[0:00:01] Well, I'm very excited to announce our next guest and somebody that you're all anticipating seeing tonight. And Sam Altman has just done an amazing job kind of taking over Y Combinator as its president with the retirement of PG. Previously, he founded Looped. Previous before that, he attended CS at Stanford. He's on the Forbes 30 Under 30, and he leads Y Combinator. Let's give a big Star Brain welcome for Sam Altman. Let's hear it.
[0:00:41] Well, welcome. Thanks for having me. It's like this every day when you walk in, right, or no? Is it? I wish, no. Yeah, there's no one there. It's just the partners. There's no companies there. There are no companies in there. Okay. Well, I think I'd love to start with this kind of theme that there's no single way to build a startup. And along this question of, you know, what are some of the good and the bad things that you see people do in the beginning? Yeah, I think, like, the foundation of these startups is incredibly important.
[0:01:08] And I think that if you get it right at the beginning, it's great. And sort of on the other hand, if you get it wrong at the beginning, it's usually fatal. And so the first thing is before you even start, why you're starting. And I think that people that start a startup for the sake of a startup, before they figure out what they're passionate about, what they're going to do, that's much harder to make succeed. As soon as you declare something a company and not a project, there's this pressure to figure something out quickly.
[0:01:38] And it can't sound too crazy, because it's like, now it's my startup, and I have to tell people about it as my startup. And when it's still just a project, it can sound pretty crazy. And these things are just so delicate, these ideas, before they develop, that I think one mistake people make is they have this startup, and now they're like, I need a startup, and I'm in my co-working space, like hanging out and doing whatever, and now I need my idea, and I'm going to do that, and it's going to be great, and I'm going to rage code, and I'm going to get on TechCrunch.
[0:02:06] And that is not the path to success. So I think the most important thing to do at the beginning is actually take the time to figure out what you're going to do and let it be a project before it's a company. Have the idea first and then the company. Have the company be the way to support this thing that you think is really awesome to make it happen. Once you get started and once you're building and once it eventually sort of develops into this company, then I think,
[0:02:36] The most important thing is to have a focus on growth. We talked about this a lot at YC, how startups and growth are incredibly interrelated. And the best startups have growth goals from very early days. The Airbnb guys used to draw a graph they wanted to hit, a forward-looking projected graph, and then they would put it up on the mirror in their bathroom and on their fridge door and everywhere. And every week they were just trying to grow 10%. And our best startups do this.
[0:03:05] And it's so important, and it is this function that shapes the company. And it's one metric to optimize for, and it's one way to answer every question. And the startups that do this usually succeed, and the startups that don't do this argue about everything else. Well, I want to talk about growth. Let me go back to your first point of doing something you're passionate about or just doing a project versus starting a company.
[0:03:27] You know, wanting to be something and trying to force it. What is that balance? Because, like, if I say to myself, hey, I want to be an entrepreneur or I want to work for myself or I want to be my own boss, but at the same time, like, I can't force it. It just kind of has to happen. I have to kind of find a problem that needs to be solved. Well, the way you can force it, sort of, is you can just decide to spend a lot of time thinking about ideas and building stuff as projects. Like, until you sort of say, this is my startup and we're going. That is when I think...
[0:03:57] Sometimes stuff goes off the rails. And where is that? When do you say this is it? Where is that point? I think when you get conviction behind what you're going to do. But I think when you're like, you know what, this may not work, but I really believe in this. I want to try to make it happen. Is there some data piece behind that as well? No, because again, at this stage, you can certainly make a company and raise money and whatever before you have built it. But raising money, for example.
[0:04:24] If you raise money and you still don't know exactly what you're going to build, even generally what you're going to build, that's usually really problematic because now you have these investors and you feel like you owe them progress. But this phase of thinking actually takes a while. And once you commit to something, once you decide to build something, you want to go as quickly as you possibly can. The hours matter. But it takes time before you get to that phase.
[0:04:52] And you can commit and say, I'm really going to spend a lot of time trying to figure out what is broken in the world and what I'm passionate about making better. Do you subscribe to the lean startup customer development methodologies? Are those necessary at the earliest stage or not? I think there are good parts of that for sure. For example, I think understand your users is one of the most important pieces of startup advice in the world. A startup that does not do this will fail. A startup that does this often succeeds. So that I really agree with.
