20VC: Why To Win in AI, Investors Need to Change Their Approach | Why VC is Run by Principals and Associates and is a Broken System | The Bull Case for Anthropic & Whether Deepseek Changes Their Strategy with Nabeel Hyatt @ Spark Capital

23:11You know what later stage, you know what later stage wants. But you're also closer to public markets. You have a closer understanding of what makes a fundamentally great business. Again, this is all amazing and wonderful pitches for 2021. Who knows what the public markets are going to like in AI in seven years? What are you doing? And so if you spend all your time trying to internalize what the public markets are thinking about this month, you are absolutely going to make the wrong choice. What do you think of these VC armchair investors who love macro? Do you just got a hope that the cycle comes a background for them in exactly seven to 10 years when those companies want to go public and if it does, they'll do great.

23:46You said service as a word quite a few times. Keetra Boy says on the show and very publicly, the best founders don't need the help of the VC. You agree and how do you think about that in conjunction with service? Look, I had a bunch of venture that I raised as a founder. I raised eight rounds of venture capital as a founder. I never had a horror story from a VC. I had nobody who was like terrible and horrible. I basically had mostly VCs that were fine. They fall into the Keith for Boy camp like they show up to board meetings. It's okay And I'm just gonna run my company. I just imagine that in any other context of life If I had somebody who was on my team anyone else on my team and they were fine What would be the feedback?

24:27Marge

24:35Actually it wouldn't be fine. It's not lost. It's not just happy good matter If you're executive assistant to your VP of engineering or you had a product was fine. So why are we accepting? You'd be your band.

24:51Why are we accepting mediocrity at that level? I think what you want is people who are engaged, who look at the details, who are invested emotionally, who want you to win and are doing the detail work to be able to not just give you armchair VC advice that they got from the one other board meeting they were in this week with the other hot company that they're in because every single startup is a different journey. So you can tell you get out of the covers of what's really going on inside of this org, AKA, how do these co -founders fight? What are the problems inside of the executive team? Then you're going to give different advice to somebody if you really understand them.

25:25So it's worth understanding them. If you got in race, look, I've had really tough rounds to raise, both as a founder and as a VC who's back to seed company or a series A that's not working out well. So look, if you cannot find that amazing and wonderful VC that you think is going to be deeply engaged and use your product, then go for the no -op VC. Go for the VC who at least do no harm. Oh, yeah. Hi Price. Fuck off. Fine. I get it. But where you flip that to being the goal, mediocrity is the goal? That's not the goal. Every time you have an ability to have an investor or an employee or really anyone enter your orbit as a founder, your goal should be somebody who is going to be obsessed with you and think about your mission and try and help you.

26:06otherwise you shouldn't be engaging. And if you fail at it, fine. But many second time founders I meet say, listen, the thing I've learned about the first time is, what I want from my venture investors, good money, good terms, get out of the way. And again, I just say they're setting their bar too low. That's the bottom line. Like they're settling for mediocrity because they're afraid of risk. And that's why it's important you do more marketing. Mm -hmm. There we go. I win this debate. Can you do that at scale? I mean, you do what two checks a year? Two to four at the most, but yeah. Yeah, like two checks here.

26:38You can't do many. You can't have high service and high volume. Absolutely not. It's a model. What do you do when you lose faith in the founders? Hopefully you have had many, many conversations before you get to that point. When I use a word like service or falling in love with a founder or being dedicated to loyal, that doesn't not come with tough love. That doesn't not come. You use the marriage analogy. If you're just smiling all the way through marriage, you're not executing it right. You need to have tough conversations. And so, yeah, sometimes you're having a tough conversation that you feel like that person has lost their way.

27:09When they've lost their way, what did you not see that you should have seen? It hasn't gone well for mostly few reasons. The first is that their conflict avoidant, and I didn't pick up on it early enough. And it's very hard to pick up on early because you're going through a period where ideally, you love them, they love you, there's not that much conflict. Maybe until, like, term sheet or something like that, but, like, there's not that much conflict. And then you go through 25 conflicts in the first month of the company or three months of the company and you realize that they're conflict avoidant.

27:37They're not facing the problems of that company because every company has a thousand problems obviously. That's the first one. You try to read for it, but you can get it wrong. The second one is that every founder is like, this is one of those things that I did not have as a framework 10 years ago, but after you make like a bunch of mistakes and you look back on things, like they can become more clear. Every founder you want to move really, really fast. We all talk about execution speed, you want it. So you can imagine somebody on the very, very far end of execution speed. We also want them to have taste and judgment.

28:04We also want them to be, especially in the world of AI, and only taste is going to matter in the future because execution is just going to happen, right? So these two things are directly in conflict. If you are always the shoot first, ask questions later person. You probably are not really deeply introspective about the choices that you're making. You're just shiny, penny running after whatever happened on Twitter yesterday. And if you are deeply, deeply, deeply full of taste, you didn't ship anything. Like you just sat naval gazing forever trying to find the perfect thing. And so the casting for a founder needs to match the opportunity of that startup.

28:37You have to have good taste, especially in this world, and you have to be very fast. But where you are on that spectrum is incredibly illuminating. You mentioned earlier, you would totally get a company that was like very, very competitive, end of 35, should I invest in this thing. and it's like, well, that's an execution play. That person needs to be in the top 0 .005 % on the planet of being able to execute, because the roadmap ideally is probably pretty clear, and even if it's not clear, some competitors about to do it tomorrow. And so you could see it and you can run faster than them, and if you can aggregate everybody's innovation that's happening in the industry across all 25 of your competitors, you will win.

