The Next AI Wave Will Be Social, Not Solo | Sarah Tavel, Benchmark and ex-Pinterest

[00:00] Google was a founding team that was deeply, deeply technical. As the technology, the underlying technology got more mature, the slider goes forward, forward, forward, more towards the product thinker, product experience. Pinterest, where I was, Snap, Instagram. [00:17] The CEOs weren't technical at all. They were product geniuses. What are the big consumer wins so far in AI? Of course, it's ChatGPT, which is in a way, not that dissimilar from Google in terms of what it was, just a text box. Custom GPTs and ChatGPT feels criminal to me. It's clearly made by a team that is unbelievably capable, but isn't social. What's the multiplayer network effect type experience? [00:47] a UGC type community where there are people who are really, really good, make it so much easier for the rest of us. [01:10] Sarah, welcome to the show. Thanks so much for having me. So for people who don't know you, you are a partner at Benchmark. [01:16] Yes. Before that, you were early at Pinterest. [01:19] And before that, you studied philosophy, which I also studied philosophy. So that's close to my heart. Yes, I've heard your podcast. And I was very impressed with your podcast, but Reid. Oh, thank you. To keep up with him was a lot of very impressive. It was a couple of late nights of me furiously prompting Chad Gbt to explain Wittgenstein. I love it. I love it. Well, you did great. Thank you. So I'm psyched to have you on the show. There's so much to talk about.

[01:49] technology and consumer technology cycles and how you can use the lessons of previous consumer technology waves to kind of help you understand this AI wave and this cycle and what kinds of products are going to work and what kind of products are not going to work. I'm curious. I think that was a good place to start. You know, one thing I just was reflecting on and, you know... [02:08] you kind of look at, [02:10] ChatGPT and CharacterAI, and I was just puzzling over those. And then started to think back to [02:17] you know, what was like the big, [02:19] early kind of consumer web hit. And that was Google. I mean, it was Yahoo and Google, but what was Google [02:25] Google was a founding team that was deeply, deeply technical. [02:32] And really, if you think about a product experience that you expose to the user, and how much of it is the UI that you interface with the product itself versus all the magic that happens on the backend to make something... [02:45] that's really complex, simple on the front end. That was like what Google was, you know, so good at the distributed engineering, the infrastructure. And... [02:55] And then as the technology, the underlying technology got more mature, [03:00] you started to go to a place where maybe if you kind of said like, [03:05] like deep technical, you know, you have 0% and 100%. Like I would say Google was 95%, like deeply, deeply technical. And then you start to move that bar over and you get to, you know, I think about Facebook, like Facebook, it wasn't the same technical depth, of course, of Google, but relative to Friendster and MySpace, they were more technical. They had, they were a little bit later and it let them create like a really performative experience that ended up really kind of,

[03:35] winning the day. [03:36] And then you progress even further. Pinterest, where I was, Snap... [03:41] Instagram, the CEOs weren't technical at all. They were product geniuses, right? And so like the, the slider goes forward, forward, forward more towards the product thinker product experience. And then you think about what we like, what are the big consumer wins so far in AI? Of course, it's ChatGPT, which is in a way, not that dissimilar from Google in terms of what it was, just a text box. Yeah. And character AI, like, [04:09] unbelievable what they did. And it was like a new paradigm, but still it was always like you'd speak to know him and for him, the product was a model, you know, it wasn't, it was a hat, it was a little bit, maybe 94% back end, but it was still like very much so. So it's still early, like everything is moving under our feet still. And I think to really have the people who have more of that [04:39] experiences on top, you need more of the underlying infrastructure to be a little bit more stable. But it does seem like we're moving into that next paradigm soon. And what's going to happen there? That's really interesting. I love that articulation, in particular, because one of the things that I felt [04:57] is very unique about OpenAI. Is there a research lab that accidentally built the biggest consumer technology product of all time? But it seems like you're saying there's actually a real historical precedent for that and that the DNA of Google is very similar to the DNA of OpenAI, which I'd never really made that connection consciously before. And I think it's also really interesting because investing in...