[0:05:22] You know, this idea that I'm going to deeply understand my users and iterate very quickly, I really agree with. There are other parts I don't. Like what? You know, I hate to like... Oh, Eric Reese isn't here. Don't worry, we won't tell him. No, no, I just say, like, I think that the central idea that I'm going to... Well, okay, I'll say one thing that I think can get missing from the Lean Startup methodology. I think, and this is sort of like not his particular version of it because at this point it's been adapted so many times.
[0:05:53] But sort of very ambitious startups often take a long time to work, or sometimes they take a very long time to look ambitious. You know, when Reddit started, it was this sort of like joke of an online forum, and now it's this sort of like weird collective consciousness. When Facebook started, it was college people sharing photos, or not even photos in the beginning, it was like profiles. That was it, I remember. And now it's this like incredibly important sort of like universal connectivity.
[0:06:26] Other things, like if you're trying to build a nuclear reactor, just take a long time to get started. So sometimes these big companies take a very long time to get going. And I think it's important to allow something like that to happen. I like this phrase you said, the hours matter. That's a really great way to describe the earliest days of taking on a project or building a company. How do you prioritize? How do you look at that? What does your day even look like? One thing I've really gotten good at,
[0:06:56] In the last six months, just as YC has continued to grow in prominence, is saying no to things that aren't a good use of my time. And that's been this incredibly freeing thing. And I've really had to figure out what the highest impact things I can do are. And that is figure out how to grow YC, how to figure out how to get the best people in the world as YC partners, the best companies in the world to apply in YC, making sure we pick them well, making sure we advise them well, and making sure that we develop the YC network.
[0:07:25] This YC made a company to be this organization that can support these important companies to happen. So I try and spend my time on those things, not a lot else. And I almost never speak at things, but these are our people, so we're here. I'm here. These are the people we want to talk to. But generally, I've gotten just really good at not doing the things that don't matter. The best founders are just, they execute so quickly.
[0:07:55] One of the ways that we can identify people that are really great is how much progress they make between when we meet them. So the bad or mediocre founders, you'll meet them a week later and now they're like, oh yeah, I'm doing those things from a week ago, still kind of early. And the best founders, you see the next day and they're like, I did all of those things and these others. And the founders that execute slowly have always such great excuses with such detail about why they couldn't move faster.
[0:08:25] Like what? What's a good excuse you've heard? Oh, you know, like, well, the product wasn't quite good enough, or we learned this other thing, we wanted to take more time, or our lawyers found a problem, or, like, you know, there was this PR issue, or this competitor did something, or, you know, this VP at this other company got sick. Who knows? But there's some reason that things didn't happen. Ah, the old VP sick thing, man. Kills me every time. Yeah, right. Deal killers. Let's talk about, let's see, you've got 114 companies in this batch. Yeah.
[0:08:55] You've talked about this a lot, making greater efforts with women and underserved minorities. I'd love to hear where you guys are at on that. Where is that at with this batch? What have you done in terms of your distribution to change some of those things? Improve them, I should say. Look, so the way we look at this is we want to fund all of the best startups in the world, as close as we can get to that. And to do that, we need to reach a very wide applicant pool.
[0:09:25] And so we've made huge efforts to get more companies applying internationally. And as you mentioned, we have done a lot to reach out to women and racial minorities in the US. Good news, I think that's worked pretty well. The applicant pool has tracked upwards and upwards. We made a lot of progress, especially with black founders in this last batch applying and then funding them. We've looked at this in detail. We fund people from other countries or women or minorities.
[0:09:54] At about the rate they apply. So if 3% of the applicants are black, we fund just about 3% of black founders, a little more, a little less, depending on the batch. But I think just doing the outreach, going around and visiting HBCUs or having a conference for women that are thinking about starting companies, that goes an incredibly long way. That's moved our numbers more than anything else. Great. In terms of your distribution, are you more bullish on
[0:10:19] The growth of your YC network and alumni or on, you know, I think historically Hacker News was this kind of incredible and PG's essays were this incredible distribution channel for getting great people to kind of know who you were. But are you more bullish on the alumni network in the future? Are you more bullish as Hacker News, you know, as important as ever? It's one of these things where the way that we recruit the best people in the world is by reaching them multiple channels. So the alumni network is really important. Hacker News is really important. Putting great content online is really important.