29:15There's many other situations, especially for the deals that we do at Spark, which are often creating a new market that didn't exist before. We have to have a crowd of real -grown owners of the world, respectively. I've put in the like incredible taste. Yes. I'm sure he's graded execution too, but there's real taste. Real taste. And I'd say, Grinnell is a great example of a situation where everybody else competing in that market would have taken the execution angle. They would have built a MeToo product, slight arbitrage. Looks like fireflies or any of the other things, with like a little bit faster or maybe a chatbot in a slightly larger box or whatever it is.

29:48and they would try to run faster than their competitors. Chris, his product changed completely from the seed round to the series A, which we led, like complete reset. Even though the internal metrics were okay, and it was because it didn't feel right to him, he could kind of like self -inspect and realize it wasn't working, and that takes real taste. But he's in a competitive market to be clear, and he knows it, like he's got to be pretty high on the execution path. That's why. Oh my God, he's a whole combination of both. And that is where I think the best founders can manage and understand at any given moment, what is the muscle that I'm using?

30:22And how am I using it? I think the mistakes for founders is realizing that one, I like got them wrong on one of these accesses, you know, quite a lot, or I cast it correctly. Like maybe you cast it, somebody was very execution oriented with a good amount of taste, and then the market flipped something crazy happened. Well, this is what I was gonna say, which is something crazy happens, and the sustainability of value -stay seems to have completely eroded. And what I mean by that is something crazy happens, and OpenAI release a new model, and it just completely kills Grenola overnight. Well, the data provider example that we have, I don't know if any of the large foundation models decide that actually that's the prime easy market for them.

30:59They have all the overnight the data providing goes and it just goes rolled into their core products. Yeah. How do you think about sustainability of value in such a changing world? I think you have to find a founder who is continually innovating. You can ask all the simple questions about barriers to entry and all the rest of it and have some decent answers. But the truth is, if you are not reinventing yourself, everything I did deep sleep tells you the evil no idea about bars. I'm saying you have to justify it to yourself to go to sleep at night and maybe have some base where it says like, for right now, this is what I think the barrier to entry is.

31:31This is how I think their next year is going to be good. But some kind of compounding effect where we think no matter what happens after this year, it's eBay. They just launch a product and 40 years later the product click basically exactly the same and it's just fine that those ages are not right now. Those might metastasize inside of smaller vertical markets and AI in the next couple of years, but by and large, it is a sea of speed and taste at the same time right now. Do you give a shit about market size? You said about kind of the market creation angle there. How do you think about market size?

32:01It's such a simple heuristic for investors to fall back on. We don't talk about or look at market size at all. Unless it's sometimes there's a, you know, a confirmation that it's a small market. If the guy's starting a nice cream truck, then like it's probably not for us. But usually if you're creating it, but with respect, yeah, I would still push back and say crumble one of the fastest -grown businesses in America, which is the cookie business. Yeah, that is looking like a venture -sized outcome at this point. Starbucks was also a venture -backed business. I didn't say they can't be good businesses.

32:32I said they're not spark businesses. Again, we're not trying to canvas the entire world for every single possible thing that we can invest in. We're not trying to be the Britafilter of venture capital firms that has to look at absolutely everything. I need to do a good deal a year. That's the job. It's not that hard. It's incredibly hard to execute. It is a simple thing in its essence. And so if I try and win every single war across every single front, I will be average across the whole board. So we try to be good at what we're doing. We try to partner well with founders who want that product.

33:05We try to look for new market opportunities, which by the way in the world of AI is, you can imagine why I'm a kid in a candy store right now. Like the most excited I've literally ever been in my entire career, including as a founder. Like it's just an amazing opportunity. And if you're asking what makes a new, what makes it new market opportunity, I think it's, you're looking for a new behavior, looking for a new behavior where when you try it, it just like sears into your brain, you can't stop thinking about it. That's, it sounds, again, sounds simple. But if you just do that, is this really a 10X better product?

33:33You just don't see that many of those. That simple thing, you just don't see very often at all. You have such a product -centric investor. I spoke to Kyle at the bot company. I spoke to Andrew at Descript. I spoke to Ritu before. I really stored the shit out of you. Very impressive. But everyone was like, his product -centricity makes him such a unique investor. And I thought it was just so interesting because product is the one transient element of investing. If you think about market, people on product, it's the one thing that will really change. I mean, market can do, but often less So people, it's right around the same market.

34:05Why do you focus on the one that is so transient? So you used market people and product as your three cores. I think if you just look at right now, market, do we understand any of these markets? How fast are they all changing in the world of AI? They're all shifting like crazy. And who knows which ones are going to become commodity markets with absolutely no margin whatsoever anyway. So if it's a big market, maybe it was a big market two years ago, and it's about becoming a really small market and the same thing in reverse. People is very interesting and I think there are firms that do really good job at just making people bets I think you have an instinct about people that just you get over the line and you make your bet on people Thank you.

34:41Yeah, my turning down of Chris it. We know they're in the pre -seed chaser You can't be a snobilt. You can't be a hundred percent all the time. Yeah, or Alex it deal. Yeah, yeah, or Christina Ivante I have my fair share of issues. I just managed to bring the real best Yeah, thanks. And I don't think of product like, I'm I, the product master. I think of product as an instantiation of what the founder does. So let me recast it a different way. How do we separate hoxters from good executors? Because they're here pitching us as VCs. The way we separate hoxters who do a good pitch from people who are real executors is you look at the thing that comes out of their hands.

35:19You look at the thing that this petri dish of humans have created in the world, and you try and evaluate and ask questions to get that. So when you say I'm a product investor, I would push back slightly and that I've never evaluated a company by like looking at the product using the website and being like, oh, this person should have a $59 check. I don't think that's right. You look at the product and you try and learn about the humans behind the product by evaluating the product. It's the questions that you ask somebody like Kyle who started cruises now doing botco and you ask him why he made the decisions he made in this thing that you were using.

