Ben Horowitz: Why hesitation kills more startups than bets

1. 0:00 – 4:09

   Introduction to Ben Horowitz

   1. BHBen Horowitz0:00

      (instrumental music) The worst thing that you do as a leader is you hesitate on the next decision. The thing that causes you to hesitate is both decisions are horrible. Probably one of my bigger ones on that was we went public with $2 million in trailing 12 months revenue at 18 months old. That's obviously a bad idea. But the truth of it was the alternative was going bankrupt, and that's a worse idea.

   2. LRLenny Rachitsky0:23

      It's very difficult and painful to be a CEO, to be a founder. In spite of that, so many people want to start companies.

   3. BHBen Horowitz0:29

      The psychological muscle you have to build to be a great leader is to be able to look in the abyss and go, "Okay, that way's slightly better. We're going to go that way." If everybody agrees with the decision, then you didn't add any value, 'cause they would have done that without you. So the only value you ever add is when you make a decision that most people don't like.

   4. LRLenny Rachitsky0:47

      You are famous for writing one of the most popular pieces of literature for product managers.

   5. BHBen Horowitz0:52

      What I was trying to get at in Good Product Manager, Bad Product Manager was the job is fundamentally a leadership job, and it's a tricky leadership job because nobody is actually reporting to you.

   6. LRLenny Rachitsky1:06

      There's always this kind of sense that the PM is not the mini CEO, how dare you call yourself that. I actually think that's exactly what the PM is.

   7. BHBen Horowitz1:12

      It doesn't matter if you write a good spec or you have a good interview or you do this or do that. What matters is that the product wins.

   8. LRLenny Rachitsky1:19

      (instrumental music) Today my guest is Ben Horowitz. Ben is the Z in a16z, the world's largest venture capital firm with over $46 billion in committed capital. They're investors in OpenAI, Cursor, Anduril, Databricks, Figma, basically every generational tech company. He's also the author of two New York Times best-selling books, The Hard Thing About Hard Things and What You Do Is Who You Are. Ben is endlessly fascinating. He started a rap group when he was younger. He started his career as a product manager and wrote the now famous Good Product Manager, Bad Product Manager piece. In our wide-ranging conversation, we cover a ton of ground and Ben shares stories and insights that he's never shared anywhere else. A huge thank you to Shaka Senghor, Ali Ghodsi, and Adam Neumann for suggesting topics for this conversation. If you enjoy this podcast, don't forget to subscribe and follow it in your favorite podcasting app or YouTube. It helps tremendously. And if you become an annual subscriber of my newsletter, you get a year free of 15 incredible products, including a year free of Lovable, Replit, Bolt, N8N, Linear, Superhuman, Descript, Whisperflow, Gamma, Perplexity, Warp, Granola, Magic Patterns, Raycast, ChatPRD, and Mobben. Check it out at lennysnewsletter.com and click Product Pass. With that, I bring you Ben Horowitz. Today's episode is brought to you by DX, the developer intelligence platform designed by leading researchers. To thrive in the AI era, organizations need to adapt quickly, but many organization leaders struggle to answer pressing questions like which tools are working, how are they being used, what's actually driving value? DX provides the data and insights that leaders need to navigate this shift. With DX, companies like Dropbox, Booking.com, Adyen, and Intercom get a deep understanding of how AI is providing value to their developers and what impact AI is having on engineering productivity. To learn more, visit DX's website at getdx.com/lenny. That's getdx.com/lenny. This episode is brought to you by Basecamp. Basecamp is the famously straightforward project management system from 37signals. Most project management systems are either inadequate or frustratingly complex, but Basecamp is refreshingly clear. It's simple to get started, easy to organize, and Basecamp's visual tools help you see exactly what everyone is working on and how all work is progressing. Keep all your files and conversations about projects directly connected to the projects themselves so that you always know where stuff is and you're not constantly switching contexts. Running a business is hard. Managing your projects should be easy. I've been a longtime fan of what 37Signals has been up to and I'm really excited to be sharing this with you. Sign up for a free account at basecamp.com/lenny. Get somewhere with Basecamp.