[0:10:48] I taught a class with a bunch of guest lecturers last fall that got millions and millions of views. For that applicant cycle, the number of applicants we had went up like 40 or 50%, 60%. So I think there's a lot of things we can do. That's a great point, too, because this is something that's so interesting about YC is your numbers, you have 40 to 60% increase batch over batch in terms of applications.
[0:11:16] But these aren't companies that apply to other incubators. I don't have any hard data on that, but you just know, you get the sense of people saying, hey, I won't apply to any of the incubators, but if I got into YC, I'd definitely do that. How have you built that brand? How have you built that buzz with entrepreneurs to say, hey, in this whole space, I look at them and I won't kind of...
[0:11:43] You know, I'll focus on that, but not really anything else. Well, we've been very fortunate to fund some incredible entrepreneurs, and I think that becomes a self-fulfilling cycle. You know, I think that there are a lot of things that we do differently from other... I don't even like to call us an incubator, because that sort of brings to mind all these terrible things that these other people do. But I think that... What are you called? What are we? Yeah. Like a... Well...
[0:12:12] What Y Combinator is, the mathematical definition is a function that starts other functions. And so in this way, we're this company that starts other companies. Marketing-wise, it's kind of. And that was where the name came from. But I think what we really are is a source of funding and advice and a network for startups. But they don't work in our space. We don't have them all live in some satirical dorm. They kind of.
[0:12:39] They're their own companies, and we're there as a passenger in the back seat, and we help them when we can. How many countries are represented in the latest batch? High 30s. High 30s, and where has it been in the past? It's been steadily trending up, but it was one in the first batch. Okay. That's up. Yeah. Could you ever... Sorry, I didn't give a good answer. Go ahead. I think, like, the thing, to the degree that people think of YC differently than any other accelerator, I think it is the strength...
[0:13:09] Of our alumni network and sort of the track record of the companies. And I think that I hope part of it is also the advice of the partners. We don't have a bunch of mentors, we only have full time advisors. We only have people that spend all of their time getting better at helping startups. And we have this alumni network that is incredibly dedicated to helping other companies. We've also funded now many billion plus dollar companies. Yeah. And the last time I checked, no other accelerator had funded a single one. That's right. And so I think people look at that and say, well, something here is working.
[0:13:39] I could see where they draw that conclusion. Which international markets are you impressed by? Where do you see an impressive group of talent coming from? The thing that's so amazing, I spent a lot of time traveling around the world last year, is how good so many different places are. I remember this one moment at the end of the year, actually, flying back from New Zealand. New Zealand is not where you would expect to be a startup hotspot.
[0:14:10] And I was like, man, those founders were so good. Those companies were so good. But really, everywhere I've traveled in the last couple of years, the quality of founders has just gone up and up. And I think that Silicon Valley will remain dominant for a long time. I'll sort of note that Beijing is especially unbelievably impressive. And I think many founders around the world still want to come here. Not all, obviously, but a lot of good ones do.
[0:14:41] But if you believe that intelligence and determination is sort of evenly distributed around the world, then 5% of the best founders will naturally be born in the US. 95% will be born elsewhere. I do think there are things that the US does that are particularly good sort of culturally at encouraging sort of disruptive, crazy innovation. So maybe some people will come here. I hope they keep coming here. I hope that we get out of our way on this immigration thing
[0:15:11] Really wish we had done so earlier. But I think that the talent around the world is amazing everywhere, and it's like impossible to pick one great place. Our hashtag is startupgrind. I once heard that you guys take more founders from the Waterloo, kind of Toronto area than any other international place. Is that an accurate statistic that you know? On a per capita basis. Yeah, amazing place.
[0:15:40] Yeah, I was there last year, too. You know, there is something about the UWaterloo in particular that trains people to think like founders. You have the right combination of an engineering background and sort of thinking about what users actually want and getting experience building stuff. So yeah, Waterloo is especially great. We brought three buses from Waterloo, so. Seems like a lot of people. It's been a three-day trip. They smell terrible, but they're nice people. Could you...
