[0:00:14] Thank you very much. It's nice to be back here. Paris is my favorite city. So this is the end of my trip through Europe and I added an extra day and I'm really happy. Like questions the best. So I was planning not to talk. I think hopefully people know what Y Combinator is. And I also apologize. I'm getting over a sore throat. But if people have questions, I can start answering them. I'm happy to talk about startups, tech, AI, anything else, things that are working on a Y Combinator. I'm sort of especially interested in...
[0:00:44] Kind of technology and the impact it's having on society these days. But really whatever you'd like. So if anyone has questions, I'm happy to start answering them. And if not, yes? With all the things you've got going on in your life at the moment, how do you balance your time and choose where to dedicate it? I think one important skill that took me a long time in life to learn is not doing things I don't want to do. And not doing things that don't matter. So I try to, every morning when I wake up, I make a list of the three big things I want to get done that day.
[0:01:11] I have a list of like all of the projects I'm working on in a given year. And I try to be, I'm not still as good as I'd like, but I try to be really good about not doing things I don't want to do. And I try to be really, really good at delegating. So I think when you're starting a startup, you have to focus on one thing, mostly because the company is not big enough to delegate big pieces. But as you get bigger,
[0:01:36] You can kind of hand over huge pieces of the business to other people, which is what I do at Y Combinator. And as soon as I get bored with something, I try to hand it over. Someone gave me that advice once, which is as soon as you get bored with something, delegate it and do something else that you find interesting. And that's worked really well for me in this role. It didn't work as well when I was running a startup where it's more about doing the same thing over and over again and not getting distracted.
[0:02:02] But you know, I figured out this set of stuff that I care about and that I think the world needs to get right around like startups and education and AI and biotech and how they all fit together. And so I try to work on things that are in that constellation and not work on things that aren't. Saying no and figuring out what you can say no to, I think was the most valuable professional skill I learned.
[0:02:27] Hi, Sam. Thank you for being here. I want you to know your advice on, I've seen your request for startups on the Y Combinator page. You have some topics there that require very long time to market, like energy. I'm actually one of the startups who can get into something like this. We access to market in 2020. So is Y Combinator actually looking for this kind of startups or do you want something more like digital to help energy or real deep tech energy? I don't think this is true for most investors, unfortunately.
[0:02:55] I think most investors do want things that'll have results in a short timeframe. We love things that take a long time. I'm on the board, I'm chairman of the board of two nuclear energy companies. I'm on the board of a supersonic airplane company. So the companies that I work on in particular are ones that take a very long time to market. I think the biggest arbitrage opportunity in the market right now is Time Horizon. And if you're willing to dedicate yourself to something that takes a long time right now,
[0:03:25] There's a really good opportunity because most people don't. Most people want things that are going to have a result in two years or three years or one year. And even then, that doesn't usually work. But there are very few people who are willing to say, I'm going to start on this project and it's going to take five years before we even know if it's going to work. Now, the trick is even if you do something like that, even if you do something that takes a very long time, it's important that the company operate
[0:03:52] And move as quickly as a company that can ship their software every day. And we've learned a lot now about companies that do this versus companies that don't. If something's going to take five years, it's very easy to say, well, I'll just start tomorrow and I'll go to this thing today instead. The hard tech companies that we've been fortunate to be part of that have been most successful are the ones that even though what they're doing takes a long time, they work as if their work is as urgent as if they had to ship a piece of software tomorrow.
[0:04:22] Our most successful exit of last year was Cruise. It was a self-driving car company. And the thing that amazed me is even when they were years away from a self-driving car when they first started, they still worked with the same intensity as Airbnb needing to ship a bug and fix a bug by tomorrow. So I think we like it, but it's really hard to get the culture right. And it's really hard to instill that value of...
[0:04:47] The hours matter as much as they do in any other company. We're working on a new marketplace for co-working. And I know from all the information that you shared, let's take Airbnb, the marketplace situation focus on the quality and not the quantity. I wanted to know if you think that's always the case or are there situations like probably ours where we should focus on the quantity? Because I started meeting with spaces here now and I...