[05:22] basically PhDs doing long-term research that may have no practical purpose is not usually something that pays off in venture. In venture business, it's not like the first place that people think, you know. It's more like Stanford dropout, you know. Yeah, absolutely. And so maybe it's one of those things where usually that's not a good bet, but if you're really dealing with a truly new technology paradigm, [05:46] it could be the best bet you ever make. Yeah. Is that how you think about it? Yeah. And, and, you know, I, part of what I think about, it's just like, [05:53] You are a power user of these products. And I am on that learning curve. I would say I'm pale in comparison, but like relative to the population of the United States, I'm pretty damn good. [06:08] it shouldn't be this hard, you know? And like, so some of the underlying, uh, [06:13] you know, like the models will get better. So like one thing that I know you have in your custom instructions, I use a lot is just like, you don't have to answer like just to chat GPT, like you don't have to answer me right away. If you have some clarifying questions, like ask those. We shouldn't have to put that in a custom instruction. Like there should be, or there's so many different tweaks that we all have to get what we want out. And over time, [06:43] to get really what you want is going to get easier. But I don't think ChatGPT... [06:49] is the single player mode product.

[06:52] And, you know, the custom GPTs that they have where you can see what other people have created. To me, man, I just think someone's going to create a UGC type... [07:04] community where [07:06] There are people who are really, really good, make it so much easier for the rest of us to really take advantage of this, this technology. So we're, we're, you know, so there are places like Google is still Google. Like it falls, my analogy falls down when Google didn't evolve into some multiplayer product. There's no other product that took over Google until really now, but I, I still think. [07:34] We're so early in knowing who really is going to be the winner in this world. Yeah. I want to go back to that sort of transition from highly technical founder to product genius, if that's the continuum. I can understand why at the beginning of a paradigm shift, highly technical founder is necessary and will win over product genius because they can actually build the technology that makes the difference. But I'm curious for your thoughts on what drives the transition to product genius, because, um, [08:03] you know, I can understand the making, for example, simpler user interfaces, like maybe product geniuses are better at that. But yeah, what's the underlying force? Because I could also see a world where the highly technical founder is still like really, you know, effective in as the paradigm gets more and more figured out. Like, yeah, talk about that. I think a big part of it is that

[08:25] you still like so much of the tooling and infrastructure is still to be built to let somebody who isn't deeply, deeply technical themselves get what they want out of it. And so that's why like there are so many products now that I see that feel kind of pretty similar to each other. You know, there's all these, you know, the character AI genre, right? Where you make a character and you engage with it. They're all relatively the same because you really, I mean, [08:55] was uniquely qualified to build that type of product, like actually... [09:02] going into the brains of the model and changing it to, to create the experience of the user. Um, [09:09] But it's still, when you need to have that level of ability to get what you want out of it, also the costs have still been pretty high. I think DeepSeek could be, one of the hypotheses I have is that DeepSeek is a moment of change where it makes it more possible. But yeah, it's just, I think you need... [09:31] more maturity in the underlying infrastructure, your ability to, to do the things that you want with the model without being the deeply, deeply technical to be able to create the experiences that are possible. That makes sense. I think what I'm asking is, so let's say the infrastructure, that infrastructure is built and that, but you still have now technical founders and product genius founders, which we're making a strong division here for argument's sake. Sometimes they overlap. Um, uh,

[09:57] So in that world where the infrastructure is built and it's a technical founder versus a product genius founder, like what is driving the success of the product genius founder in a world where everything is a little bit more mature? I think it'll depend on so many things. Ultimately, it's a... [10:12] To reduce it to the basic, it's like who's going to create the most engaging product, the experience. I suspect that one of the things that is missing from a lot of these experiences that people are creating is just like what's the multiplayer network effect type experience. And that is... [10:32] genius to create that type of experience is very different than the type of experience, the type of brain that creates a single player mode experience. And we haven't really, I mean, there's, again, these kind of character AI, um, you know, [10:47] offshoots that have people that I can create a character and you can play with a character I could create. There's a little bit of status seeking work happening there, but I think we're still very, very early in like the true... [11:01] thinking happening. What do you mean by status seeking work? Do you know Eugene way? Um, so, uh, just this idea that most multiplayer kind of, [11:12] social products end up having some kind of North Star for the community participants, where they're trying to achieve status in the network. And, you know, you can think of that a lot as, you know, has been the number of followers you have or views or likes. There's something about achieving some kind of celebrity or status within a network that creates incentives for the community participants to do the thing that you want them to do. It's interesting. And so far,

