20VC: Scaling to $1BN+ in Revenue with No Funding: Surge AI | The Most Insane Scaling Story in Tech |

23:11But if you really want to go big, if you really want to build a generational foundational company, I think it really should be about an idea that is almost like unique to you. You said about people maybe gaining value or self -worth in raising big amounts, going to conferences. That is how most people do gain self -worth. When you think about where you derive your own self -worth from, sorry to be personal, but given yours is clearly not that, how do you think about where you get self -worth, self -value from? I think it's kind of funny. So if I think about the things that maybe happiest in the past few years, I can think of like two things I'll tell my head.

23:46So one is sometimes our customers, whenever they launched their next big model, one of their first things that they do is they'll reach out to me and they'll be like, hey, I just want to send you a note that we couldn't have done without you. And I think that's just so amazing to hear like, again, if you think about like how often do you get to play a role in building some of the most important technology of our time and then right after they're launched, like these very, very top people who are very busy. One of our first thoughts is to thank you because of how critical your return operation.

24:15I just think that's so cool. So that is one of the things I often think about. And then I think the other thing that I often think about is, again, in many ways, surge is embodiment of me and my interests. And what I've always loved doing is analyzing data and figuring out how to use that data to make models better or to make products better. And so every now and then, when I just get the chance to write in an Alice myself of the latest frontier model. Or I get to read some of the analyses or internal employees are creating based off of the data that we're providing. I just think it's so cool that a lot of data we're providing.

24:45It's just so insightful and it helps people build models in ways that they just wouldn't know how to otherwise. Going back to that story then. So you build out the MVP, you post it, and then you said luckily, you said it very nonchalantly, Edward, which is very sweet. People came and people Lighted what did that look like how did the initial demand come to you? Sorry. I think I say it not shalonly because it felt very not shalon I think what would end up happening is so I would find all these people who Really were desperate for a lot of really high quality data I mean the way it work is they would just email me with their request or we would just jump on a live meeting and we would just get started and Might take a week or a couple weeks to Negotiate some sort of So, SOW or contract, just because a lot of this does have to live within the confines of their company.

25:33But, yeah, I think we're really lucky. And again, I had a lot of experience in this space. And so, I had a lot of experience working with ML engineers and research scientists and the ways that they wanted to get data and the way they wanted to look at it. And so, I think things just moved very, very quickly. In the early days, everyone else is acquiring supply side of town, correct? All the other people that compete in the space. And you're not acquiring that talent supply, you're building product, correct? It was both because I mean, obviously we need a talent supply in order to make our product work.

26:08So there are some companies in this space who will something think of it as a pure supply problem and they don't give any consideration to the technology, like both the technology to underlying technology, like how do you identify these people, how do you make sure that they're doing good work, how do you remove the bad quality work, like they're just literally not thinking about any of the technology aspects at all, and you're also not thinking about a product at all. Like, how do you present the data to the customers? Like, one of our principles, like one of the principles that I've always had, even prior to surge when I was just an M1 engineer or data scientist, one of the things that I've always tried to encourage is what we call this visceral understanding of the data.

26:45Like, I really just want you to go in and get your hands dirty and look at the data. Like, historically, a lot of M1 engineers, they kind of just don't take to time to look at the data. And maybe that's because the data just isn't all that industry. Like when all you're doing is drawing bounty box around cars, sure. I don't need to look at a thousand bounty boxes. But when you're doing is creating poetry, creating mathematical equations, creating new research, like you want to get your hands dirty with the data to see what it is that you're producing, where you're teaching your models. And so I think it actually really is important to this aspect of, or necessarily understanding the data that you're getting.

27:17And so we're there doing both building product and acquiring the talent supply in unison. What did we end the first year at? Did we have immediate product market fit? I mean, I think it was very, very obvious that there was just huge demand for this product and there was so much more that we could be doing. So Edwin, when there's huge demand for your product, this is even more so the time when everyone goes, now raise money. Higher CS teams, higher sales teams, why did you not raise money then? I get it at the start when hey, you didn't want to do whatever I lost it Why not raise money when it was a hair on fire problem and you had so many people cooling you I would say there was nothing that Raising what help us with I can we were very lucky to be profitable from month one and so we didn't need the money We didn't need a sales team like I didn't actually didn't want a sales team going out and selling our product Like I wanted people to buy us precisely because they understood the value of high quality data they saw all the gains that our data was producing.

28:14I didn't want them to buy it simply because they heard about it since I'm tech -crash -ronical. Because that would almost put them at odds with the kind of product that we were building. Like one of the things that I think is actually really important, especially early on, you want customers who believe in your product and not people who are simply giving you a little bit of money. Because your early customers will shape the kind of product that you're building. Because yeah, you're building for them. You're building for their needs. Like they're giving a lot of really, really great feedback. And so you almost want customers who shared a shame overall vision.