35:49And that's where you can get a sense of who this person is and what they're going to do from there. I had an absolute fucking meltdown with the team the other day because I have like a seat you're going to hate this. But like, you're a freaking hate this. Like a CEO template for how we analyze CEOs and they took that CEO template and put it on a CPO and asked the same things. And I'm like, that is criminal. We changed the template entirely to what product decision are you most proud of? What would you most like to build? But you have constraints that mean you can't build it. What are you most embarrassed about building?

36:19And this is actually what I don't actually care. the specific analysis is the way that they think around those questions. That's right. It's the way we conduct any deep level investigation on a person. You're not asking them, like, give me your Tam answer and give me your margin answer. You're trying to figure out how they think about the things they're doing. And what are the best, if you're talking about an early stage startup, what is the thing they have thought about the most? They produce a Tam slide or they produce something, they did that for the deck for you. So they put what, two hours into that, three hours into that?

36:47or nowadays they probably sent it over to Gem or some other product and just had it spit out those slides. If you're trying to get to what they have been obsessed about, it's like this thing that you're using, they've spent the most time thinking about. So it's where you're going to get the most insight into the humanist. How many companies do you meet a week? Almost a no, 20, 30. 20 to 30 a week? Sometimes. Yeah, email. Email. But I own a cool. I don't do that many on a call. I probably wanted to a day. Okay, once to a day. How long do you have? I think the one hour call is the worst call anywhere, because it's like too short to get a real read and too long to get the kind of speed dating version of the world.

37:24So a half an hour, I'm just trying to figure out whether I like the person at all and I want to have a second call, right? And then I'd rather go from a half an hour to two hours. So you do the first one and you're just trying to get a read. Is there a kismet? Is there a connection? Is there any chemistry here? Do you feel it? And then after that, it's like, yeah, we should just, let's just go for a walk, man. Let's go have a conversation. Let's go really talk about everything. Will you have an invest if you haven't met them in person? Probably not. There's some world where you met them five years ago, you knew them quite well.

37:51A chrystrum granola is a prior spark founder. And so he was one of the first people I met after I joined Spark, because I was supposed to go sprinkle growth fairy dust on him in New York and talk about growth marketing, growth hacking, and all the rest of those metrics things that I don't really aspire to now. And I said, you know, So, Discord, Jason, I knew for seven years we were founders together before investing in Discord and also smartly passed, why so I'm discord before investing. So like, you know, like a lot of long -term relationships and a lot of short relationships where you passed and then spent time.

38:25Crews back in the day with Kyle, I passed. We did a huge deep dive on why I thought his business wasn't going to work. He disappeared for nine months and wouldn't return my emails. And then he comes back nine months later and he's like, yeah, this is, we've repivoted. We've gone from trying to put, you know, aftermarket things on top of outies, and we're now gonna build a full self -driving stack. I'm gonna show you a demo, you and like, five other investors a demo, because I really liked your last conversation. And so you knew somebody for eight, nine months, you've thought of how they internalize information, and you really know them.

38:53Now that can't always happen with every investor. The last investment I did was a company called Wordware, and that was a very, very fast, very, very competitive. They had term sheets for quite a bit higher. But, you know, I'm getting dinner with the founders, I'm going for walks in the morning like I met with those guys six seven times before we invested in the span of a week You don't have time but in the span of a week because you care word wear Ranolla Descript these real pretty big valuations actually and pretty hot rounds they were I'll price sensitive. I'm not that price sensitive. I mean there's always a number We could go through the deals that we didn't do because the price got away There's always a price where it just doesn't make economic sense anymore.

39:36But if you're... Like, we're looking at a series A now and it's like, we put down like 10 on 60 and the founders like, I want a hundred and then the Vortna's like, we would do 80 and I'm like, we'd do 80 but not 100. I feel like for us valuation is always a test on conviction. Like, if you liked it at 60 but don't like it at 65, like... But 60 goes to 100 and that is different. That is different. But 60 to 80 is not that different. And then 80 to 100 is not that different. Do you see what I mean? We can push it. Harry, this is hard job. This is our job. What did you turn down because of price that you most regret?

40:11Because we run a fund the way that we run, a small number of investors, six investors with a $700 million fund, is kind of broken, Adventure Camp. And so what it means is usually it's not about valuation, usually it's about check size. So in our model, if you really believe in the company and you want to have the ownership that you have, and you believe that they're at the right stage, then it's about whether you're gonna write a $5 million, check a $10 million, check a $15 million, check a $29 million, check. And so sometimes you say valuation, but I root back to maybe that founder is raising around and you don't think they're gonna spend $20 million very well and it will mess up the company.

40:50There is absolutely a belief, for me at least, that too much capital can mess up a company. And so sometimes it's not about valuation, although obviously it's algebra, about these things are all related. It's about, you know, a $25 million around here is probably gonna kill this company. And so if it was a $10 million around I'd be in, and also we had the ownership properly, and it'd be a good partnership, but I just, I think this company will be different with this amount of capital into it. And so the company changes. Which one stands out most? My minus Figma, it was quite a while ago now. It was still pre -launch, so it was trying to write like a very large check pre -launch.

41:25For me, it's that because also there was just like a connection with Dylan. It's not just that it was a large valuation. It's that I think that that journey would have been fruitful and interesting and an amazing way to spend five to ten years of your life. Before we do digging on AI2, you said there about kind of capital inefficiency within companies. One challenge and real concern that I have is a lot of growth investors who have too much cash, but I'm going, I'm willing to pay up because I believe it's going to be a $5 billion company. Fine, I might not get a 5X, but I'll get a 3X and I'm playing a deployment game.