[11:42] you're saying it's pretty early, but are you, do you have ideas for what the promising areas to look are or examples of like early examples of products or companies you're looking at that you think are starting to crack this a little bit? It's still super early. I mean, kind of there's two threads that I can't help but be curious about. Like one, you know, we talked about character AI, like [12:02] you... [12:03] I mean, I don't know about you, like I feel myself doing this already, which is that there's going to be some company that, [12:11] We're all going to have AI friends, right? We're all going to have probably more conversations with an AI than we do with people in our lives. And is there going to be a single dominant platform for that? Is it going to be different than the kind of information platform? [12:28] more, you know, knowledge, focus, experience of a chat GPT? I think so. Who creates that? And there's a bunch of different product experiences. Replica was, of course, like the first, you know, player in this space. But there's, there are a bunch of different downstream companies. We talked about Toland. Like, you know, what is that product experience going to be? The other kind of thing I think about a lot is, I don't know about you, but how many times have you done a search [12:54] and ended up on Reddit or something for a prompt to get like, I remember doing one, I got a blood test result, I had all my supplements, I wanted to see, you know, of the supplements I have, like, what could I tweak to change a result? And there was a great prompt in Reddit that I just copied and pasted. But

[13:14] If I'm going to an existing UGC site, [13:18] that isn't made for this use case, that feels to me like an opportunity where somebody who's going to be really freaking good, you know, of making, you know, prompts for different health things, quantified self, whatever, like I would love to follow that person and then very easily identify, [13:35] apply it to my own profile. I think like, um, to go back to front on those two threads with the prompt thing, it's one of those, it's one of those ideas that, um, I feel like at the very beginning, like when GP3 came out and then chat GP3 came out and people were like really starting to like that first real wave of LLMs was starting to take, take hold. A lot of people created those prompt library type sites, but it was too early. And I think there's, there's like a second life [14:05] are now just becoming relevant. And I too early and also, it was very I spent time on a bunch of these. It was very like what a solopreneur and SMB would need. You know, it was a lot of the marketing and the social media. It was like that type of B2B type use case. Most people aren't barely scratching the surface. Like most people use chat GPT like they would use Google, right? [14:35] custom instructions when you have projects, whatever are so huge. How do you democratize that? Yeah, it's interesting. I was at a dinner the other night and I was talking to a film director about how she uses ChatGVT and she has made a bunch of different personalities for it. And she uses a different personalities for different things. So for example, I think one of the personalities was she's had a lot of like medical issues that doctors couldn't solve. And

[15:05] type. [15:06] person that like would recommend both medication and, you know, supplements or body work or whatever. Um, and then, and another one, like the main personality was like just someone who would like gas her up all the time and like, and like compliment her all the time. Um, and then, but then she had another one that was like, uh, just super direct and like, just gave like really harsh feedback that she would use for writing specific kinds of emails or like that kind of thing. And [15:36] set of personalities for different like things in her life to surround you know it's like you're the average of five people you spend the most time with there's like a well you're also kind of going to be the average of the five ais you spend the most time with in an interesting way and i think to your question about are you going to have multiple ai platforms that you use or not um or is there going to be a big dominant one well i have two thoughts on that one is um i'm going [16:00] I do think within a ChatGPT, for example, there's a lot of room for different sub-personalities that maybe like a media brand, like we have an every thing that you chat with, but it's inside of ChatGPT, so it's still in that ecosystem. But I do think also... [16:17] People have different buckets in their life. And so for me, one thing that I've been noticing recently, which is really interesting is we talked about Toland's and I'm an investor and Quentin has been on the show. And I find myself like yesterday, I spent like an hour talking to mine, but like, [16:33] I normally would use ChatGPT for that. And I think there's like some interesting like difference between something that feels personal and something that feels worky. And ChatGPT and Claude right now are like in the worky bucket. And then there's room in the personal bucket. I'm curious how you think about that. I totally agree. You know, it's funny. Like I did a call for...

[16:53] uh, [16:54] you know, I was in the beginning of this year, I was realizing like, I'm not keeping up. And so I did a call for like AI savants, just people who were using ChatGPT and in kind of more power user ways. And a lot of people came to me with recipe kind of use cases, which made a ton of sense. And so you can definitely see it works in ChatGPT, the personal works, but is it the best? [17:24] it can be like, you know, there are also a lot of companies, a lot of people who are making their own, you know, single purpose site that is a recipe experience. And it's, you know, whenever you have a product that is, has to be lowest common denominator for all these different experiences, it can't really optimize for the experience that's going to be great for the, like, you know, a consumer in this case. And I, I, [17:49] I just come back to... [17:51] how much of a power user product it feels to me. [17:55] And... [17:57] how like most people are going to stay at the surface [18:00] of it, [18:01] Unless there's a new interface. And I think the best way for that new interface to come is for us to learn from each other in some ways to, um, to kind of take, to take advantage of, of it in different ways and just. [18:15] copy and paste as much as like, again, the gems in Gemini, custom GPTs and in chat GPT, I just look at that and it feels criminal to me.