2. 4:09 – 10:15

   Important leadership lessons from Shaka Senghor

   1. LRLenny Rachitsky4:10

      (instrumental music) Ben, thank you so much for being here and welcome to the podcast.

   2. BHBen Horowitz4:17

      All right. Thank you, Lenny. Excited to be here.

   3. LRLenny Rachitsky4:19

      I'm even more excited to have you here. I want to start with a question that a close friend of yours suggested I ask you, uh, Shaka Senghor. So Shaka, he's... Like, we could do an hour just on how interesting this guy is and the things he's done.

   4. BHBen Horowitz4:32

      Three hours on Joe Rogan He's very...

   5. LRLenny Rachitsky4:34

      (laughs) Okay. So we're not going to do that. Just to give a glimpse, he, uh, he was in prison for 19 years. He was in solitary for seven years. He, uh, led a huge prison gang. You wrote, you wrote, uh, about him in your book, uh, as a great exemplar of great culture in the prison gang that he ran. Uh, so interesting. But, uh, something that he learned from you that he told me I need to ask you about is about success and how to be successful and how it's not what people think. And he said that you learned this lesson from a pilot. What is that story? What is that lesson?

   6. BHBen Horowitz5:07

      Well, it's kind of a... I mean, I'd say it's a long life lesson, but the, the pilot story is, um... I actually... You know, I ask people, like, silly questions sometimes when I meet them and so, you know, I met this gentleman who was a pilot and, you know, it was right around the time, uh, JFK Jr. crashed his airplane, uh, and, and ultimately died. And I asked him, I was like, you know, "What happened?" Because there's always... You know, the story in the press, and I know this from them writing about me or anything, is it's always what's the best narrative, not what's true. So you can never kind of actually find out what happened. You just find, like, the best story version of what happened. Uh, and, you know, the story in the press was all about, oh, he wasn't trained on instruments, instrumentation, was flying at night, and I wanted to know is that right? And the pilot said, "Well," he said, "Really it's like... All plane crashes are a series."... of bad decisions. And n- none of the decisions by themselves is that bad, but when you add them up, it's bad. So the first decision was he needed to get wherever he was going and that was the priority. And in flying, that can't ever be the priority. (laughs) Uh, you know, because there are conditions, there are things that happen. And then the second one was, well, like his timing of when the sun would go down was wrong, so he thought he'd be flying in sunlight and he wasn't. And then, you know, once he got up there, it was, uh, you know, when the plane was, uh, you know, going down, kind of making it go up, uh, was a bad decision 'cause he was upside down. And like, you know, so, so it's... And I can't remember all the things, but this guy had like 17 steps, you know, 17 bad decisions in a row and the big thing for me that, that I felt was really true is, it's, you know, one decision leads to another. Uh, and so, you know, if you can (laughs) break, you know, if psychologically you can take the sunk cost, um, then that gets you out of a lot of bad paths and then, you know, a little good decision may be difficult, but you have to believe it's gonna lead to the next one. And, you know, a lot of success is about that. You know, it's a, it's a small thing, a small thing that's hard to do that doesn't seem to have a high impact, um, but it leads to the next small hard-to-do thing and then eventually, uh, you get an outcome, so, so that was kind of the concept, but...

   7. LRLenny Rachitsky7:46

      So the lesson there is just success is just a bunch of little things. It's not this, like, "Oh, I got here," and a big thing and now I'm successful.