[0:05:15] I got five spaces in one day. So if I go on like this, it will take a long time and we won't be able to offer the best experience for people looking for co-working. Sure. Two thoughts. One general about startups and one specifically about marketplaces. The first one is a general observation about startups. We went through and looked at our history of our most successful startups. Everyone says they want to capture a large market. They want to dominate a large market.
[0:05:45] And there are two ways to do that. One is you start with a small market, take all of it over, and then expand. And the other is you start with a large market and take a small part of it over and then expand. And in either case, you sort of go horizontal first or vertical first, but eventually you try to have all of a large market. Now, the thing that surprised me is when we looked back at our own data, all of our successes, I think literally all, started with...
[0:06:13] Capturing a large part of a small market. And in fact, I think many of our failures that I was very bullish about failed because their initial markets were too big. And I can talk more about why that doesn't work if you're interested. But in any case, I think what you really want is to find the smallest possible cohesive subset of your market. The most narrow, smallest market you can find with users that desperately need what you're doing and go after that first. Once you take over that market,
[0:06:43] Then you can expand. And I think this is true for marketplaces, but it's also true for Reddit or Dropbox or Stripe. Anything we've funded has had this characteristic. The markets that are too big are too hard to really take over all of, and you do need to get to a point where you have very high penetration in some specific market. And by far, it's easier to get there if you have users that really, really want what you're doing. So, specifically on marketplaces,
[0:07:13] The hard part of a marketplace is that you have two sides that you need to move up the ladder at roughly the same pace. And the reason that quality generally works more than quantity is the same idea of you need to start with the most passionate users. And if the marketplace gets too big on either side, someone's going to be unhappy. So in practice, there's this ramp up period where you want to start usually, almost always, with a small number of people on both sides and super high quality, as in people that really, really want it.
[0:07:43] Hi Sam, you might know that universal basic income is one of the proposals of one of the major presidential candidates. You do have a program about it, so can you share why you decided to take that on and what you've learned so far? First of all, I'm not sure that universal basic income is the right solution. It's one that we want to test, but here's the problem. The problem is that I think all repetitive human work, which is 75-80% of what most of us do,
[0:08:11] is going to be done better by machines in the next 10 years. And this is a shift that most people never live through. We will live through once in our lifetimes, but this is on the order of the industrial revolution, the agricultural revolution. And this is what technology does. It has small periods where it gets a little better, and then all of a sudden something fundamentally comes along that reshapes how entire society works, how we live together, what we do with our time.
[0:08:40] What our economic system looks like, what our social system looks like. And I think what's about to happen is going to be on that order. And because it doesn't feel that different than what's happened in the last 10 or 20 years, I don't think we're paying enough attention to it. So the question is what happens when 75%, 80% of the jobs that exist today look very different or don't exist at all in 10 or 20 years. This is in no one's living memory.
[0:09:12] We can read history books about this, and I suggest that you do. It's pretty interesting. But we have a major, major change to get through. And we, all of us, you know, the people that work on technology, I think it's our responsibility, or at least we're certainly going to be held responsible by the people whose lives we change as AI rolls out. So the question is, what happens on the other side? And how do we get through the middle? The other side I'm more optimistic about.
[0:09:41] Human demand is limitless. People will find new things to do, new things to want. There will be new jobs even if we can't imagine them today. Probably someone in 1800 couldn't have imagined computer programmers as a job either. However, it's going to be a very rough transition and it's going to happen faster than the agricultural revolution and the industrial revolution happened. And so how do we get through it?
[0:10:08] How do we give people enough money to live on who get displaced? How do we give people the same sense of purpose, fulfillment, social structure, validation that they get from a job today? And that's what we're trying to study. Universal basic income is one idea. I don't think it's the entire thing. I think you also have to think about retraining new social structures. But specifically, what we're studying with universal basic income is not what people think.