[0:16:09] Could you ever do a batch outside of Silicon Valley, do you think? We could. It's not high. I mean, we have a lot of crazy plans coming up. And I would not put running a batch outside of Silicon Valley on the short list. Most great founders, many great founders that we meet, still really want to come to Silicon Valley for a little while. They may not want to stay. But to come for three months, develop a network here,
[0:16:38] People move their families here for three months during the summer and move back. It's a hard thing. You don't encourage them to stay once they get here. You say, go home, build your company. Many of them get here and love it and decide this is the best place to build their company. But many of them are like, you know what? My users are in India. I'm going back to India. And we say, that's great. So we could run a batch somewhere else. But still, Silicon Valley works so well for startups. I mean, if you compare Silicon Valley to the other startup hubs,
[0:17:06] So like everyone talks about New York as a startup hub, right? I love New York. I think there's cool things happening in New York. As far as I know, since Bloomberg, New York has not produced say a five plus billion dollar tech company. And like why is that? I'm not totally sure. I think one thing that Silicon Valley has that is great is that although the people here very much want to get rich a lot, they're willing to wait a long time to do it. And unlike most other cities where there is
[0:17:34] The other jobs, you know, finance in New York or movies in LA or whatever, sort of have this focus on like shortish term cash comp. In Silicon Valley, people are willing to wait eight, 10 years, whatever. And it's cool to like have equity in a startup. And startups are like the number one thing here. It's not a number two to the finance or number two to the movie industry. And so I still think Silicon Valley is like the place for us to focus. The costs are getting out of control. If something is going to upset the sort of Silicon Valley dominance.
[0:18:03] There's a good opportunity. It'll be something like the price of housing just gets untenable. It's awful. And it's so stupid that we can't fix this. All the renters out there, really. Man, it's like we want to drive the startups away. It's crazy. You talk about, in the future, you've ramped up the number of startups and the number of partners, and you kind of average like a one partner to 20 startups. Is that about right? 15. 15, okay.
[0:18:36] But you've talked in the future about funding 1,000 companies, or you hope to fund 1,000 companies. It sounds like you're trying to figure out how to do that and scale how to do that. Is YC something that could ever go public? I have no desire to run a public company. And I think like most other people that don't either. So I sure hope not. Look, we're very fortunate. We invest our own money. We don't have outside LPs. LPs are actually .
[0:19:05] Generally, okay. Public shareholders are not that great. I really don't want to have some random hedge fund or bank calling me saying, you missed your quarterly earnings by a penny. This is the end of YC. I'm having to deal with the stock price. I'd rather just focus on what we do. At the end of each application, you or...
[0:19:28] You know, I think PG did this before. And Jessica talked about this last year when she was on stage. You sent out this blog post, you know, kind of, hey, we're so sorry, we can only fund who we can fund, and we got these thousands of applications coming in. Do you guys, you know, Bessemer has this kind of anti-portfolio. Is there anybody on your list that you say, hey, you know, you kind of say kudos to those guys, we missed it, but, you know, they've done a great job. Is there anybody like that, like in the YC anti-portfolio? Yeah, we do track this obsessively.
[0:19:57] You know, like much more, I think, than people would realize. Do they have a spreadsheet anywhere? We have one, we do. We have actually custom software that just tracks this. There are a few companies that are doing pretty well. We have not yet missed a huge hit, but I'm sure we will. I'm sure we will. Huge hit being a billion dollar company. Yeah, we haven't missed one yet that's applied. But the thing that makes me sad is that I'm sure there is someone we would have funded that we didn't, that had we funded, would have been a multi-billion dollar company.
[0:20:26] Probably many of those. There are a lot of companies that we fund if we didn't fund them, which has never happened. Anyway, I'm sure in the last few years we've missed someone that's going to be a huge success, because our application volume has just gotten so high. It hasn't happened. The company hasn't become really successful yet. Just a matter of time. What do you say to these entrepreneurs that come in? Some of them are pre-products. Some of them are just initial product. About raising capital and when to do that, what are some of your
[0:20:55] Kind of best practices on when to do that, what is the right point, and what the kind of the triggers are to get that machine going? You know, generally you want to raise capital either when you have to or when it's really easy. And so if the company like desperately needs money and they can't figure out any other way, then they need to raise money. Or if someone's like offering easy money on good terms, sometimes happens at the end of the YC program, you should take it because you can use it for good things. One of the things that I think has gotten really screwed up recently
[0:21:24] is that people, if they don't raise a lot of money, they don't think of themselves as a real startup. It used to be cool to do a lot with a small amount of capital. When YC started, we ourselves used to invest $12,000 to $20,000. But now people are like, well, I need a million dollars to build my product. And that's a real problem, I think. So I think it is still an incredible win for startups to be able to operate on small amounts of capital.