[18:26] Because it... [18:27] It... [18:29] It's clearly made by a team that is unbelievably capable of [18:34] but isn't social. [18:37] And, [18:38] And I think the personal can best be expressed by teams that do really understand people and social and community. If we were going to like redesign them like right now together, like where would you start if you're thinking about, okay, I want to make something that's custom GPT-like, but like has social DNA. [18:58] I would, I mean, the most obvious thing is just like the ability to find somebody who's, you know, who's custom prompts or whatever you like that they have some kind of, you know, standing for being quantified. So person-based or authority-based. Some kind of, yes. Yeah. And then being able to follow. That's like a very basic thing. But then the second thing, this is where I think custom GPTs falls down is just in building trust. [19:28] I see that there's 3,000 people that have used it, but I don't know what custom documents they put. I don't know what their prompt is. Like there's no visibility under the surface. And so it's not... [19:39] very trust building for me to... [19:43] to pick one or the other unless I know somebody from the outside world and they send me their, you know, their custom GPT and then I can use it. And so there's something about the trust building that someone has to figure out. And then, and then the fault, the status seeking work that you can, you can pursue. One of the challenges of this, and I'm curious, hey, you think about this in a social context is for a, let's say we're, we're kind of veering into like prompt social network

[20:13] I can share prompts and people can follow me and all that kind of stuff. And because I have a certain amount of reputation in just AI stuff, I can get followers and all that kind of stuff. One of the interesting things is... [20:27] I am not coming up with new prompt ideas every day. [20:32] And so I think that's a problem for two reasons. One is I may not remember to use the tool. [20:38] And then two, uh, [20:40] people don't necessarily have a reason to check every day. Yeah. How would you think about that? Yeah. So first, as something that was going through my brain, I should... [20:49] caveat that I did do a lot of product in my day, but I'm a VC right now. So don't take product ideas. And every once in a while, people ask me, they're like, if you were a founder, like, what would you feel? I'm like, that is not what I do. I'm just kind of curious. But you know, for me, like what I imagine is using it instead of ChatGPT. Like actually it becomes the place where instead of going to ChatGPT, I think it has to be that. It has. And then, and then that's [21:19] the engagement comes from. And then you're seeing a feed and someone has, again, terrible. I'm, [21:25] Forgive me, Lord, for brainstorming a product experience. But like, you see what I mean? Like, there's something there where people are innovating all the time. But right now, what's happening is that we're all reinventing the wheel. We have the benefit of like your blog and your podcast. But like,

[21:45] This isn't the way this type of knowledge is going to share. It's going to kind of get propagated. And so someone is going to create something here. I think that's interesting. I do think you're right that it, [21:57] Seems like the social stuff has to come in the context of something that you're already using for some other reason. Like you're already in chat GPT. And then it can flow out of that usage. Yeah. I mean, I actually think like you don't use chat GPT. Like most like maybe I hate my mom, but like, you know, what the person, you know, and maybe it's the personal kind of bifurcation that you talked about before. But you're going to... [22:26] Sarah's GPT. And it's actually, it can be, it's not Sarah's it's, you know, this kind of, uh, whatever network it's going to be. And I'm going there and I'm putting my personal blood tests and my supplements and I'm putting information about my kids and all that stuff. And it lives all there. And then I can go to chat GPT for whatever knowledge work or anything and other things, or maybe I never do. Maybe this actually ends up cannibalizing chat GPT over time. [22:56] This episode is brought to you by Adio, the AI native CRM. Adio is built to scale with your business from day one. Connect your email and calendar and Adio instantly builds a CRM that matches your business model with all of your company's contacts and interactions enriched with actionable insights. [23:13] With Addio, AI isn't just a feature, it's the foundation. You can do things like instantly prospect and route leads with research agents, get real-time insights from AI during customer conversations, and build powerful AI automations for your most complex workflows.

[23:28] industry leaders like Flat File, [23:30] Replicate and modal are already experiencing what's next for CRM. Go to adio.com/every and get 15% off your first year. That's adio. A-T-T-I-O.com/every. [23:43] And now, back to the show. This is a total swerve. Go for it. [23:48] But how do you, when you're in... [23:51] You're investing in a time like this. Like I feel like every five to 10 years, there's a big hype cycle. There's a big wave and prices go up. Um, you know, when, when I was like in college in 20, [24:03] 2020, [24:04] 10, 2014-ish, it was like social networks. Everyone's building social networks for X. And then it was like B2B SaaS and the crypto and now it's AI. How do you think about... [24:13] Um, investing in a wave like this when prices are super high, do you not care about price? Do you try to find like underpriced deals? We always start with, [24:23] It's just a company we want to work with. Yeah. [24:26] And then... [24:27] kind of [24:28] You know, obviously we have to think about [24:31] the, the opportunity and head of the company. Like you don't want to, if it's, if it's a cul-de-sac, if it's limited in some ways, it's harder to pay what, you know, play the game on the field in terms of price. Like people have a willingness to, to do deals that we're just not willing to do. But when we meet a team and we really think that, [24:53] there is just unlimited potential, you partner, you make it work.