28:42And so that was actually very important for us. Like I didn't want sales teams who would email 10 ,000 people and be like, hey, any thoughts on getting good data, it was just very, very counter to kind of product that we wanted to build. How do you think about what you just said there in terms of building with your customers, being so close to them, letting you shape your product? But then also not doing the Henry Ford of building a faster horse, and then also not building a product that bluntly isn't relevant for a wider audience base, and you really just kind of tie yourself into a few small, what, few clients.

29:12So I think this is where we actually have a really, really strong vision of what a product should be. So again, like going back to what I said earlier about, how most companies within our space but maybe also at large, they don't have product principles that they try to adhere to. Again, like we had very strong product pencils from a start. We wanted to focus on quality above all else. Like if whoever thought that we couldn't give the quality that we wanted, we would just say no. As opposed to these other companies where they're almost like desperate and racing around just trying to get any traction that He can to try to prove to the recies that there are numbers are always going up They're almost like focused on getting $10 $100 a thousand dollars wherever they can and so as soon as some customer comes to them Even if that customer is counter to the kind of product they're building if they're offering money They'll just make sure I'll do it just because they'll give me another logo from a website They'll give me another case study to show another customer.

30:05They'll give me another talking point with my VCs. I think we're very lucky to not have to worry about that, because we could go for the long -term vision we had, as opposed to, again, as opposed to pivoting every few months. We just wanted to double down on the idea that we actually believed in. Is there a time when you lack quality slip in any area of the company, and with hindsight, what did you learn from that? No, I think we've never let quality slip. It's such a principle and grain into everybody at a company like one of the things that we simply tell everybody when we first join Quality is the most important thing.

30:38It's more important than anything else If you have to make a deadline slip because for whatever reason you don't think the quality is there if we have to say no to a project Because we just can't handle it right now We can generally handle a lot of things, but we just want to ingrain this principle that is it okay to say no It is okay to kind of let other things maybe slip just because we care about quality at the end of the day Most founders have a challenge where they need to hire now, but they haven't found the perfect person And so they hire a seven out of ten they let the quality boss lip because they need someone in the role How do you think about that?

31:10And what would you advise them? Yeah, I think the funny thing is like again I've been at all of these other companies oftentimes when people are saying like yeah I my hair is on fire and I really need this engineer So I know they don't need to borrow I'm gonna lower the bar to hire them like actually the engineer doing like what are they doing? They're building probably a feature that nobody cares about. They're building an internal tool to improve the productivity of everybody around a company by 2%. While at the same time, having so many meetings with them that they take up 5 % of 10 % of the time, just talking about the feature.

31:42A lot of the things that people hire for just actually aren't all that important. And so again, when you don't feel like you have the higher for the sake of hiring, Like when you have the mentality that, okay, if your company only grows by 10 % or even 0%, that's actually positive. Like, I think people right now they have this view that if someone would tell you, oh, yeah, my engineering org only grew by 2 % this year. Your initial reaction is going to be, okay, you guys must not be doing well. And so there's like the snake events ended where people feel like they need a hire, just in order to prove to other people that.

32:14Do you think now we're in an opposite world to that? The way you see the reduction in force from saying Microsoft and you see bad performance once and ever from them on a revenue per head. Do you think now we're seeing the counterbalance of that, which is the desire to be the smallest team, the fastest team to XAR and the smallest team to it? And now revenue per head is the most important metric. I also don't pay enough attention to, to like, these kinds of Silicon Valley Twitter discussions for me to have a sense of whether does mentality is becoming more pervasive. I can believe in it, I can hope for it.

32:45I don't know if it's sure right now. Do you worry that by not being so ingrained and social you miss out on certain elements that is important to be in or do you think that purity of mine you get is really so valuable. It's kind of funny because again I used to work at Twitter and I love Twitter back in heyday but I actually really am glad that I'm not surrounded by the default ways of Silicon Valley thinking. So every now and then if something is important enough like maybe there is some big new product is actually really cool or there's some really really interesting you've researched paper.

33:18It would be big enough that even though I'm not monitoring Twitter every day, it would just reach me in some other way. Yeah, one of more employees will tweet, will we'll post it in our stock channel or somebody will email it to me. So like the really important stuff will manage the percolated self to me in other ways, but I actually haven't really got that I'm not. Worrying about what people are saying about us on Twitter. I love that, especially given the irony of being in Twitter for a number of years. I do have to also first year ends. What do you end revenue add on the first year? Yeah, let's just say we've been we've been doing really, really well from the start.