[0:21:53] Build their own product, not have to hire developers to build it. And this has somehow gotten really screwed up. We're going to take some questions in just a second, so be ready. Apparently we're going to start with him. All right. Better be good with that kind of excitement. What are YC's core values? Helping entrepreneurs. Honestly, the thing, to the degree that we have been successful, I think one thing that we have done differently than most other investors.
[0:22:23] Is that we are just all about the entrepreneurs. One of the things that I really respect about PG before I took over is I have seen him do things that were like way that hurt him or hurt YC financially or otherwise. But we're in the, you know, the right thing for the founder. And, you know, we're here to help founders. Like that's that's why we get out of bed every morning. And then and funding important companies like making, you know, like making stuff.
[0:22:48] One thing that I deeply believe, and I think all my partners would say the same thing, is that if we can't get growth back in this country, we're in a really bad place. I think it just breaks everything down. And the only thing at this point that's going to drive growth is innovation. And the only thing that seems to be driving innovation, Bell Labs, the U.S. government funding lots of science, that's over. It's going to be startups. And so I think there's sort of this moral imperative to...
[0:23:17] Make more startups happen. That's everyone in here's best hope. On that note, I just want to ask you one or two more things about along this line, which is you see so many entrepreneurs and they do so many different things over 700, 900. 900? Can I pick companies? 800 and something. Okay. So what's the dumbest thing you've ever seen a founder do? Nothing I can say on stage. But really dumb stuff, I'll say that. Like jail dumb? Sometimes.
[0:23:50] Really? No YC founder that I know of has ever... I have never been involved with any YC founder in jail, but I have been involved with other founders. Like you weren't a part of the crime that he committed? Is that what you mean? No, no, no, no, no. It's just like it wasn't the person the time that I have been to a jail to bail someone out was not a YC founder. Okay. What do you do if that gets you across the bridge when it comes to it? Is that something where you say, hey, look, you've crossed our values.
[0:24:20] Here's your equity back or something. Yeah, yeah, yeah. What do you do? Yeah, so we have an ethics policy. And if someone violates it, then we give them their equity back. They get to keep the cash too, but we remove them from the program. And I think it's important that we take a hard line about that. We haven't had it come up too often, as far as I know. I think it's only happened twice ever. It's actually remarkably infrequent. Yeah, it is.
[0:24:47] We funded more people than are in one class at a Stanford or Harvard or something like that. Yeah. And they seem to have much more trouble. A lot of those guys go to jail, so it is really. That's true. Clay Christensen's book, How to Measure Your Life. Jeffrey Scaling, Harvard. OK, let's give it up for Sam Altman. Thank you very much. Thanks for having me. Thank you very much. Let's take a couple of questions. Can I just ask this? You've got a room full of 1,000 people here, so please ask something not so specific to you.
[0:25:18] But something that would be helpful to the general group, and please try to ask it relatively quickly. We'd appreciate it. Yes, sir, right in the front. It's all you, buddy. Hey, guys. Sam, thank you for How to Start a Startup. It's awesome. Thanks. I wanted to ask about the trajectory of something like Y Combinator. Typically, you have to realize that in order to ride the wave, you have to start paddling a little bit ahead of it.
[0:25:43] So do you guys usually look for startups that are one to three years out of capitalization, or do you think more in the now? No, we are the most successful when we fund things that other people don't yet think are going to be a really big deal, but two years later become a big deal. And it's very hard to predict that. And the best thing that we've learned, and the thing that has saved me as an individual investor so many times, is just fund the smartest, most determined people you can find.
[0:26:14] With the clearest vision, even if you don't understand yet why it's going to be a big deal. So yeah, we try to fund things. And we love it, actually, when other investors are like, why is YC funding this crazy stuff? And then two years later, begging us to fund more of it. Because at that point, it's sort of like too late, and we funded our companies. But yeah, we try to fund companies that are working in markets that could be huge in 10 years, even if they're small today. We try to fund companies that other people don't like right now.