   8. BHBen Horowitz7:52

      Yeah, or if somebody were to write a story about me, they would be like, "Then Ben did this really smart thing and blah, blah, blah," you know, happy ending. But, you know, it really wasn't like that. I don't think it's like that for you or anybody. Um, and then, you know, I spent a lot of time with Shaak on, you know, how... 'Cause, 'cause it's always your own psychology that, you know, gets you, and one of the kind of most insightful things he said to me is, you know, 'cause he was in soli-... Most people who are in solitary for seven years, (laughs) uh, that's it, you're insane, like, you're never coming back from that, it's just an impossible thing. But if you, like, study his story, he actually kind of really was massively self-improved coming out of solitary which is, you know, like... And he wouldn't recommend that for anybody, just to be clear. It wasn't solitary. But what it was, you know, is he changed... In solitary, he was able to change a big set of beliefs that he had about himself that kind of got him out of that. And the thing that... His conclusion from it, which I thought was really interesting, he's like, "Look, I was in prison for 19 years, I was in solitary for seven. I come out, I can't rent an apartment, I can't vote, uh, you know, I can't get a gun, I can't do..." Like, like, no rights. None of that was anything compared to what I did to myself. And I think that's very true for CEOs in general and people in general, is that, like, all the things that you perceive that are happening to you th- that are bad, you know, be it, like, the system's against you or somebody undercut you or racism or sexism or this or that or the other, is very small compared to... Like, it means a lot if you believe it. (laughs) You know, if you believe what people say about you, if you believe what they did to you, um, then, then that destroys you, but if you, if you go like, "That's not me," um, you can overcome almost anything and that's... You know, he's got a new book out on that anyway that, that I think is very good because that's the... You know, I say more than anything, that's the key to success.

   9. LRLenny Rachitsky9:51

      Something you... It's like if you look at all the writing you've done, it's essentially about the struggle and pain and suffering of being a CEO at your heart. You know, your first book-

   10. BHBen Horowitz9:58

       Yes, right.

   11. LRLenny Rachitsky9:58

       ... The Hard Thing About Hard Things. There's a lot of talk these days about just how important struggle is and how valuable it is to go through struggle. Jensen's big on this, you know, he talks a lot about just you have to go through pain and suffering to be a great leader.

   12. BHBen Horowitz10:10

       Yeah. (laughs) Yeah, but y- you don't really have a choice, you know, like that, that, that's true.

3. 10:15 – 19:35

   Running toward fear and why hesitation kills companies

   1. LRLenny Rachitsky10:15

      There's something that I saw you share that I love which is running towards fear versus running away from fear, something that you tell all your leaders to, to work on. Uh, easy to hear, uh, hard to do. We don't like doing things that are scary, running towards things that are scary. Why is this so important? Why is this something people need to learn to do?