33:48You said publicly about being in a billion in revenue now. Did it look like relatively even growth with their elements where it was much more accelerated than others? I'm just intrigued and say whatever you feel comfortable to in terms of that. Yeah, so we've always been very, very successful from from the early month one. Things definitely hit an expression point with ChatGbT because I think people just saw how incredibly valuable human data in our Jeff was. So definitely chativity was an inflection point for us, but even before that we have a strong roof. So post -chat GPT you really see the inflection point.

34:23Another one that I guess is probably quite an important one is scale of seed selling and the movement of customers away. How did the world change for you with the scale acquisition? So it's interesting because I think it was an open secret where a lot of top researchers already knew who we were. They really knew that we were the biggest than the best in the space, even though we've been pretty under the radar, and so most people were already working with us. There are a lot of teams who are using scale for legacy reasons, so they just didn't happen to know about us. So we've been getting a lot of new interest for them too.

34:51I think the more interesting thing has been, it's kind of been really fun seeing how we've opened their eyes to what really amazing, really high -quality data can actually look like. Like a lot of them have tried getting human data from other teams, and it tells it's been this slog. They'll spend months trying to improve the data quality for a really basic stuff, and it will look like it's better for a month, but then it will just quickly regress. We have this concept where we just want to get started immediately, we want to show them really, really high quality data immediately. Like when we know big concepts for us as a company is, we always want to be producing data that you simply couldn't get anywhere else.

35:24Like there's so much richness and complexity and types of things that we do that we just want to open up new avenues of research and open new avenues of new types of products. I think a lot of these new companies or these new teams who've been coming to us It did help, it's just been a breath of my share for them. I spoke to Garrett Hanshake right after the acquisition, he said, I'm just saying all night, there's just a tidal wave of scale customers moving to us. Did you have the same known terms of that tidal shift in customer demand shifting to you as well as the realization that you mentioned there?

35:55Yep, I mean, so I would say I'm pretty sure that a lot of these other companies, like Edinodate, people want high quality data and they don't want to be working with body shops. And so I think we've seen like a massive wave interest because like yeah, like the space is really large And there are a lot of teams who are still using scale for like a series ends But it's like at the end of the day We were ready the biggest invest in a space and so even When there were teams at some of these larger companies who weren't working with us already They kind of like knew who to turn to do you think everything has a price Edwin?

36:28I mean, I think for some people and like they have a price, but I think we don't you said you wouldn't sell to suck for 30 billion dollars But you sell for 50 billion dollars. No, I mean, I definitely want to sell for 30 billion or even 100 billion. If you think about us as a company, I already have everything I want. We're profitable. I have complete control of our destiny. And so I'm really lucky to have all the resources I want to already do anything that I want. And there aren't many companies who can say that. What are you doing this for? I've introduced you to thousand founders. In the nicest way, I've almost never met a founder like you.

37:01In a nice way, it's really special. But with the pure mindset like you have, what are you doing it for then to build a business that you can pass on to the next generations? To build a legacy. What is it for you? I mean, I think it really is to help achieve AGI. Like, do you think about every, every, like what do kids dream of? Like, yeah, when you're a kid, you literally dream of building AI deck in, do all these amazing things. And now we have the chance to do it. Like I really do think we are such a critical aspect of what all these companies are building. Like again, a lot of our customers at these fronters labs, they would just often tell me they wouldn't be able to build what they're building without us and just amazing at what we do.

37:37And so being able to be this critical part of what is the greatest technology of both our time now, but also maybe one of the most important things we can ever build, that's amazing. And so why would you get acquired and stop doing that? Because getting acquired would be really limiting. It would be this admission of failure and jumping ship because you can't make it on your own anymore. When we're to upset, we're incredibly successful. And there's literally nothing else that I'd want to do. It is 2040 and we still do not have a GI. What is the primary reason why that would be the case? So I think there are two reasons.

38:12One is that there will always need to be more breakthroughs. Whether it's breakthroughs and how you leverage all this data. Or breakthroughs in different times of algorithms that they're building. And another one is just how you gather that data. It's like in order to cure cancer. How will you gather the data that's needed to make those breakthroughs? Maybe you're going to have to run real world experiments, real world studies. And those studies will simply take time. Will there be a way to speed up those experiments through various kinds of simulations or just other forms of gathering data? I don't know, but they're just a minute of question.

38:43How do you get the data even faster? Which I think will be very, very important. Speaking of kind of evolutions with AGI there, do you just want to also like the changing nature of data? How would the data needed evolve as AI gets smarter and smarter with each evolution? So a lot of people talk about this shift to PhD level data. And yeah, I think it's important. Like we basically have the biggest group of the smartest people in the world working on a platform. We actually have Harvard professors and Stanford PhD students and Princeton computer science theorists working on all these really interesting problems with us.