41:56That's right. But it's the wrong actual thinking because you're assuming that it's equitable and that putting in Preemptive round in place will still lead to that five billion. Yeah, but me and you both know if I try and shove cash in before it's ready I could destroy that potential five billion and make it a one billion It's another good example of where we have a different world now that we had four years ago That company will probably raise another hundred million and then they might just die And in fact, they probably will die. Their probability of dying eventually, it'll take a long time because it got a lot of money, goes up.

42:29So then if that happens, if you feel like that company's raised too much capital, you can watch their hiring velocity, you can look at the quality of the people they're hiring, their execution speed, you can see it all teetering, and then maybe invest in something else in the market, even though there's a lot of money in the market. When you say no, don't raise that round. Do they listen? Never. But do you engage in secondary markets actively? No, why not? With the huge influx of private late -stage capital and the continuing delays of public markets, we need liquidity. At some point, you have to deliver cash to investors, I do too.

43:04Do you not think that becomes an ever more important part of our role? I'm not saying you never sell secondary, I'm not saying it never happens. Again, the primary job of trying to figure out a little bit about the future, listen to those founders who have of that little glimmer of the future. Have a beginner's mind enough to be open to it so that when somebody comes into you with some cockamaming idea that was way off -piece from how you thought the world was gonna work, that you're open to shifting to it. Like that takes time, energy, research, it takes trying every product, takes curiosity.

43:35Like the question is like where are your hours coming in the day? And so sure, there's secondary that happens. Sure, there's latest stage valuations that happen. Sure, you can decide to figure out you wanna do growth. Sure, you could run a conference every month. Sure, there's a thousand things you can do, but doing the simple thing at the highest level takes time and energy, and I don't even think I'm good at it yet. So I'm still just trying to get good at my first job before I do the second, third and fifth job. When we think about AI companies specifically, you said to me before, and I love this, there's three categories of AI startups.

44:06I love like a framework. I know you do. So, but we're not in the age of framework, I said to you all, guys, just remember that. So there's the guy with three categories of AI startups Just to say. It's helpful to bucket things and have lenses. Totally, no, very much. We say lenses and not frameworks and then with you. Lenses works well for me, but what is the three categories of AI startups and how should we think about that? This actually is a framework that came in the mobile revolution for us at Spark and then reapplied. So this isn't, this is an old lens reapplied, which is adaptation, evolution and revolution.

44:40And there are versions of this that have existed as people have talked about AI generally. It can I use the mobile analogy to get you there? So adaptation is the obvious like, I'm gonna take the thing and I'm gonna make a copy of the thing for AI. And so in the mobile revolution, this was The New York Times makes New York Times .com on mobile and that's the product, right? And that's obviously a world where 2023 was kind of the big adaptation push. And that was when Adobe Firefly launches. It's when Spotify DJ launches. It's when Canva create, I think it's called. which when like, you know, the big boys came to town with their AI products and everybody had like a year to go think about what they were gonna do after the GPT era and like ship their incumbent advantage stuff.

45:21Like that's all adaptation. Evolution, I think the easiest way to kind of separate it is it's when there's a new workflow. It's once the behavior has changed slightly. The good example of this is in the mobile era is Instagram where you're suddenly doing a different behavior than you used to when you think about Flickr or prior photo websites. It's a new behavior that is native to that medium. This can be done by incumbents and by starters, by the way. Sometimes a really fast -moving startup will do it. Sometimes it's an incumbent. Granola is a good example of this. They are an evolved product.

45:55You are treating, I don't know if you want to call that an AI meeting note software. I don't know if you want to call it transcription software. I don't know if you call it just Apple Notes with AI in it. But it's a different behavior. It's a different way of using the product. I think Repplet agents, the way that they've rebuilt. another really good example of like evolving the medium in a way. Descript is another example. You cannot take the incumbent UI and just slap -descript on. It is a rethinking of how you would do audio and video editing from scratch with AI in mind. So that's that's evolution.

46:27And then the last one revolution, this is the canonical like I'm sure this gets talked about every week on your on your podcast, but this is Uber, right? This is like it's an entirely new platform that would only exist because this technology exists. Where do we have the most and where do we have the least? Oh, we are in the industrialization of startups playbook land where everybody's trying to churn out some piece of ridiculous arbitrage every week in order to get through the end of their incubator and raise their seed round. So we mostly have evolved products that are not good enough or we have adapted products with a coat of paint on top that says AI.

47:04What do you find most interesting? We invest most of our largest exits at Spark over time and most of my the most satisfying work over time has been on the revolution and sometimes the evolution category. So we have no desire to invest in anything that's an adaptation. We're trying to lean towards the more disruptive, the higher risk, knowing that that won't always work out, but at least it's a journey worth traveling. When we think about value of cruel in the new landscape, Kyle at the Bot Company said that, But you've hedged this and you have that's in foundation models and models. And then you also have that as an application layer.

47:37Yeah. How do you think about where sustainable value or crews and the GPT wrapper? Oh, there's no value in application thin layer. I wouldn't call that hedging. I believe both could win. We were in a very early investor. You maintain that to stay with models. Absolutely. Yeah. I mean, we were early investor in Anthropic and very happy about the investment and think there's a lot ahead for it and feel really, really positive about it. Absolutely. And then on the other end, Can you paint the bull case to me with unsurface? Sure. The thing to understand about entropic and open AI, CHEP GP, their direct competitor is obviously, is to think about it in the terms of this adaptation evolution, revolution framework.

48:16I'll just use this thing we just talked about a minute ago as a way to make this point. If you are trying to make the next best model and you're running out of data, then what do you need? You need to understand how people want to use your model. If you want to understand how people want to use your model, then you need a lot of people using your model. And so the fact that those two companies have a user interface which has both insights to how a user would use it, but also, frankly, data exhaust on how people are trying to navigate a model and get through it, gives you insight that no one in some academic lab somewhere just popping up a model is going to be able to advance as fast as you.