[0:26:43] But we think may like in the future. And that's hard to predict. We often get it wrong. Hi, Sam. You mentioned earlier that the biggest difference is that some of the entrepreneurs don't execute each week. Have you been able to parse down, is this an issue of not being able to deal with the stress or anxiety or uncertainty in terms of hitting those targets each week? Have you seen any patterns? What differentiates the men from the boys? Or the women from the girls? I think I said each hour.
[0:27:16] Not each week. And that is an important difference. But you guys measure growth at what, 7% growth is like the optimal week over week, right? Like 10. 10, okay. You know, it's partly just a personality thing. It partly can be learned. One of the things that makes me feel like we actually really help at YC is that we can often take someone in that is like a slow mover and kind of limp and make them a fast mover and really fearsome.
[0:27:47] So it is possible to teach someone to do this sometimes. There are some people that never do it and there's just a disposition issue, but a lot of the times you can just really focus on this and learn to be decisive and move quickly and yeah, I think that's a good thing to strive for. Hi, my name is Gloria and I'm curious to know, earlier you mentioned that you guys are very active in terms of supporting female entrepreneurs. What suggestions do you have for female entrepreneurs
[0:28:18] That is not technical. Well, I won't answer that for female entrepreneurs specifically because I'll just give the answer of what I suggest for founders that are not technical. And I think the answer that I give most is work with a technical founder. Don't just pick someone random off the street that you don't know. But spend whatever time it takes working another job. Develop a working relationship with someone to found a company with. I think.
[0:28:49] Not everyone has any desire to learn to code, and that's fine. People shouldn't. But it's very hard from the data that we have to have a company with no technical founders. It's not impossible, but it's a significant headwind. And there are all these reasons we've talked about. You can believe them or not. It's certainly something that in our experience we've seen is true. Companies with no one on the technical team have a much harder time getting the product built cheaply, a much harder time hiring.
[0:29:19] engineers because it's hard to evaluate or they want to work with a great technical founder. So yeah, my advice for anyone, male or female, is to find a technical co-founder. Sam, what advice would you give to an entrepreneur that is trying to change things in the government space? First, I would say that's great. I think a lot more people should be doing that. I think that in many senses, the government is the largest enterprise customer in the world.
[0:29:52] As any other enterprise customer goes, the bigger they get, the slower they move. And you have to find ways to not have that kill a startup, which is what usually happens. So figuring out ways to move quickly and start with smallish, kind of like bite-sized projects that government agencies can act on quickly seems to be really important. But in general, this is one of those areas that I think in a few years will be big that today people aren't paying attention to. And we funded many government-related startups.
[0:30:20] In the last year or so. So, Tom Dennison from Chicago. Greatest city in the world, by the way. Thank you. I was born in Chicago. Nice. What's your question, Tom? So, I was a co-founder of a startup that went through YC. And we got down to the finalists, whatever. We were flown out and did the interview and all that. Very candidly, how on earth...
[0:30:56] Can you choose whether a startup gets into YC in 10 minutes? Yeah, so this is a question that we get asked all the time. The first answer is we can't always. And the interview process, I think, for companies is not pleasant. It's pretty harsh. It's pretty quick. Often founders are sort of not super happy with us afterwards. But here's the deal with YC. A lot of what makes us work...
[0:31:25] is that we don't have a filter. We don't require intros. You don't have to be in our network. You don't have to have an in with us. And so we see tens of thousands of startups per year. And we have to spend 90% of our time advising founders. So there's a limit to how much time we can spend selecting the next batch. And so we've learned as much as we can what makes a great company. And we try to make our best guess in 10 minutes. And we try.
[0:31:55] To learn what the characteristics of a founder are. Clarity of vision, clarity of explanation, determination, passion. Some evidence that they've done something great in the past or that they have on this company. But honestly, it's really hard and we screw it up all the time, I'm sure. And the thing that sucks the most is that I know for certain there are these great companies that we've rejected.
[0:32:23] That had we accepted would have become great anyway. But the hard part of our model is that the top of our funnel is just so huge. And so we haven't figured out a better way yet. Yeah, it's not a perfect science. Let's give Sam another big round of applause. Thank you for being here.