39:12Like it's kind of crazy. If you think of all the PhDs even at Google or Meadow or Microsoft, we have way more than all of them combined doing work for us in a single day. And it's also true that they're not just writing random JavaScript code to improve ads. They're actually pushing different tiers of science when they're collaborating with these models. But I think what people underestimate is that having a PhD isn't enough. Like a lot of PhDs, they just aren't good at this type of work. There are a lot of body shops and recruiting shops in our space that basically just look whether you wrote down that you have a PhD on your resume and it'll just instantly give you work if so.

39:42But a lot of PhDs just aren't very good. Like I think 80 % of the community science PhDs I know they write shitty code because they're only good at math and algorithms. And you think about people like Ernest Hemingway. You don't have a PhD. I don't think I even want to college. And so I think there are two things are important. There is this underestimate aspect of our space where you actually need a lot of technology in order to make sure that you're delivering for the high quality data. Like I think it's like a lot how like Vimeo has a lot of so -called high quality videos. But yeah, they don't have any algorithms.

40:09And so YouTube's videos are way higher quality and more engaging in the end. And then the second is that just that a PhD isn't enough. Just because you have a PhD doesn't mean that you can make some breakthrough in physics. We also need is street smarts. Like you need that creativity and the mental 42 to think of really interesting problems and find these problems in probe albums and see whether they can solve them today and then teach them in really interesting ways. Because otherwise, if all you're kind of doing is throwing PhDs out of the problem, all you're doing is teaching models how to hack silly benchmarks and get good at basically the equivalent of SAT problems.

40:41If that's the landscape mode today, which is PhDs on enough, puts a lot of PhDs on great quality, How does that change over time? Will you have a dramatically larger supply side? How will the tooling of the supply side change? How will our ability to turn around work change? Again, I think this boils down to technology that we build. Over time, it's simply true that people are going to be trying to solve more and more problems. We have hundreds of thousands, millions of people working on our platform. When you do that, and you have a thousand projects, like 10 ,000 projects that are a little of running in any given week, how do you make sure that you have a lot of money?

41:16you are building technology to identify who are the top 1 % top 2 % of people who can really push the boundaries of physical elements with these models. Or how do you identify the top 2 % or 3 % of people who are writing the most amazing poetry? How do you find those people? And then also how do you remove the worst of the worst? The people who will and inevitably try to cheat you and spam you and they will basically regress the models if you allow their data through. Like it actually is a really, really profound problem and you just need a lot of technology to build this. And at the same time, these are researchers who want to move really fast.

41:52Like researchers to all these frontier labs. Again, all the algorithms are changing every day and so they want to try out new projects every single week. And so if you're not moving fast enough, like if you're unable to create a new template or you're unable to find the expertise that you needed like literally within the next day or the next week, it's just gonna be too slow for these researchers. And again, if you don't have to name technology to manage these 10 ,000 projects and automatically create them and automatically identify the really high quality data, it's just going to be too slow for them.

42:20Speaking of slowness of data and quality of data, I would love to push you on this. When you think about like, bottlenecks to progress today, if I were to rank them one through three, one being the most pressing bottleneck and three being the least pressing, you would access to compute, you would algorithm and you would data quality. If you would rank them one through three, how would you rank them? I would definitely rank data quality first, followed by compute followed by the irons. If compute continues to prove to be the unlock, we're throwing more compute at it, unlocks more and more performance.

42:54Does that denigrate data quality in the prioritization stack? I mean, I actually just fundamentally don't believe that you can throw more compute at a problem. Because if you're not getting the data that the compute is essentially trained on, or if you don't have the right objectives, any value -wishing metrics that again, your compute is optimizing towards, you're just going to fall into this trap of seeing progress that actually isn't there. I can give you some examples. So let me talk about why I think data quality is such a problem. So I think data quality issues have already been a huge step back for the other frontier labs.

43:24Like one of the things that we often hear from teams over and over is that before they use us, they tried getting data in other ways. And so they train their models, they devaluate their models, and their metrics kept going up. But after six months or even a year, they realized that their trading data was shit. Their evaluation data was shit. And so all the progress that they thought they were seeing was actually completely misleading, and they either made no progress or their models after six months were even worse than when they started. Like for example, we see this a lot with LM Arena. So LM Arena is this popular leaderboard of LM Models, and it's basically the equivalent clickbait.

43:54What happens is that you have people going on to what's called a chatbot arena. They'll enter a prompt, they'll see two model responses and then they'll vote on which one's better, but they're not taking a time to really read or evaluate the model responses at all. Like one at models could have made, completely made everything up, And these participants still vote on it because it has emojis and nice formatting. We've literally seen this in a data ourselves. Like, one response will just be a complete hallucination. But because it has an emoji and because there's a couple of words voted, people will just like, okay, yeah, like that looks good.