48:54You can make a faster algorithm, but you can't get new data and new data exhaust from consumers. But the love of consumer data is probably 10x more for open AI than the new Zanzarope. I didn't say it was the only competitive benefit. But for both of those companies, you said make the case for these. I think for both of those companies, I think that's a major, major case, right? They will have a user interface. They will have user data that will help them be smarter. I'm sorry, there just everyone not like deep -seek sitting on number one is getting more consumer data and more consumer insight than anyone else.

49:25I mean, Xi Jinping is having a data fast. But I mean, respectfully, there's so many that do. I mean, you .com has a pretty good user interface. I wouldn't say it's that much worse than Claude. It's like, it's okay. Do they have enough users and enough growth? No. Do they have enough scale? No. I think you want to own the interface with the customer. And if you own the interface with the customer, you have the chance to iterate on it and interface with the customer and stay of everybody else. So if you just look at Anthropic and you look at artifacts, right, which now open AI is copied. If you look at the next phase of the things, there's going to come out of insights with the customer and those things matter.

49:59It's not just model quality. So I don't think foundation and models are just model quality. There's also your ability to continue to innovate, your taste, your execution speed, and then do you have a relationship that's very direct with the customer, which you can still keep iterating with them faster than everybody else. So if I compare that to somebody who's spending a bunch of money on compute for an economic lab for a very large foundational model, I don't think those things are correct. very well over time. I think you need to be full -stack. Does deep -seek change how open AI and Antrop it should operate?

50:25I just had Jonathan from Groc on the show and he said, if I was Simon Mortman, I would open source today. You will die if you don't open source. I don't know that deep -seek changes that much for the way that I think about the future strangely enough, and maybe that's an odd thing to say when everybody in the world is freaking out about deep -seek this week. I don't know that I ever really believed personally that the $100 billion or $500 billion, $1 trillion dollar training was the only barrier to entry for making these models. In fact, if we had believed that only capital was going to win, then we would not have invested in Anthropic because you surely, Sam was telling us and everybody else, like, we're going to win the capital game.

51:02Capital game is the only way to win. And so there's no reason to build a competitor. So we didn't believe that back then. Or else we wouldn't have invested in Anthropic. And so it's still true today. And do you think Anthropic wins them because of product taste? I think Anthropic wins the same way that every company wins, which is that, I think, They are executing very fast with taste. They are listening to their customers and delivering what their customers want. What do you think will be a bigger business, the API business or the consumer? I think when you're at the front end of innovation, when you're moving really fast, I think you want to be vertical.

51:34I think you want to have as much connectivity layer between the model you're trying to build and the thing that the consumer is trying to wrestle with the world to make happen. You said about data exhaust. And you said to me before about it, I just want to make sure I get the quote right. The data resource is more important than models. Yeah. That being the consumer insights later. Well, I'll give you an example. If you are a descript today, you don't just sit on this amazing and wonderful interface, which kind of changed the market and reinvented what this was. You also sit on every single edit that every single person has done to try and go from a rough cut where somebody might, like me comes on your podcast and says ridiculous things.

52:17And then, you know, cut that down to something that actually sounds like a wonderful production. That's internalizing the wisdom of experts. One of the things that we're seeing right now, in this world, open AI and lots of other companies are paying a bunch of PhDs to buy hand, go figure out PhD math equations so they can internalize the model. I think Web2 .0 was very much the kind of wisdom of crowds. And we're at an age where it's the wisdom of experts. We're not trying to get what the average output of every single human is. We're trying to get what really amazing people at whatever their field is across every field in the world would do in this situation.

52:53And so if you are running a next -gen product with AI deeply embedded and you have the best users using that product, that will inform you to make better products for those users and inform the models that you're building. This is about trying to build a top of market product and then trying to make decisions for, I mean, AI, the promise of AI is this thing, this little alien near computer, is going to help you be smart and as smart as the best person who does this thing. I just walked with Manny Medina from Outreach and he's building essentially Stripe for agents. And he was talking to me about the future of agents and he was saying, hey, the future is actually hyperverteclosed in what would have been non -interesting small markets, but because you were replacing labor, it's actually so much bigger than you could have ever thought.

53:38And so an example is that happy robot, this Andrieson company, which is brokers, cooling truckers to organize loads and transports. Before you're like, that's not very interesting. But actually, if you replace the broker, that's a multi -multiple -indole market. How do you think about the future of an agent economy in that respect? Are you excited by that? Do you spend time thinking about it? I spend a lot of time thinking about it. I think most of them fall into, if you just really think through the second order effects of it, fall into kind of like near -term arbitrage, which might just take that whole market to zero, especially if you're meeting the market where it is today with the models of today.

54:18You need to assume that you still have a second, third, and fourth act in your business because you're going to need to keep innovating, or else how are you not going to get lacked by 25 other competitors who are also going to build call agents into your or a vertical market tomorrow. And so I think you do have to ask some of these questions. Sensibly, how hard is the job to be done? This is a direct contrast. If I just took my 200K check and I just did an incubator and I'm trying to show 10 % week over week growth so I can raise my seed round in three months, then I want something that the models can solve for tomorrow.

54:51And so I might go into a market that, I don't know, a little bit of transcription from AI, it like solves it and we're kind of done. If that's really the extent of your innovation, and you're probably going to be awash with 50 other people who also joined all the other incubators and are also doing the exact same thing. And so your marginal benefit to the world is zero. If you are doing something at the very edge, I mean, I definitely am constantly trying to advise founders and encourage founders to think about what the models might do in six months or nine months and start to chart there. Because it's not a worthy exercise.