[0:10:33] When politicians talk about universal basic income, it's always about cost. How can we possibly afford this? How can this ever make economic sense? And I think this is one of the stupidest arguments in the world. If the AI really does come and take all the jobs, it will do that because it can do them for free. And so the cost of a really good quality of life will go down a huge amount, a factor of 10, maybe a factor of 20. And it's not that hard to imagine a world where you can tell a really smart robot
[0:11:02] to go dig up raw material, refine it, and build you a house for the cost of energy and land. And no one thinks about this. So the cost is not an issue to me. If all the jobs go away, the good news is the cost of a really good quality of life is going to come way, way down. We'll easily be able to afford this. What I don't think people have really addressed or studied is what does this do to communities? What does this do to people's fulfillment? No matter how much people complain about their jobs,
[0:11:30] They like having somewhere to go every day. They like to be needed. They like being important. They like having colleagues. They like having a sense of mission in their lives. And what are we going to replace that with? So what we're studying is that. What systems build up in a community when no one needs to work? And what do people do for these sort of very human needs? The study will run for five more years, and hopefully we get some good data out of it. Hi, Sam.
[0:11:59] I saw that you recently created a vertical group for AI companies at YC. What's your personal opinion about AI and do you have any special expectation for this tech? So, two things about AI. One is the thing that I'm most interested in is the research and development. So, Elon Musk and I co-chair a group in San Francisco called OpenAI that is focused on trying to invent the algorithm for intelligence.
[0:12:27] I think we have a pretty good chance at it. I think we now have 50 or 60 of the most talented researchers in the world. We have some incredible results coming later this year. But we're trying to figure out, you know, can we write all the code? Can we invent, figure out, whatever you want to call it, the code to make a computer smarter than a human in every way? And I think we'll be able to do that. Even if we can't, though, just with the research we have today.
[0:12:53] I think it has gotten to a place where the easiest layup in all of startups, as close as a guarantee as you can ever have, is to take existing machine learning technology and apply it to every vertical. Self-driving, farming, radiology, law, accounting, everything. And you can build a way better thing than exists in any of these markets with today's machine learning technology. And I don't think this was true two years ago.
[0:13:23] We have this vision where we fund one company in every vertical and they use machine learning to start a startup. And we think that's so important and so powerful and going to be such an incredible economic return that we decided we would, for the first time, make a special track in YC and provide those companies with extra support, things that they need differently, and also advice from AI experts. We're going to try that just for one batch this summer. We love to try experiments at YC.
[0:13:53] And if it works, we'll keep doing it. If it doesn't work, we'll stop. Almost certainly, we'll tweak it. But I think this is... I normally hate tech trends. And anytime someone starts talking about a trend, I think that they don't really know what they're talking about. But I think this one is real. Hello. What is your opinion about a copycat startup? Is it a good idea to do that in his own market? And how to protect his startup from a copycat?
[0:14:23] You know, people worry about copycat startups a lot. And I don't think you need to. We did a study on this as well once. There are very few startups that are true copycat startups that go on to be successful. The copycats almost always fail. They're always a year or two behind. The people aren't as good. They don't have the same sense of mission. So there are some examples. There's some debate about who copied who in Uber and Lyft.
[0:14:52] But that is an example of where you have two serious players with the same idea. It doesn't happen that often. Usually it's a pretty new idea. You could sort of say Facebook was a copy of MySpace, but it really wasn't, honestly. It was a pretty new idea. You could sort of say Google was a copycat of the 12 search engines before it, but again, it really wasn't. They had this one big new technical idea that made it super different.
[0:15:19] There are a few things that are more stressful as a founder than having a copycat launch. But they don't really cause problems. And I think it takes many years to learn to ignore them. But I believe that is the right answer. One rule that we have at Y Combinator is that we won't talk to you about your competitor that you're stressing about until they're beating you in the market. Because that almost never happens. Now, there is one subset of this.
[0:15:49] Which is copycats in other markets. And the normal version of this is, let's say a French company looks at a US company and says, that's great, I'm gonna copy it in France. That doesn't work either, or at least doesn't work for long. Usually the company that started first will eventually muscle you out. But here is the important point that I'd like to make. We have had great success in particular funding French startups, and it's nice to see some of you here.
