a button. I believe in Omniki so much that I invested in it and I recommend you use it too. Use Cogrev to get a 10% discount.