55:23And what I mean by that is it's so unpredictable. It's these to nine months of product roadmap prediction for model providers. You'd like? You can try and figure out what it... I'll put it in more simply. But then in case 18 months they decide to change, deep seat comes out, they hit by that. Let me try a different wording. Pick a job that's hard. Venture. No, I mean, if you're a startup, pick a job or decision that is not solved by today's AI perfectly. pick a job that you think it will be worth trying to pursue for the next decade. Because that's the nature of the startup. And so if you can solve it fully, which are the most satisfying things as a founder, I can satisfy this fully by time I get to the end of my three month incubation period in my little seed round and I'll be done, then you probably are gonna get lapped.

56:12If you think the best you can get is an MVP that hallucinates constantly because this particular problem is incredibly hard, but people will pay a little bit and I'll get a little bit of traction, and if I do that, I can make a little bit more progress, and I can imagine working on this project for the next 10 years, and still innovating seven years from now. Okay, so now you're in a right path. And then, by the way, if the model takes an extra three months, or more three months left to hit its level of fidelity, you'll be okay. What's been your biggest loss in the bill? Loss? Yeah. My mother?

56:43I mean, what do you mean if you had a zero? Oh, have I had a zero? I push back, man. I don't know that you should look at your losses. I mean, maybe if you felt like the discusing that you made at the time, if you look back on it, and like you just were in a bad place when you made the choice and you should try to not be in the bad place, the next time you make the choice, you made the choice for the wrong reason. But I don't know, we're in the business of the things that work. So you study your successes? Yeah. How did I find that founder? What were the signals that happened there? What are the lessons I can learn?

57:12What are the things, what are the types of founders that connect well with me, and I connect well with them? What are the market dynamics at that time? and what was the product like at that time? Those are things worth deeply investigating. Are you incredibly bullish about the future of the US right now? I am incredibly bullish about the long term of the US right now. You're in a market dynamic where, I don't understand your markets. Why? Because too positive? No, because a lot of you were like, oh, you know how it's, how it's, how it's, Trump comes in does a load of really efficient stuff. Mark, it's go to the fucking moon and you're still like me.

57:48I'm like, you ungrateful, ungrateful, champagne socialist. And we sit here with the Lego had Chancellor who does negative growth on us. And women, did you take it? I don't know that the president affects the economy in the US as much as you would think that any president affects the economy. I think you would normally be right, except Trump. The confidence that is instilled now in the US public markets, I think is unpowered. And I think - But I'm not investing in the US public markets today. So when you ask me, how do I feel about America? Am I optimistic about America and all the rest of that stuff?

58:27Am I immediate? My pronation of thinking about the world is like, oh, well, why do I think about 10 years from now? That's my thinking. I'm a long, long, long, like I don't get to do anything today. I get to invest today for something to 7 to 10 years from now. But I think a huge amount of people want to move that money to the US. want to invest in the US in data centers, in real estate, in you name it. Yeah. Because of the state of the economy. Sure. I think that trickles down. It does. It does. But what's your point? I'm optimistic. I would be optimistic about the US in either case. No matter who won the selection, I'd be optimistic about the US.

58:57Are you optimistic about Europe? You've got Grenola here. You spend some time here. No. There's some exceptions to every rule. What do you think the challenges that Europe faces then? If I was a founder starting a company, my default state is dead. Things are really hard. It's hard to recruit, it's hard to raise money, all of it's hard. You're pitching the worst of the world that you're dedicating your life to this thing and you're all in, quote. And so if that's true and you're trying to risk mitigate all the things that are going to kill you and it's the age of AI, I don't know why you're not in San Francisco.

59:25Forget the opposite case, can somebody succeed in London, can somebody succeed in Berlin? Like of course they can. But the real question is like, if as a founder, why would you make that choice? So that's the problem. The problem is that more of the people who are actually all in, not just telling you they're all in, more of the people that are actually all in, who are actually trying to do everything on the planet to put themselves in the best case to win and are willing to sacrifice for it, they're going to want to be at the dinner where they're learning about AI people, they're going to want to be able to recruit the best people at San Francisco right now.

59:58And so why wouldn't you do it? They are just the only challenge that I push back on is talent acquisition is so freaking in hot there. Competition for talent is so high. Salaries are so high. Churn is so high. You guys are very promiscuous with your jobs. But oh, well, you know what, this isn't that hot anymore. I'm sorting off to somewhere else that's way hotter. Oh, well, anthropics in you up around isn't as big as the X's. So when moving, Christ, you jump around. So what is that incentivize? That incentivizes the system where you have to keep innovating and you have to have speed or you have to have synthetic growth.

1:00:31There's a downside to I agree. You have to be smart enough to separate those two things. Final one before we do a quick fight. So much for job is like you sit down with the founder after investing and they're like what do I need to get to raise my A? And you're like well and you kind of put the past aside. Just like this conversation. Yeah. And you kind of plot the past to A. It goes back to that. I don't like that conversation. I understand it and I have it. But it goes back to the packaging. It's just putting a ribbon on you and then passing you along. How do you feel about that conversation?

1:00:59of what do I need to get in May, and how do you approach it? Yeah, how do you get to the next round? I usually try and start from asking them a lot of questions about how they think about the future and trying to separate them from the way of VC thinks about the future. Because ultimately, we're listeners more than we are really tellers. Like, I get that part of this job is for us to be tweeting for us to be on podcasts like this and so on and so forth. But the future is invented by founders. The question is, what can you surprise an investor with in the next year that they weren't asking versus the other around?

1:01:29A down -round. Like, if that conversation has often led to situations like, well, everybody kind of expects us to do 8 to 10 million ARR. I'm like, oh, well, everybody expects you to do 8 to 10 million ARR. Then I got to tell you, they probably won't invest if you do it. Because what they want is for you to exceed expectations. What they're trying to invest in is the best. If they think you've already got 8 to 10 in the bag, then it's not going to work. So I'm so glad we have this conversation. Now, what do you think would really surprise yourself about this business? If you woke up a year from now, what would shock you?