[0:16:18] French startups have been an incredible... I guess I'm saying this live, so I may get in trouble for this. We've had much better luck funding French startups than we have with most other countries in the world. Some phenomenal, phenomenal founders. You all can compete with anyone in the world. And you know things that are different than people in the US. You have new ideas that we don't have. Just anyone in a different context, living a different life, will come up with different things. So don't start the French version of X.
[0:16:49] Why? Which no one in the US has thought of, but will be a worldwide company. Just because no one's doing it in the US, we're not any smarter than you are. We don't know anything you don't know. We don't have any more talent than you do. So come up with your own idea and make us copy you. Hi, Sam. With my friend, we launched three months ago a startup called Peter. It's a chatbot on Messenger that helps students do their homework together. It's focused on...
[0:17:19] And we got a crazy growth. Like today, we have hundreds of thousands of active users, purely organically. And my question is simple. So today, we have students helping each other, exchanging on Messenger through robots, images, text, et cetera, to help on homework. And do you think that these data could be valuable to one day create maybe an AI of a tutor?
[0:17:49] For sure, yeah. I mean, that's an easy question. I don't have a long answer for that. Yes, I think that's a great training set. Congratulations, by the way, on the growth. Hi, Sam. Thank you for being here. So, Alice told us about the family's vision for Paris. What's your vision of the Parisian ecosystem? And please, you don't have to be nice to us. Speak your mind. Again, I really do love Paris, and we have been particularly impressed with the traction we've seen with French startups. Here's the one thing I would add.
[0:18:18] The thing that is magic about Silicon Valley is that it is a place where you can have a crazy idea. You can say, I'm going to go build electric cars, even though I don't know anything about cars or batteries. And I'm going to do that to save the world. And people don't make fun of you, either to your face or behind your back. There's not this thing of, I'm too ambitious. You're too ambitious. That's never going to work. People don't.
[0:18:47] They don't build themselves up by sort of making fun of your ambitious ideas. And because they've learned that 95% of the time when you do that, you're right. But the 5% of the time that you're wrong is a very expensive mistake. And so you have this very unusual thing that people aggressively believe in the future and they aren't mocked for trying to build it. And that I think is how you end up with
[0:19:17] Very ambitious companies that go on to be super successful. That is missing almost everywhere in the world. That idea that the society as a whole accepts that people are going to do these things that are very risky and they're probably not going to work, but we still support each other. We never make fun of each other for that. If there's one thing I would say to try to cultivate in a startup ecosystem, it's that. Because that's how you get the really outsized, really important companies. Hi.
[0:19:46] Earlier, you talked about building computers and machines that are smarter than people. And my question is, what happens exactly when we are there? And we have technology that outsmarts people on a daily basis. And the whole society is dependent upon this technology, but only a few people can actually understand and master it. I'm thinking influencing customer behavior, having people do things that they don't really realize they're doing. Who is responsible for that?
[0:20:14] So it's a great question. That's something I think about hours every day. What happens when we get computers that are smarter than people? How do we control that? Who gets to decide? The reason we started OpenAI is I don't want to live in a world where one company controls all intelligence in the universe. When we get to a place where the smartest human is only one one-billionth as smart as an AI, I don't want one company controlling that.
[0:20:44] This is something that feels like should be owned by us collectively. So OpenAI is a nonprofit. We don't have any obligation to maximize corporate profits, anything like that. Our only goal is to do this and have it go well for everyone and to distribute the benefits and the control to everyone. We don't think the four people that get the program right are the ones that deserve to make all the decisions about what happens from T0 forward. And we don't think that those people should get all the economic benefits either.
[0:21:14] So we really want this to be a collective extension of human will where we all get to decide what happens with it and we all get to benefit from it. And I don't have like a lot of free time. I don't normally start other companies, but this one seemed so important. This technological shift belonging to the people and not to one company seems so important that it's now something I dedicate like a day a week to. Hi Sam. Hi. I just read that the first Moroccan company ever joined Y Combinator.