44:23That looks much better than this other thing that I didn't take the time to fact check at all. And so one of the things that we've learned is that the easiest way to improve indisterina is simply to make your model responses a lot longer. Like, one of the funny things is that you especially take the top model honestly to the board, the number one model and you ask it when did the Pope die? It'll give you a really long response that seems oppressive, but it gets the answer completely wrong. It tells you that Pope Francis is still alive. It'll even tell you that there are search results that indicate that Pope Francis died in April, but actually these were just rumors and misinformation and he's still alive.

44:52It's wild that this model will say this. It's okay, like what happens is that there are a lot of companies who are trying to prove they're a bird or a drink. And so to see progress for six months because all they're doing is unwittingly making their model responses longer. They're adding more and more emojis, they're adding more and more formatting. And so they see their models cladding on this leaderboard, and so they think they're making progress. When all they're doing is training their models to produce better clickbait. And they may finally realize six months or a year later, but it means they've actually spent the past six months making zero progress.

45:19This is what happens when you kind of like throw compute at the problem without understanding the underlying training data that you're like again like throwing the compute towards. It actually just sets your models back. When you look at like Grock, obviously announcing their recent developments and how they performed in the latest benchmarks and came out as number one Are those benchmarks bullshit then how much weight should be placed on the importance of those benchmarks and how reflective are they truly of Model quality if you watched a rock for a launch the rock for a live stream I think you would even have heard Elon himself saying like yeah, these models are really good at I forget the word He used but like they're really good at homework problems.

45:54They're really good these academic very We narrowly scope problems. It's basically the equivalent of making them really good on SAT problems, but I'm making them good at problems that people are actually facing. We used to be surprised by how far Elon has been able to get with Groh as far as he has done or not. I am not. It's kind of funny, so before we worked with the team, I didn't really have a conception of what an Elon company was like, but yeah, I mean, we worked really closely with the XAI team and it's actually just incredibly refreshing to see how they operate. They are all very very mission oriented.

46:28They're all incredibly smart and they work incredibly hard like it will be 11 p .m. at night and I'll DM them and someone will want to jump on a meeting and yeah I jump on a meeting with them and I see them during the office and there's a ton of people behind them So like they're all like crazy hacking together on all of these problems And so I actually think it's incredible and it's this kind of embodiment of what a startup can do when you do, really believe in something and are willing to do whatever it takes to achieve it as opposed to living within the confines of this giant bureaucracy. So I think it's actually really impressive.

47:00Is there anything that Elon does specifically to inspire his team to have that form of culture when they're not a small company? I think it's almost that you know what you're getting into when you work at GROC or when you work at XAI or any of these other companies. You know when you interview that these people are incredibly mission oriented. You know when you interview that everybody works super hard. You're that if you want to work there, you're going to have to be the kind of person who has the same values. Otherwise you just shouldn't, you just shouldn't, you don't want because you'll be miserable.

47:29It's this fact that it has such a strong culture and such a strong belief in what you're doing. It just attracts people of some more talent. I do want to touch on kind of the working as you mentioned that, but everyone poses synthetic data. That's a big threat. And what happens to your business when we have synthetic data that is obviously created automatically and labeled automatically? How do you think about the role of human label data in a world of predominant, asymptotic data? What's your thoughts there? So I think synthetic data is actually really useful in some places, but I think people over us may, what they can do.

48:03I'll give a couple of examples. So right now there are a bunch of models that have been trained really heavily on synthetic data. But like I mentioned earlier, it means that they're only good at a very academic homework more style, bench art style problems. They're actually terrible at real world use cases. So yeah, synthetic data, it's made models good at synthetic problems, not real ones. And we actually hear from a lot of companies who tell us they spent the past year training their models on synthetic data, but if only now, just realize all the problems that's caused. And so they've spent actually months throwing a lot of it out.

48:30Like a lot of them tell us that even a thousand or a couple of thousand pieces of really high quality human data that we generated for them, it's actually been worth more than 10 million pieces of synthetic data. And so a lot of the work that we do, it's simply cleaning up all this synthetic data. And if you think about what wide is happens, it essentially is because the model's collapse on its very, very narrow scope of similarity that the synthetic data creates. And so just doesn't give the models the kind of diversity and generalizability that they need. And then one other point is that there's also this interesting phenomenon where models simply make a lot of mistakes.

49:00And I've certain misunderstandings that humans never will. Like I was actually just playing with one of the frontier models recently. And it kept on just randomly outputting Russian characters and Hindi characters in the middle of its responses. This is a mistake that would be obvious to any human, to any second grader, but the model just didn't know. And it's like shocking that the model in 2025 are frontier model in 2020, how I would do this. And so it's almost like you always need this external value system as a kind of safeguard to make sure that the models are working properly, just because the models themselves have such a different set of way of thinking.