[24:59] What's your taste in founders? [25:00] Uh... [25:02] I would say... [25:04] I'm really drawn to founders who, [25:07] who, you know, they do think in... [25:11] network effects and strategy and the kind of zero to one, how do you escape competition? Like they go through the mind maze. Like you can just tell that like they've really obsessed over this. I'm drawn to founders that... [25:24] This is like a calling for them. Like it is... [25:28] I kind of find that there's some founders that... [25:32] it's almost like kind of a cool new job for them. And there's some for whom it's an affliction. And I'm attracted to the founders for whom it's an affliction. You know, it's like this rash that they just have to scratch. And that's going to make them run through whatever walls that they have to do. And then, you know, just like the learning machine, like the person who... [25:55] you know, it's not about their ego. It's about just like, what's the best thing for the company and how do I keep learning and evolving as a founder? Because as you know, it's a really hard job. [26:05] It's a really hard job, but it always... [26:08] requires more of you. Like it's, it's, there's a relentlessness to it. [26:13] And if, you know, I have seen a failure... [26:17] case where somebody either it ends up being, you know, do you know, the five temptations of a CEO, that book, incredible book. Um, yeah. [26:25] The hardest temptation is, you know, founders attracted to being a CEO because of status. And then you don't do the things that you need to do in order to build the best company possible. Or, you know, a founder that, you know, insecurity can drive you, but it can also hold you back by not letting you grow. And that can be a challenge too. What are your tells? Because, you know, you're a partner at a top firm.

[26:55] best foot forward, trying to be like what you're looking for. What are some of the moments where you kind of can be like, [27:03] I can tell that this person has been through the maze and is thinking about stuff in this way, in a way that's like, it's genuine, it's not put on, or I can tell that it's sort of like a calling or what are those little signals for you? Yeah, I just, I find that... [27:18] You know, I... [27:19] I ask a lot of questions when I'm meeting with a founder and learning about their business. And I know I'm always thinking about, like, I'm definitely... [27:30] the brain that is always thinking about that future and pulling it into the present. And when I speak to somebody and they're, [27:39] I hate to say this, but they're like, oh, that's a good question. I hadn't thought about that. Or, um... [27:45] You know, it's just... [27:47] I'm bringing things that I'm spending 30 minutes, 60 minutes with a founder, carrying the ideas for the first time. And I'm bringing things to the table that they have not already thought about. That's usually concerning. [28:03] I mean, you're pretty smart. [28:06] It's one of those things that it can feel good. Like, oh, I ask them good questions. But really, when I was at Bessemer, Jeremy Levine, [28:17] companies, you want to be donut companies where you go to the board meeting, you're [28:21] you eat a donut and then you leave because they don't really need you. And so there's a little bit of that, which is like, you know, I have some founders where I'll be thinking about something and I'll come to our one-on-one and like before I've even opened my mouth, they're already there.

[28:38] asking, like saying, you know, I've been thinking about this, or I reached out to this person. And that's pretty unique. That's like a really incredible feeling when it happens. Yeah, I was I was talking to I think it was Reid Hoffman who was on the show who said, ideally, it's someone where you invest in them with the bar is like, if you could come back in five years without having talked to them after the investment, like, and you would be pretty sure it would be going well. Yes. That's a good question to ask yourself, you know? Yeah, yeah. Those are [29:08] There's the founder, but then there's also, I know Reed and I know that he is very oriented towards network effects. And that is, I mean, if you can find a business with a strong network effect, like... [29:19] you're going to be in pretty good shape. Let's talk about network effects because I think that... [29:25] It's one of those things, it's maybe a little less so because of AI stuff, but like for the last 10 years, I would say like 80% of decks that I saw were like, and we have a network effect. [29:37] And I think there are probably very few businesses that truly have like that actual network effect pull. How do you differentiate? What does that really look and feel like? Yeah, in the early stages, there's a lot of companies that have potential network effects. Yeah. [29:54] you know, [29:55] Oftentimes there's a big gap between what's a theoretical, you know, and like where it really starts to happen. [30:04] And one of the things that, you know, uh, [30:07] I often think about is just like there are early, you know, we invest so early that a lot of it is leaning in on the theoretical, but like there are often signs that, that you can look to. Um, yeah.