1:02:03What would make you feel amazing? Do you think you can storytell that to VCs? Can we work on packaging that? Let's make about 20 % of the time, 25 % of the time, the founders are down for it. After we invest, I immediately try and have a conversation and get a pitch deck together for the next round. Just a glossary, just a table of contents, what would you want your next pitch to be? Could be story, could be product, could be data, could be anything. It's storytelling. It's always storytelling. Let's get the story down. And I think people default back to numbers when they have no other story to tell.

1:02:33So let's start from the beginning. Tell the story about what you want to be able to tell the world in 18 months about this product that you're building, this company that you're building. And then we can figure out whether we think that that's actually like viable enough for our use sandbagging. I love that. I'm going to take that. It's just a one -page it. It's like a memo. Yeah. Okay. When we look at all the cohort of enterprise companies who've raised seeds and aes over the last three to five years. And they're brought up on the triple, triple double double style, kind of pathway. Yeah. And they're at 8 to 20 million in ARR.

1:03:02What happens to them? Because new growth, growth investors are going, doesn't fit my AI company, you grow cycles, lovable, that 10 million so fast, X's bolts are so fast, did they just... They don't have a smart answer here, man. I don't know. You don't know. I don't know. I don't either. I don't know. I'm worried. Yeah. Yeah, I can tell you that some of the founders that I have worked with that that are stagnating and don't have that next chapter They're doing the things that founders do. It's not even a stagnant. It's a doubling and that's not enough enough now for VCs Who are used to AI revenues and they're like yeah, I even use the word stagnating and you write it's not there's still growing Yeah, they're still a great.

1:03:42Yeah, isn't I want to do a quick find my friends so I say short statement. Yeah, you ready? I'm ready Okay, so what if you changed your mind on in the last 12 months? I don't think you should evaluate any company by the models that are underneath it even model companies. Well, the business school just gone out of business. What a problem. I'm thrilled. I can't work in excelsbred shares. Like, you know, teams like, I have you two equals some. I'm like, oh, came up with a bracket. What about the way that your parents brought you up? Did you do differently with your kids deliberately? I don't know that I did very much differently because I think my parents did not understand me at all and and yet were incredibly open to me walking my path.

1:04:25My mother was a first -generation immigrant. Like she just wanted me to be a doctor or a lawyer or whatever, except unlike a lot of first -generation immigrant families, she never told me that. I never even felt it. She could tell that I was gonna walk a weird path. She didn't even know what entrepreneurship was. She didn't know any of that stuff. And she just wanted me to find my place. So no, I'm more trying to mimic my parents than I am the opposite. Are you hands off as a parent? No, I'm pretty hands on as a parent. But you let them do what they want to do. Insigliary. I like the role of consigliary in the world.

1:04:58We can talk like crazy with a founder or my two sons about what they're going through and then with earnest and like deep heartfelt coming from a real place, be like, it's your decision at the end. It's okay and I don't know what the truth is. So I'm not telling you what's right or wrong. You gotta walk around path, but it doesn't mean we can't like exhaustively talk about it all. I said the other day to my mother, the best thing he ever did for me was Fuckle when I left university and I was a law scholar to do a podcast. And to be fair, it's not Fuckle, it's trusting in your child and their conviction enough.

1:05:30Yeah, I mean, look, I had a moment where I came home after my sophomore year in college as a computer science major, which is you're like, oh, he probably gave me a job. It wasn't even sure then. And I was like, I want to leave. And I'm not even sure if I'm going to go to university. I might just start another company. I don't know. I just can't do this. I ended going to art school and my parents were completely supportive, could not have been more supportive. Does being rich make you a better investor? If I had any superpower joining venture, it was that I had no long -term desire to be in venture.

1:06:02I was very happy to be in venture if it worked out and I'm still very happy to go start a company. I would be very happy as a founder as well. It's just not what I chose to do and I really love this job deeply. But if it had not worked out, it was okay. And so I came in with a nothing to lose mentality, which allows you to sit on the front end of creative risk and being willing to take that extra risk, which of course that's what this business is about. Versus this kind of protectionist, like I just want to make sure I have my job. I just mean when I get to the next fund, which is where I think you make most of your mistakes.

1:06:32If you were sitting down with the HBS, Banker's style, young investors today, you've been used to the last five years. What would you advise them to prepare them for the next generation? You know, there's this thing where musicians struggle with their second album when it works, and that is completely untrue about athletes when they get to their second year of being a pro. If you're a tennis player and you go to your second year, you get better. You get smarter, you get better. Musicians, it actually gets harder. Why is that? Why do you have the sophomore album problem? And it's because one is a core creative exercise where you're not trying to a your tests.

1:07:07like you don't know what the next answer is. With athletes, you know, it's a fixed game. Venture is not a fixed game. It is more a creative exercise than it is a math exercise. And that is because the markets are changing constantly, the venture market is changing constantly, the founders are changing constantly, the products are changing constantly. And so it's just not like putting a ball in a hoop. I don't know how to talk to somebody about navigating that, but I try to get them to a world where they understand how to navigate uncertainty with confidence and without terror. Which company in the last 24 months did you not do that you reflect on most?

1:07:41I think many of the decisions that are made wrong in how people build venture firms, in how people hire people and how people invest, is about not being fundamentally attuned to that fact, and because it's so unnatural for humans to work in a kind of incredibly intrinsic way. Do you feel like you did a good job today? I came in six months after this job, And I'm like, and I look, I sold a company to Zingab beforehand. I was in this early growth hackers of Silicon Valley kind of groups. I was AB testing everything with some of the very first people who are helping Mixpanel and Amplitude build out their data dashboards.