[0:21:43] I'm pretty excited about that. Excellent. Yes. I'm from Morocco, and I'm putting together a conference to discuss the effect on automation on the developing world, the poor world, basically. Would you be interested in speaking at it? Is that something? Yeah, sure. I mean, my schedule gets booked out like a year in advance, but if we can make it work. So that's going to be when you're available. If we can make it work, I'd love to come. Send me an email. Okay. You can tell me how I can get in touch with you. Thank you. Sure. Hi, Sam.
[0:22:11] I just want to have your opinion about platform business models like when you put consumers and businesses and you capture value on transaction. So you have like very great examples like Airbnb which really work well and you have also bad examples like Homejoy. So what's the secret? So how would you describe the...
[0:22:40] So we value them. Sure. Talk about a few areas that I think you really need to pay attention to. One is how strong the network effect is. So Airbnb is one of the greatest platforms of all time that I've ever seen because the network effect is so strong. You can only list your house on one platform because of instant booking. And you as a visitor really care which house you go to.
[0:23:07] This house is very different than this one. And compare that to a business like a ride sharing company or a house cleaning company where you don't care too much which car you go in and most people don't care too much who cleans their house as long as it's clean. And a car can list across multiple services. A cleaner can be on multiple services. That's a much weaker network effect, a much weaker lock-in. So I think the most important thing to understand about what makes
[0:23:36] One marketplace good, one marketplace bad is how intensely are people locked into that one platform. Does that make sense? And you want a business obviously where there's exactly one platform and everyone has to use it and people care a lot about which specific good they get. Very low fungibility of goods. So I think that's the most important thing that makes it different. Second is unit economics. So there are some platforms where there just aren't enough.
[0:24:05] There are some marketplaces where people use them super frequently.
[0:24:31] The other thing is too weak. There's some where all the dynamics in the marketplace are really strong, but people just don't use them enough. It's something that you buy not even once every year. We had a scientific instrument marketplace that had this problem. So you need some frequency of use. And then the final thing is in the very, very best, buyers become sellers. So a marketplace like HomeJoy doesn't have this much, maybe never has it. A marketplace like Airbnb or eBay,
[0:25:02] A lot of people will buy some collectible or go on vacation and then say, wow, this was cool. I can sell my stuff or I can list my house. And that's a really nice thing in the flywheel. This is not really tech related, but since you've been pretty outspoken politically, I feel like I can ask. It seems like the US is very divided right now. And I'm afraid that France is heading in the same direction. What do you think could be done to avoid
[0:25:32] This kind of division between people could be tech related because a lot of tech companies took stands mostly against Trump or people for, so I'm curious about that. Great question. One of the questions of our time for sure. I think tech has an opportunity and an obligation to build tech that connects us better. Right now, a lot of tech companies will do the things that makes user engagement go up the most.
[0:26:02] And this is what we show in a feed. This is the next video that we play for you. And the effect of this has been an increasingly divided country, world, where everyone gets into their own filter bubble. They see the content they want to see. They see the content that keeps them engaged. But most people don't actually want to hear the other side. They want to believe that they're right. They want to confirm their existing misconceptions or conceptions, let's say.
[0:26:32] This has led, you know, there was a time before the internet where we had to talk to our neighbors if we wanted any human communication at all. And so we kind of were forced to have some diversity of opinion. Now we can go pick and only read exactly what we want to read and only communicate with people that think exactly like us. And I think tech companies have an obligation to stop that. I don't think people did it intentionally. I think something about human behavior has just been exposed. But now that we know that, it's time to fix it. I heard recently that
[0:27:03] One of the big video sites, someone had done, just last week, done an analysis on this. And if you start watching political videos, and the algorithm, no matter what you believe, and the algorithm is just trying to make you keep watching videos for as long as possible, and keep you engaged for as long as possible, eventually you end up watching Le Pen videos. All roads lead there. And that's the dead end of the cycle, and you just keep watching them. And, you know, this is what...