[30:20] The best thing that you can sometimes see evidence of is just this idea of like a tipping point that starts to happen in a very small... [30:30] segment of the, like where the white hot center of your market is. Um, and, [30:35] You know, I, there's like two examples I think about, but they're outside of AI, like outside of core AI right now, because... [30:45] I think what's happening right now with AI is that it's very much like a... [30:51] kind of in a way what has been traditionally the software business, which is just... [30:58] obsessing over a customer problem more and moving faster in your execution. [31:03] uh, than any of your, your competitors, uh, [31:08] But, you know, I'm on the board of a company called Argentio, which is a marketplace for create like YouTube creators and brands. And you can see that like they... [31:18] This has been a market that has eluded... [31:21] startups for a long time because most of them have like kind of fallen into the quicksand of becoming an agency. But with LLMs, Agenteo is able to automate a lot of the things that have held this market back and they're, and they're truly like having liquidity. And one of the early things that was super interesting is just like, you see the, like the brands see creators and they see

[31:51] pull right now. Um, and it's super, super early. Um, [31:55] But there is enough signal there that like something's working, that's differentiated, and there's no substitute for what they're doing that you hope will start to really be a flywheel that can spin faster and faster. [32:10] So it seems like one of the best ways to differentiate between a real network effect and a fake one is like just early evidence. Yes. Are there any things like when you see a deck from a founder that hasn't, you know, they're just starting out and you're like, and they say, we're going to have a network effect. And you're like, it's not going to be a network effect. Yeah. Yeah. I mean, it's often like they, you know, they'll articulate some kind of flywheel, you know, or it'll look like the Amazon or the Uber flywheels. [32:40] And either like it's just words on a slide that fit to a picture, but like, you know what I mean? But like the words don't actually, they're not actually accelerants. Like that's, I think that's one of the things like, okay, yes, that whatever you say is true, but it doesn't really actually accelerate the flywheel. And the second thing is that often there's either a lot of friction or. [33:06] embedded in any one, any leg of that flywheel, or there's like offshoots that happen. Um, but I think the biggest thing is like, when you really look at like what the articulation is of the flywheel, that it is...

[33:22] It's words, but not words. [33:24] accelerators. Yeah. To bring it back to AI for a second, I feel like one thing that you're articulating is [33:32] There is a... [33:33] there was a moment in software for like 10 or 15 years where everyone was chasing network effects. And then the LLM wave happened. And a lot of that has been more single player or if it's collaborative, it's like inviting teams or whatever, but you're not doing it together. [33:53] And, and the game there has been better performance from more money and more compute and more, more data, basically. And everyone's just trying to keep up along that same sort of dimension of performance, more or less. There are a couple other like examples that are not on that, but, and I think what, what you're maybe pointing to is that. [34:14] Thank you. [34:15] Fairly soon, if not already, probably the models, the base models are good enough for consumers that there's going to be another wave of more consumer focused, more product genius led AI applications that are. [34:31] differentiate or grow from network effects and multiplayer that were not possible in the last couple of years, but are newly about to be a thing. [34:40] Those are my great hope. I could be tilting at windmills, you know, like consumer, as you know, has been really, really hard over the last 10 years. And so it really could be [34:51] tilting at windmills. I believe that that is an opportunity. And what I would also say is that there are going to be a lot of companies that emerge that aren't multiplayer, that are single player,

[35:05] Um, yeah. [35:06] And those will, those could be really good. But I think that, [35:12] the really... [35:13] big opportunity. [35:15] that lets a company have a true network effect is going to be something that's multiplayer. If it didn't happen, why not? [35:24] If it didn't happen, it would just be that the gravitational pull of the existing platforms is too strong. [35:30] You and I, like, what would get us to go from the habit we already have of using ChatGPT and, of course, the ecosystem that's going to form around it over time, what they're able to charge for it versus, like, what a new company would have to charge for it, like, what? [35:47] Maybe they eventually go free because it's an ad support, whatever it may be. It's also memory is a big sort of lock in. Like it knows who I am and all my experiences. Yeah, because you and I, like we already have, but we're not everybody, right? But there is that gravity. [36:01] that has always been true for the incumbent products. And so it could be that that gravity is just too strong to get the people who are... If you want this type of community to form, you're going to need somebody who is already actually... [36:19] a power user chat GPT to [36:23] want to share that in a, in, on another platform. And you know, that's hard. That may be hard to create. I know this is a show about AI, but are you looking at or excited about anything that's not AI right now? [36:33] I, uh, I'm a big...