49:29I'm an investor in poolside, which if you don't know, obviously it's kind of in the same spaces, say a cursor or a windsurf, but they seemingly are much more behind because they've built their own models, and they believe very much in the power of verticalization of models and specific models, or specialized models, so to speak. How do you think about the future in terms of monolithic generalized, very large scale models versus the requirement to have very narrow, very specialized models for things like code creation and development? I think there's an opportunity for both, and the reason I think that is, it's because on one hand, you have these giant all -powerful models.

50:05And sure, they can be really, really good and really, really powerful. And in like a raw capability sense, I think they'll be able to encompass all these different use cases. But in the same way that a company, so take a company like Google or Facebook, there are simply some products that they can't build because building those products would be counter to like culture or the business goals of like the overall parent company. And so in the same way, sometimes you need to be able move faster and to take big bets on certain kinds of products and all powerful models just can't kind of let that happen because if you let it happen within just like one small domain, we'll kind of almost like pervade the entire model.

50:42So sometimes you do need to like the smaller models to break through if they have like a really unique view on how you're operating. You are very composed as a leader, as a CEO. Where are you not meeting the bar? Where are you not great and you are aware of it? So I think one area where I'm not great, which is kind of funny, but one area where I'm not great is I'm really bad at understanding financials. So sometimes people around a company don't try to tell me like, hey, have you been paying attention to revenue numbers? Have you been paying attention to or costs? Have you been paying attention to your margins?

51:14Do you even know what they are? And I don't. They were just like these financial measures that I could not tell you what EBITDA is. I mean, I know what the acronym stands for, but the difference between that and revenue and profit and Net margin and operate like I actually just don't know any of these terms and it's just like this blind spot I know matter how much I tried to try to understand these things I just can never remember what single mattress defines the house as a business to you What mat trick if I showed it to you every morning you'd be like okay? I know the state of my business like if I could paint my perfect nor star And this is something that I think we we want to work towards like it's something that we were actually we you actually want to build for an industry.

51:52It's like, are models progressing in fundamental ways, already actually getting more intelligent? Are there capabilities improving? Again, as opposed to simply climbing up a meaningless clickbait leader board. So are these models progressing? And then how much of that is due to us, whether it's due to our training data, or whether it's due to the evaluations being fried, or whether it's due to the insights that we provide all these researchers for ways that they can improve their models. Like if they're a way to measure that, I would love it. I think the closest proxy we have for it today is just like the variety of projects that we're creating.

52:29One thing I really, really believe in is we want to make it easy for all of these researchers to come up with new ideas and to not be blocked by data. So the more complex, the more diverse, the more creative projects that we can provide, that is that comes across proxy for that overall more. Final one and then we'll do a quick fire, but you mentioned Elon and X and the hard work in that culture being so ingrained. I recently said that Bluntney Silicon Valley and China have increased the intensity required to win in terms of work ethic. You must work seven days a week if you want to build a $10 billion plus company and the ability to put your phone on the side and not check an email does not exist anymore if you want to build a $10 billion plus company.

53:14You've built a $10 billion plus company. Do you agree with me? So I would say, I think you have to be willing to work hard. Like you have to be willing to jump on a call at 2 a .m. And yeah, I'm customarily like, I think one of the things that I love is that sometimes customers don't call me. Although they call me at 2 a .m. 3 a .m. and they're like, hey, our models are freaking out. I need a bunch of data to fix it by 6 a .m. Can you do it? And maybe going back to the question of things that make me happy. Like nothing makes me happier than knowing that, yeah. We can deliver us, yeah, we can deliver 10 ,000 data points to you.

53:45And next few hours, even if you call us at 3AM, to fix some critical bug, critical fire that you're facing. That is actually something that makes me incredibly happy. And so I think you have to be willing to wake work hard. I think a lot of people do confuse working hard with creating value, like again, it's maybe a trope to say, but you have to work smart and not just hard. Like if I think about a lot of what I'm doing, like oftentimes the best ideas come to me when I'm just walking around, not necessarily when I'm at my computer. I mean, I think we all work really hard, but I wouldn't confuse the number of hours we spend with actual progress.

54:18What traits of yourself do you love most or is your favorite trait, Edwin? Good question. So at least the thing that I really enjoy is, like, I've always really enjoyed writing down insights and reform. And I think I'm pretty good at it. And so disability to to deliver some novel insight about a model, or deliver some novel insight about an algorithm, or deliver some novel insight out of dataset, and communicating that to two or customers. I think I'm pretty good at it, and it's something I really, really enjoy. Dude, I want to do a quick file. So I say a short statement, you give me your immediate tools.

54:52Does that sound okay? Yeah, that sounds great. So what one widely held belief about AI, do you think is completely wrong? So I think a lot of people think AI safety is overblown, but I think they ignore the paperclip maximizer problem, where you have AI models that are accidentally trained towards their wrong objectives, even though this is a big problem that all of them hold its face today with all the issues around alumorena and benchmark hacking. So I actually think it's a really important problem that people should be thinking more about. So you think AI is much more dangerous than we let on?