1:08:15I was an all -in data guy. And so you can imagine, I'm three, six months in, and I'm like, am I doing a good job? And I'm trying to measure literally everything to figure out whether I'm doing a good job. And I give so much credit to Bigeon, who's a person who really recruited me into Spark, who just kind of would always reflect back to me. He was just like, did you enjoy the work you did today? Do you think you put all in? Do you think you want to come back into it tomorrow? That's it, man. Just do it well. Did you make much money from selling your company? What I'm going at here is how did making money change your mindset?

1:08:44I don't think that making money changed my mindset very much. I don't actually think I execute that differently than when I was, when I had a term sheet pulled on me and I had to sell my car in order to make payroll for my team. And those stages of early entrepreneurship, I don't know that I process the world that differently, mostly because I like playing the game for the joy of playing the game, whatever the game is. And the score takes care of itself. And I get that we all get measured on a leaderboard. Like, I don't get to keep doing this if we don't make a lot of money for our LPs and our founders aren't happy and all the rest of it.

1:09:23But I just enjoy the work. And I still enjoy the work. and I can go do other things if this doesn't work out. What question have I not asked that I should have asked? I know the big broad thing that I think about right now is I can complain a lot about the way the VC market is structured today and think that it's not structured that well, frankly, for innovation and it's not really in service to founders in the right way. That doesn't really answer how should this whole market actually work? If you could wave your wand and it was 10 years from now, we might agree that especially in worlds of high levels of innovation, you can't look at a checklist and decide whether a company is amazing or not.

1:09:58And that's not how you make exceptions, and we're in the business of investing in exceptions and exceptional companies. But that doesn't really lay out like, what should this whole thing look like? If we really believe that startups are the font of innovation and that they help the world move forward and also create great capital value for people and all the rest of it, what should it feel like and what should it look like? I don't know the full answer for that, but I think that's a more productive conversation. because once you've talked about the way you think it should all work, then maybe piece by piece together, we can all try and nudge the world there.

1:10:30I'm seeing more and more founders want no pref shares, like all common shares. How do you feel about that? Venture as an industry was a very weird thing when it was invented, right? This idea that you wouldn't take majority share in a business and you'll be a passive investor with just a board seat, a small voice instead of a loud voice. That was a unique thing when it happened. And so we've always been a world where we were in the lean back instead of lean forward control, PE mechanism of it. Whether that means we're preferred or common or the term sheet changes and our lick preps are different and all of these things have like altered over time, like open to it.

1:11:07Like I just want to make sure founders can build good companies and that they treat the people that they're bringing into their orbit as people who should be committed to the same cause and I think treating everything transactionally is kind of like the enemy of what I'm trying to work on. It's like it's probably as simple as that. It's been such a pleasure to have you. Honestly, I love doing it in person. It makes such a difference being able to see this. So thank you so much. I look forward to you sending me a picture of when you ring the bell for Grenoter at the IPO. I'll just send a picture back at me crying.

1:11:39Thank you so much for having me on. I love what you do. It's like insane what you've done over the last 10 years and it's really, really amazing. I do just want to say the thing that really strikes you when you spend a lot of time with Nabil in person is just what a good human being here is, how genuine, authentic, sincere and just kind. Nabil is one of the special ones. This was such a joy to do. I so appreciate his friendship and you can watch the episode on YouTube by searching for 20VC. That's 2 .0VC on YouTube. But before we leave you today, let me tell you about Riverside. Did you know that every virtual 20vc interview you listen to?

1:12:15Yes, including with incredible guests like Mark, Benny, or Sam Altman is actually recorded using Riverside .fm. It's the one tool a genuinely cannot live without. Riverside delivers incredible audio and video quality, and it's just ridiculously easy to use. And the best part, my guests love how simple it is. They don't need an account or software. Just a simple click, and they're in. When you're hosting someone like Mark or Sam, it kind of really matters that it's also really reliable, which it is. And Riverside doesn't just stop at recording. Their AI editor is a personal favorite. They make it so easy to create shareable clips, turning an epic interview into bite -sized moments to go viral on social media.

1:12:54If you're thinking about starting a podcast or upgrading your current setup, trust me, this is the way to go. Check it out at Riverside .fm4 -20VC and see why I can't live without it. Now, let's tackle those messy, expense reports. You know, those receipts that seem to multiply like rabbits in your wallet, the endless email chains asking, can you approve this? And don't even get me started on a month and planet when you realise you have to reconcile it all. Or Plio offers smart company cards, physical, virtual and vendor specific, so teams can buy what they need while finance stays in control.

1:13:29Automate your expense reports, process invoices seamlessly and manage reimbursements effortlessly, All in one platform. With integrations to tools like zero, quickbooks and net suite, PLEO fits right into your workflow, saving time and giving you full visibility over every entity, payment and subscription. Join over 37 ,000 companies already using PLEO to streamline their finances. Try PLEO today. It's like magic, but with fewer rabbits, find out more at plio .io -4 -20VC. And don't forget to secure trust with your customers. Trust isn't just earned, though, it's demanded. That's why over 9 ,000 companies, including at Lassian, Cora, and Factory, rely on Vanta, to automate their security compliance.

1:14:12So Vanta helps businesses achieve certifications like SOC2 and ISO27001, turning months of tedious work into this beautifully fast and straightforward process. Their platform automates compliance across over 35 frameworks. It centralizes workflows and it proactively manages risk, all while saving new time with automation and AI. So whether you're just starting or scaling your security program, Van to connect you with auditors and experts to get audit ready quickly and build trust with your customers. Get $1 ,000 off your first year by visiting van2 .com -4 -200VC. That's v -a -n -t -a .com -4 -200VC.

1:14:52As always I so appreciate all your support and stay tuned for an incredible episode coming on Wednesday with Max Levchin founder Adafirm.