[0:27:32] What machine learning has figured out is what makes people most engaged. But the kind of Le Pen videos it shows are not the ones that bring society together. And so I think we have a real duty that's going to be hard as a tech community to realize that we have to address this problem. We have to get people out of their filter bubbles and we need to kind of come back as one world and try to understand the other side. The other thing that we talked about earlier is if
[0:28:04] If we don't figure out a way to address inequality and anxiety about the future, then there will be a lot of people who are just really nervous and vote in these ways where they say, you know what, I don't want any more change. And this can be about my job, or it can be about immigrants, or it can be about anything else. But fundamentally, it is fear of change, I think, that drives it, or jealousy of some sort. And I think there's a lot of things we can do to address that.
[0:28:32] But if we don't, I think we'll continue to have these problems. Hi, Sam. So I'm going to relate to this question. I'm a French founder. I started my company four years ago, and I raised the seed round two years ago, 300K. So we have nice numbers. We have nice curves.
[0:28:51] But we have a very strong social focus. We work for locally elected people. And our goal, I really care about improving democracy at a local level. And it also relates to a movement in California called Zebra Startup. The claim is quite ambitious. It's zebras fix what unicorn break. A zebra being a real animal, very cooperative, and working for shared profits. So since you like to experiment, my question is, when do you open a zebra program in YC?
[0:29:20] I don't think I know quite enough about what a Zebra program is from that explanation. I'm not opposed, but I don't think I quite got a... We try to have a very high bar for doing things. Maybe we can talk afterwards and you can explain to me more what it is. Hi, Sam. I have three questions for you. So the first two are related. I'm planning to... I'm assessing the idea of opening a European business accelerator.
[0:29:50] For US startup, so to help US startup to come into Europe and do business here. My question are, first, do you know how many successful startups are coming into Europe? What's the percentage between the one that come here and the one that stays in the US? And my second question is, do you know if successful startups are coming first in Latin America or in Europe?
[0:30:19] When they want to go international. So the question is, do European startups that go through YC come back to Europe? No, U.S. startups. Oh, do U.S. startups expand to Europe? Yeah. Or Latin America first. Got it. I think a lot of most successful startups eventually expand worldwide. I think the hard part about an accelerator for that is they usually don't expand until they're doing really well in the U.S., at which point they might think they're beat.
[0:30:48] They're beyond the stage of an accelerator. So I think that would be the challenge and the thing to really ask your customers about the startups in this case. And I think in terms of Latin America or Europe first, that really depends on the particular startup. I'd say more often Europe, but certainly not always. Okay, thanks. Sure. And a more generic question, do you think in the next two or three years we will see another big social network like Instagram or Snapchat?
[0:31:19] Do we think there will be a big social network in the next two or three years? I think one will start, but it won't be obvious. So I think the next big social network will start in the next two or three years, but it may not be obvious until five years out that it's going to be huge. Hi, Sam. You talked earlier about artificial intelligence on the machine replacing people's jobs. What do you think are going to be the next big shift in other fields, like energy or transportation?
[0:31:48] And the different question, what do you hope are going to be the next big shift? So one area that I'm particularly excited about is healthcare. I think there, you know, so in the US, just to pick on radiologists again, there's a lot of resistance from US radiologists because they don't want to lose their jobs to computers, understandably, even if they're better. But I don't think, I think it's like a big problem if you have...
[0:32:17] Better technology, you can use it to save lives, you need to use it. One thing though that I have found heartening is that in the developing world where there just aren't radiologists, it's much easier to start with something. And so I think, I was not sure about how the whole evolution of medical AI was going to happen.
[0:32:37] Because there'd be such resistance to it. And now I'm already seeing it happen in the developing world. And I think it'll happen there first. And then people in the developed world will say, well, wait a minute, I want better care too. So I think that's one field that I think will happen and where I'm very happy to see it. There's like many, many others. One that I'm particularly nervous about is computer security. So I think people are really going to use machine learning to break into computer systems at a level we're not prepared for. The good news is I think we can use...
[0:33:06] AI to defend that as well, but that's an important area. Could you please tell us about Y Combinator? What do you inspire you, and most importantly, from outside of the Silicon Valley, please? So what inspires me and what else? The Y Combinator, what inspires you and the Y Combinator, but from the outside of the Silicon Valley. It could be a structure, people, startup, whatever. I'm just trying to understand this question. So what inspires us that we see outside of Silicon Valley?