55:20Both dangerous, but that you can be accidentally maximized towards the wrong objectives that like today, you guys sure, like if you maximize towards these alumorena objectives or benchmark hacking, the worst that will happen is that your models were regress in progress a little bit. But the more fun of what we're problem is that people don't realize this. And so in the future, when the models are more powerful and yet you're basically accidentally maximizing AM models towards the wrong objectives and you just have no idea what will happen. It's almost like a similar phenomenon to what's happening today, but because the AMOs are so much more powerful.

55:54Like yeah, they're really building the code for an insurance company or they're really building the code for some trolling dollar company, just the consequences can be much worse. You mentioned about gaining true passion love for building towards AGI. I hate myself for asking this question. It's a shit question. I hate it. I'm so embarrassed. But if you had to put a number 2028 or 2038, which bracket would it be in and why? So I think it would be 2028 if you were talking about automated in a job at the average engineer and then 2038 if you're talking about current cancer. Sorry, 2028 automated job at the average engineer.

56:26I had Vlad on the show from Robinhood and he said 50 % of co -creative by Robinhood is now by AI. Benny often said the same on the show 50%. Are we not at that stage already? How much code from search is created by the AI? I don't think we're at that stage yet. At least if you're, again, if you're working on deeper problems that aren't just random features, like again, if you're concentrating your company on the 10 % of problems that are most important, I don't think models today can write 50 % of the code and come up with 50 % of the ideas that are actually going to be meaningful to your company.

56:59sure, if 90 % of your company is writing little features that nobody cares about or improving the efficiency of your code base by 1%, then yeah. But I don't think we're at a point if you're really working on meaningful problems. What is question should every AI company be asking themselves, but isn't? So if you're a frontier lab, the question is, are you actually improving your models in the role intelligence, or are you just hacking benchmarks? If you're product company, yeah, the question is why do frontier labs won't be able to instantly replace you? Do you think they will. I don't ever worry about that in terms of application layer being able to work by model layer just because I think that is infinite product breath that they could go off to they can't go after everything.

57:37I think they can't go after everything but there are so many things where yeah you really just want to chat with the model and it's very simplistic universal interface that again I think about Google search like I actually do feel I may have felt that maybe 50 % of the things I used to Google Google search they are replaced by Chatchy or they're even better with chat GPD? It is a very pleasing aspect of a universal, all -intelligent interface that I think people will just grab a day to. What would you do if you were send out today? Would you kill your golden goose with the ads engine? So the difficult problem, I think for Google, is they have to be willing to take a short -term hit to all of your advertisement revenue in order to build something better.

58:16That's just really hard. The ultimate one. What did you believe about the future of AI that you now no longer believe will have changed your mind on? So I think the thing that I've changed my mind on is how I see a world where there actually will be multiple multiple Frontier AI companies frontier AGI's just because every one of them will be able to go in a different direction Like you see it already you see it today playing out already with the differences and the strengths the weaknesses of Open AI and for a pick and I think that that trend will continue What does that mean? I'm sorry if you just play that out What does that landscape look like then because opening on Anthropica are so unique in their properties and characteristics?

58:54It means there'll be 10 more of them. What does that look like? I don't know if there'll be 10 more of them, but I can certainly see even like three more of them. And I just think each one will have different trade -offs that they're willing to make, different focuses that they'll have, and like even today. Like Cloud is really, really good at coding. Cloud is really, really good at enterprise and like instruction following. Whereas ChatchyBT is, Yeah, it's like more optimized for consumer use cases. Like I think it actually has a really, really great and fun personality right now. And then GROC, like yeah, GROC is willing to maybe answer certain questions that maybe you should, maybe you should end, but it's willing to be a little bit transgressive in ways that actually think are very, very interesting.

59:33And so I actually think that just like this willingness to have different personalities and different boundaries and different focuses on your models that leads to models to be good at different use cases, just in the same way that like yeah, there's like, I think analogy is there isn't a single poet. There isn't a single mathematician that is like the greatest mathematician all time. They all have different focuses. They all have different ways of approaching these problems. And I think that richness, like what we often call richness of human intelligence, that will apply to the models as well.

1:00:01Have the biggest model providers been founded today? It's a good question. I don't think so yet. I can actually see big new, even more powerful or model developers appearing in the next few years. How does that look? Because when you think about funding them, the capital intensity or capital requirements are so large, I don't know anyone who's willing to all the big players and the financing world bluntly have already got their horses in terms of this race. How does that even work? I think it's because it depends on what you view the long -term vision for AGI to be. Like despite all the immense progress that we've made, if you believe that there's only, we're only like, I don't know, 1 % 5 % of the way towards AGI.