   2. BHBen Horowitz10:32

      Well, so the biggest mistake that you make, or the, the, the worst thing that you do as a leader, like, there's things in your control and there's things, like, out of your control. And hesitation is, uh, you know, that, that's generally the most destructive and I, and I go through all the ways that it's destructive, but it's extremely bad. And the thing that causes you to hesitate is, you know, both decisions are horrible, right? Like there, there is... Like e- it, it's not business school where, like, you're going through a case study and if you had done that, then the company would've gone this way, but if you had done that, it's a great success. Like that's not actually, you know, what happens to you as CEO. What happens to you as CEO, it's like, "Okay, if we rearchit-... You know, like this product is, the architecture is not gonna actually get us where we need to go, I kinda know that, but if we rearchitect it, like, we're gonna probably miss all the features, miss the quarter, have trouble raising money, you know, et cetera, et cetera." So like, that's really bad (laughs) and then not rearchitecting it is really bad. And so I'm just gonna try to, like, and avoid this subject 'cause I don't, I don't even wanna deal with either of those. And that's tha- that's the worst thing because if we archi-... You know, if action is the better choice then, you know, that's good, and then if you don't make an explicit decision then the whole company's gonna get nervous because they know that the architecture is wack and you gotta fix it. And then, you know, like, uh, probably one of my bigger ones on that was-You know, we went public with $2 million in trailing 12 months revenue at 18 months old, right? Like, and if you add up... Like, that's obviously a bad idea. I mean, (laughs) like, there's no question that wasn't a bad idea. But the truth of it was, the alternative, uh, because of where the private markets were, was going bankrupt, and that's a worse idea. And if you look at that time, you know, March of 2001 when we went public, you just look at the number of CEOs that hesitated on that and didn't do it and went bankrupt, it's a lot. And so that's, you know, that, getting good (laughs) at making a decision that everybody's gonna go, "Wow," that was insane then. Like, you went pub- like, th- The Wall Street Journal wrote a whole, like, long story about how stupid I was, and then Businessweek wrote a story called The IPO from Hell. That was the name of- (laughs) with a, like, our idea, The IPO from Hell, which, uh, like, was accurate in, in a sense. But so, like, that's really bad. But it, it wasn't as bad a- and this is why it's so scary to make that decision, 'cause you know that's gonna happen. I knew those stories would get written. Like, there was no question. Uh, and, uh, yeah. And that's, that's the kind of s- muscle. So, like, if you think about, like, the psychological muscle you have to build to be a great leader is to be able to, like, look in the abyss and go, "Okay. We're going th- that way's slightly better. We're gonna go that way." Uh, and it's very hard to do. Uh, and I would say it's, it's a thing people struggle with. And it began something s- should I fire the head of sales? So I don't wanna have that conversation. Like, and then I'll have to replace him. And then, like, there's gonna be a bad PR story. And then, and, and, and, and, and, and, and you can kind of quickly calculate all the bad stuff that's gonna happen if you do it. But if you don't do it, (laughs) uh, that's probably gonna be much worse, and, you know, kinda, that's why you have to run towards the pain and darkness.

   3. LRLenny Rachitsky14:13

      What is the advice you share with founders? 'Cause, uh, as you said, it's very hard to do this. Just, like, how-

   4. BHBen Horowitz14:17

      Yeah.

   5. LRLenny Rachitsky14:17

      What helps them actually get better at this? Is it just Ben being by their side telling them, "This is how it is"? Is there anything else you can share?

   6. BHBen Horowitz14:22

      No. No, no, no, no, no. Like, uh, I would say this is one where, you know, I can't really coach you to be good at this. Like, I can point it out and, like, so that you recognize, um, that you were slow or whatever. But it's kind of like, I was like, and when I talked to them, it's like football. Like, y- you know, you could have a really fast, great athlete, but if they don't trust their eyes, if they don't run to the ball when they see it, if they think, "Oh, maybe that was a fake," then they're that step slower, and then they're not, um, th- they'll never be as good. And CEOs are like that. You know, if you don't trust what you see and you don't run at it, then, like, you're just not gonna be good. And, uh, it, it's hard to get CEOs not to hesitate. But, like, one thing, uh, the thing that does help is, you know, they look at it and I look at it.

   7. LRLenny Rachitsky15:16

      (laughs)