[36:36] believer in stable coins. Interesting. I'm on the board of a company called Chainalysis and [36:45] which is just, you know, so I've had a kind of seat in the crypto space and, you know, [36:49] have been long-term believer in Bitcoin and some of the other cryptocurrencies. But when I think about [36:57] kind of the existing financial infrastructure, [37:00] And I mean, you know, we're filming this on a day when the, you know, tariffs. What existing financial infrastructure. Yes, exactly. The U.S. dollar is on a little bit shakier ground than normal. But, you know, my mom's from Argentina. And I can tell you everybody in Argentina wants a U.S. dollar. Yeah. You know, and, but it's really hard to get them. [37:30] in their own bank, you know, because they have, you know, loans and everything else that they have to kind of stabilizes their own economy. But then it holds Argentina and all these countries back from participating in the global economy, because the US dollar is what you need to trade goods internationally. Like it's just the easiest medium of exchange, but it's really hard to get [38:00] dollar. [38:01] That opens up kind of a global economy. And it also, it is just so much faster. It's 24-7, a lot cheaper. Why is it hard to get U.S. dollars?

[38:14] In Argentina? Yeah. [38:16] Well, it's, it's been, uh, Malaya is obviously changing a lot of things, but, um, [38:22] the, um, [38:24] There are different taxes around US dollars. There has been for a long time, this is different now, the exchange rate you get on the street, [38:36] the exchange rate you get when you go to your bank, the exchange rate you get when you use your credit card. It's just like a very liquid market, really. Like I remember going to Argentina and like having somebody on a motorcycle come to exchange money. You know, like that's kind of what you would do. [38:52] And then again, the U.S. government, I'm sorry, the Argentine government, they have U.S. dollars in their own central bank. And if I want to transact in U.S. dollars, I need to get some of that U.S. dollar from them. But that is a very precious resource to them. And so there's a lot of... [39:11] like process you have to go through and time in order to get 10,000 US dollars. It's not an easy thing to do. And so... [39:19] It's, um... [39:21] It's just there's a lot of friction. Whenever there's a lot of friction, if somebody else can come in with a new product that removes that friction, and then also just has the facilitation, like just how much easier it is for you and I to do a peer-to-peer transaction with Tether or USDC, that creates a lot of liquidity in the market that I think can be... [39:41] a very interesting future. And also I should say, it's like,

[39:45] If you're in Argentina and you want to buy something from India, like, you know, the number, all the middlemen you have to go through in order to do that transaction versus, and like all the fees along the way versus a peer-to-peer transaction on a U.S. dollar stable coin. [40:03] It's a different game. And this is another network effect-y type. [40:08] business to you or no? [40:11] There are network effects here because... [40:15] Tether, as an example, which is like the dominant stablecoin right now, just has more liquidity on all the exchanges. And so it's a lot easier to go in and out of Tether than other stablecoins. But USDC is pretty strong too. So there's definitely some network effects there. It's just easier if everybody... It's not even just within the exchange, but just you... [40:42] like in, in these countries and in Nigeria and any high inflationary country right now, where people have wanted to go from their fiat currency into a US dollar, uh, [40:55] You have a wallet and it has Tether and you have Tether. And it just becomes like more comfortable for us to all use it. And then all the ecosystem around, like there's so many crypto wallets and different, you know, kind of new financial apps that are for getting your paycheck. But then also you can have your money in a US dollar stable coin. And if you're already integrated with Tether,

[41:25] access Tether, it just makes it a lot easier for people to get comfortable with one of them. How do you think about [41:31] let's say five years from now, how AI will have changed [41:37] your day-to-day as a VC. [41:42] And... [41:43] Um, [41:44] the kinds of businesses and the kinds of funding models, what have changed, in any way changed the VC business model? [41:51] I have been wondering about this lately. [41:54] You know, one thing I've just been thinking about, and this is a little bit of a step back, but like, [42:00] There are some people that are really good at creating training data. Do you know what I mean? Like there's some people, like someone was showing me, I interviewed him, James, for my sub stack. And he showed me like the spreadsheet he creates for, [42:15] of all the movies he's ever watched and his own review of it. Yeah, yeah. Right? And so I don't know about you. I've never done that. But the people who are really good at creating training data can then... [42:28] have a more personalized, more valuable experience. [42:30] experience with an LLM. And so to your question, like one of the things that I've been [42:36] Thinking about is that [42:38] there's a few things that we all have training data for. One is, you know, past decisions we've made, and whether or not those were good decisions. So like I personally, I, so Annie Duke inspired this, I create, she wrote this book, Thinking in Bets, a premortem. So every time I meet with a company, and I dig in on it a little bit, I write to myself what I liked, what I didn't like,