1:00:41Because yeah, we literally want AGI systems that can in the future cure cancer and send and record our ship's to Mars. And like design entirely new philosophical systems. Like these are big, massive problems, as opposed to something automating away to all of the average L3 or L4 software engineer. And so if you believe that we're only 2 % or 5 % of the way there, there's just so much more headroom. It's like almost like asking, do you believe 10 years ago that Google is going to be the final search engine in the world. Like, sure, if you're only looking forward to the next five years, if you just think of the immensity of what AGI could do, there's so much more ahead of us than behind us.

1:01:19That there could be these serendipitous, very, very creative breakthroughs and just nobody's expecting in part because maybe it's going to be created by some of the AI's themselves or AI's in concert with humans. There's just so much opportunity ahead of us that it'll be almost a miss to think that we've already solved it. Do you believe AI will be able to create 10 % increases in GDP gain or in productivity increases in the next 10 years? That's often kind of touted as the number which would create $10 trillion of value. Absolutely. Love this. This is a good round. At Final One, Edwin, you can give yourself one piece of advice going back to day one starting the company.

1:01:56Going back to starting the MVP. What do you know now that you could tell yourself then? So I think it would be to focus always on the 10x improvements that you can make as opposed to worrying about 10 % of realities. Edwin, listen, I so appreciate the time. As I said at the beginning, I've been such a fan of the incredible journey. You've been fantastic. There's been very atypical in most ways bluntly having this discussion which has been so great for me. So thank you so much for joining me. Thank you. It's been great, Johnny. I have to say that show is a real symbol of why I love what I do so much.

1:02:29Over a billion in revenue, no funding, he doesn't do interviews. The fact that he sat down with me and opened up as he did simply meant the world to me. You can find the full show on YouTube by searching for 20VC, that's 2 -0VC on YouTube. But before we leave you today, I love seeing the team come together to make this show happen. What I don't love is trying to keep track of all the information, the data and the project that we're working on across dozens of platforms, products and tools. That's why we use Coda, the all -in -one collaborative workspace that's helped 50 ,000 teams all over the world get on the same page, offering the flexibility of docs with the structure of spreadsheets, Coda facilitates deeper teamwork and quicker creativity, and their turnkey AI solution, the intelligence of Coda Brain, is a game changer.

1:03:16Powered by Grammily, Coda is entering a new phase of innovation and expansion, aiming to redefine productivity for the AI era. Whether you're a startup looking to organize the chaos, while staying nimble, or an enterprise organization looking for better alignment, Coda matches your working style. It's seamless work, it's based on how to connect to hundreds of your favorite tools, including Salesforce, Gera, Asana and Figma, helping your teams transform their rituals and do more faster. Head over to coder .io slash 20VC right now and get six months off, the team plan for startups for free. That's coderco .io slash 2 .0 VC and get six months off the team plan for free.

1:04:01Coda .io slash 2 .0 VC. And while Coda keeps our team aligned, a QT scheduling ensures our time stays on track. This show is brought to you by a QT scheduling, the flasable scheduling software that helps you focus on what matters most. Growing your business with a QT, You can manage your calendar, you can accept secure payments, offer clients a seamless booking experience that reflects your brand. I've been using my complimentary subscription and it's been a game changer for staying organised and saving time. I especially love online booking. Clients can book, reschedule or cancel anytime and the booking page looks fully branded with my logo and colours.

1:04:39The calendar management tools let me set buffer times and sync with other calendars so I never feel overbooked. And with secure payments, I can collect deposits or full payments upfront through Stripe or Paypal, making the process smooth and professional. Head over to acuityscheduling .com, forward slash 20VC for a free trial, and when you're ready to launch, use the offer code 20VC20 to save 20 % off your first acuity scheduling subscription. And speaking of incredible companies, don't forget what really keeps those customers coming back. Trust is the ultimate currency in business, and today customers expect it faster than ever, and that's why over 10 ,000 global companies trust Vanta.

1:05:20Vanta automates up to 90 % of the work for in -demand compliance standards, like SOAP2, ISO27001 and more, using smart AI to centralise workflows, manage risk and get you audit ready in weeks, not months. So you can stop chasing paperwork and start closing deals and a new IDC report found that Vanta customers achieved $535 ,000 per year in benefits. That's insane and the platform pays for itself in three months. Whether you're growing fast or just getting started, Vanta connects you with trusted auditors and experts to support you build trust with customers. Get $1 ,000 off your first year at Vanta .com 4 -2 -0 -0 -VC, that's vanter .com.

1:06:074 -2 -0 -VC. As always, I so appreciate your support and stay tuned for a fantastic panel coming on Thursday with Roryo Driscoll and Jason Lemkin shooting the shit about the biggest news in tech.