[0:33:37] Yeah, exactly. More than half of the founders that we fund are outside of the US. And probably 75% are outside of Silicon Valley, maybe more. Look, I certainly believe that talent is roughly equally distributed around the world. And Silicon Valley is a tiny, tiny piece of the world. I think the next great founders won't be born in Silicon Valley. They'll come just statistically. They're much more likely to be born elsewhere.
[0:34:09] I'd say most of the most important companies we funded, those founders did not come from Silicon Valley. They came from Ireland or Seattle or India or Boston. And the thing that inspires us is that we've now been able to build up a program where we get to draw the best people from the entire world all together and create this community of people that are really great working on very different things. And they help each other. And there's this global network effect around that, which is cool.
[0:34:39] What is this thing most of the people believe here that you don't? What do I think most of the people here believe that I don't? Yeah, what is the thing? Well, I don't know what most people here believe. That's always a hard question. You have French entrepreneurs mainly. I'm thinking. It's really hard to say what a group of people believe ever. And it's really, really hard when it's a different culture from yours. I mean, something about being too ambitious being bad. I think probably like...
[0:35:12] Something along those lines where I believe that sort of like to a first approximation the more ambitious the better and it's okay to take big risks and do things that other people think are silly or stupid and I'm probably off the consensus of the group there. Like it's a crazy thing right to wake up one morning and say you know what I'm gonna be the one to start the organization that's gonna develop super AI. Like that is somehow a deeply American thing to do.
[0:35:40] Hi, Sam. I'm the CEO of Brainy that enables every consumer to share their innovative ideas with brands. And we're entering the US in Q3, so I would like your top three advice for French startups to have a success in our first year in the US. I have only one piece of advice, but it's an important one. Don't think of yourself as a French startup. People always think that there's something embedded in there, which is we are different.
[0:36:09] Hi. If somebody were to tell you, or if somebody were to hold a gun to your head and say, Sam, stop doing what you're doing. Start something else now. What would you do?
[0:36:39] Well, I'd be tempted to get into politics at the moment because it feels like it's on such a bad path. Probably that. But I really like what I'm doing. So this was if I had to do something else. A follow-up on the question about politics, the divide of the world. I've come to realize that it's also a divide between huge cities. We're in a huge city right now. And the countryside.
[0:37:06] Have you seen anything that tries to bridge that divide, and how do we go forward on this topic? The city versus non-city divide is definitely real. At least one thing I expect is just continued movement of more and more people towards the cities. I think the other thing that will work, so that's one dynamic that may help, and the other dynamic is to push more and more of the jobs and wealth to the non-cities.
[0:37:37] Funding a number of companies that are trying to work on that. But I think the internet can really help with this. The internet for all of its problems does have the ability to equalize the playing field with remote work and education and connections to people all over. So there is still a big cultural divide there and that is going to take a while. The economic divide I'm optimistic we can make better.
[0:38:05] Hey, my name is Oana. I'm Romanian and I'm the co-founder and vice president of the global community of Romanians in IT. We've been working for it for one year. My question for you is a little bit more general. I know you've said before that the difference between the US and Paris, for example, is that people are more open-minded about and they don't make fun of ambitious ideas.
[0:38:32] I'm very interested to see what's your opinion about the difference between the tech and the startup scene from the US and the one from Europe. And more specific, if you are familiar with it, the startup and tech scene in Romania and what do you think we could do to become more competitive in the tech scene in Europe and internationally? Everyone loves to ask me this question and what I say is just...
[0:39:02] It's not that different. The people are equally smart. The differences are the surrounding level of background ambition, the availability of capital. But those are the two differences. And those are the things that I think Silicon Valley does really well. But that's replicable elsewhere. And so everyone wants to know why they're special and bad, both of those things. And the answer is like,
[0:39:33] Technology is available to everybody. And one thing that has really been working at Y Combinator has been our international expansion. We have now funded very successful companies from many dozens of countries. So I think there's more alike than there is different. All right. Thank you all so much for spending your afternoon with me.