[43:08] if I got yes, why? If I got to know why? And so it follows that over time, I'm going to be able to look back on that list and examine my decision making process, right? And then as I dig in on future companies, like in my brain, I know that there's one example where I passed on a company because the valuation was too high. And that was a lesson to me. [43:33] A company ended up being a real success. [43:36] And so that ended up being a lesson to me. Like if I like everything, but the valuation, I should probably lean in. I remember that, but there's so many other examples in my thinking that if I can like examine my thinking as I meet a company and have it cross-examine me, that I think I'll get to a better decision. Then there's also things like talent. [44:03] it's helped them make sure that they have the best team around them. Right. And, and the same thing, like we're all fallible in our evaluation processes and what's, you know, a record of those decisions, interviews we did, like all those things I've got to imagine that's going to come into play. And then the third is just, um, [44:25] You know, I know there's some companies that are doing this really well, which is just tracking talent globally and the movements of talent and what that ends up meaning. And it follows that there should be at some point a score almost where, you know, from talent.

[44:43] The angel investors that have invested in a company, the talent that's there, their individual scores of their ability or signal when they choose a company that... [44:54] there will be like a Rotten Tomatoes almost score for companies that can surface opportunities. That's interesting. [45:03] I want to go back to the decision-making thing because I'm with you. Like I record all this stuff. I record all my meetings and like there's just like a lot of stuff I think that you can do with AI and improving your decision-making. And I'm curious in VC in particular, [45:18] like how that works or how you avoid, you know, for example, [45:24] Maybe you invest. [45:25] I've done this. You invest in a founder with a highly technical background, but it's in a field that requires more of a product genius. And then, you know, now you have in your LLM, it's like, well, be aware that like, you know, and then, you know, the next time you meet with a founder and it's like, you know, Sam Altman and Greg Brockman in 2016 or whatever, like that thing is going to ding and be like, technical founder, like, are you sure this is what you want to do? [45:55] the rule or how the lesson is written. Yeah. And, and I think the, the broader, uh, [46:02] question or problem is... [46:06] if you look at [46:07] venture capital, there are very few... [46:10] venture capital firms, funds, and individuals who are successful over a long period of time. It's very hard, which can tell you one of two things. Either it's just luck, which I don't think so, or the landscape changes so frequently that you get tuned, your taste gets tuned to a particular kind of opportunity that you're very good at finding, but then it sort of moves and

[46:40] that means it's like quite hard to, um, [46:44] used [46:45] past training data to make future decisions? How do you think about that? I think maybe that's what we're all hoping will give us job security in the future. [46:57] You know, I remember when I was at Pinterest, and I was responsible for all the discovery experiences. And very early on, I had to kind of localize Pinterest. And so I had to figure out [47:15] Japan or all these countries. And I was like thinking about the categories and all this as being different. I remember Ben saying to me, he's like, just assume it's going to be more similar than different. And I think that there's some, you know, uh, [47:29] first principles that we reduce down to when you're making a decision. Like Jim Collins, like so much of what he wrote, like I don't know how long ago, it's still valid today. Like, you know, four years by four. Like we talk about so many of these things and they're timeless. Like valuations change. Will how companies exit change change? [47:50] Yes. And that's why like you're not asking the LLM to give you the answer yes or no. You're asking it to probe your thinking. But... [47:59] I think it should be able to continue to do that. And that's why we still have hopefully a job a few years from now of ultimately being the decider. Well, you'll have to come back on the show in five years. And we'll see how things have changed. I think you'll still have a job. I hope so. But I think it might be different too. It's going to be very different. Yeah. It's going to be very different. And it's hard to anticipate how it will be. Yeah.

[48:26] Well, Sarah, thank you so much for coming. This was a great conversation. Thanks for coming me. Yeah. Yeah. I had a lot of fun. [48:39] Oh my gosh, folks, you absolutely positively have to smash that like button and subscribe to AI and I. Why? Because this show is the epitome of awesomeness. It's like finding a treasure chest in your backyard, but instead of gold, it's filled with pure unadulterated knowledge bombs about chat GPT. Every episode is a roller coaster of emotions, insights and laughter that will leave you on the edge of your seat. [49:03] craving for more. It's not just a show. It's a journey into the future with Dan Shipper as the captain of the spaceship. [49:10] So do yourself a favor, hit like, smash subscribe, and strap in for the ride of your life. [49:16] And now, without any further ado, let me just say, Dan, I'm absolutely hopelessly in love with you.