   8. BHBen Horowitz15:18

      And, you know, and I kinda confirm, "No, that, that is as it appears." Uh, and you know, sometimes they're afraid of the conversation. So that one I can help with. So, you know, a CEO might be afraid. L- l- like, they wanna do something, but they, they don't know how to say it. They don't know how to have the conversation with the employee. So I can walk them through that. Um, you know, I, I had an instance where the CEO said, "Hey," you know, like, um, "I need your help in, my CTO, he's an asshole." And I was like, "Okay." You know, like, "Great." I said, "But you're not gonna fire him? 'Cause I know he's a good CTO. Or are you asking me should you fire him?" He said, "No, no, I don't wanna fire him." And I was like, "So you're asking me what to do 'cause you don't know how to have that conversation with him about being an asshole without him quitting. Like, that's what you're saying." And he goes, "Yeah, that's the problem." And so I go, "Well, like, why is he an asshole?" And he says, "Well, he's an asshole because, you know, w- the other day he made, like, a very junior young woman in our finance organization cry." And I was like, "Oh, that, that's kind of, yeah, I got you." I said, "Look, this is what I would say to him." I'd say, I'd just sit him down and I would say, "Look, you're a really good director of engineering, um, 'cause you do a great job at managing the team, get the products out, all that. But, like, you're not really a CTO because to be a CTO, you have to be effective with other parts of the organization. Um, you can't just be, like, effective only with engineering. And so, you know, and making, like, somebody cry? Like, she's never gonna do anything you want. You, you, you lost all effectiveness with all of finance by doing that. And so if you want to get good at that, I'll help you. I'll work with you on it. But if you don't, I'm gonna have to hire a CTO at some point, 'cause, like, obviously I need that." And, you know, and, and then he's like, "Oh, okay. I can have that conversation." (laughs) You know? I can't have the conversation that, "Hey, you're an asshole." (laughs) 'Cause I don't want him to quit. But I can have the conversation that's more specific. And a lot of kind of getting people not to hesitate is just getting them over that. And so, so often, like, uh, and I would say, you know, early in a CEO's career, a lot of it is just not knowing how to have the conversation.

   9. LRLenny Rachitsky17:28

      There's also, I imagine, an element of, "I just wanna be liked. I don't want people to hate me." You have this great line that, "You wanna be liked and respected in the long run, not the short run."

   10. BHBen Horowitz17:37

       Yeah, yeah. That's, that's tricky 'cause it's so... By the way, I have to deal with this in my, in the firm too. And you know, like, you know, people wanna be entrepreneur-friendly. I'm like, "No, it's not friendly." You know, y- respectful. But you gotta be able to tell them the truth in a way that you probably don't tell most of your friends the truth. Because your friend, y- you know, like, look, anthropologically, we want people to like us, like, and so they, you know, they don't throw us to the lion or whatever. Like, that, that's just kind of a thing. So you say, tell people what they wanna hear. But in dealing, you know, in a, in a company level, in a, you know, in a context of you're on the board of somebody's company, you gotta be able to tell them what they don't wanna hear. That's the most important thing you're gonna say.And yes, they're not gonna like it when you say it. (laughs) There's no question. But over time, like, it could save the company. And, uh, and then, you know, all the most important things I've said are things that I've said to CEOs that they did not want to hear. You know, and then like, that's what leadership is about. It's making... If everybody agrees with the decision, then you didn't add any value, 'cause they would have done that without you. So, the only value you ever add is when you make a decision that most people don't like. Um, and that's where leadership comes in, 'cause you know that's where it's gotta get to. Uh, and, you know, that's, that's the thing that takes practice. Yeah, I think, you know, when Jensen talks about like you gotta get to near death to get yourself to do that, that's true. You know, it's hard, it's hard to build that if everything's going great. And I, and I would tell you, like, the CEOs who kind of had an easy run of it for their... Like, let's say they just launch a product that's an instant hit, it's very hard for them to develop that muscle compared to the ones that build a company like Jensen, where he, like, you know, gutted it out for multiple decades before they had big

4. 19:35 – 22:36

   Who shouldn’t start a company

   1. BHBen Horowitz19:35

      success.

   2. LRLenny Rachitsky19:36

      Clearly, it's very, uh, difficult and painful to be a CEO, to be a founder. Uh, in spite of that, so many people want to start companies. So many people, like, dream of having their own company. What do you... Who, who is not right to start a company? Like, what advice do you share with folks that are thinking about starting a company that may not understand just what they're about to get into?

   3. BHBen Horowitz19:57

      Yeah. So, it, it's funny. So, there's a couple of things. You know, John Reed, who was the CEO of Citigroup when I, when I started as CEO, said to me something I never forget. He said, "Ben, the only reason to start a company is because you have an irrational desire to do so, because it's not worth the money." And, you know, I was like, wow, he does... He didn't even quantify how much money. And this guy's running Citi, so he's a very numbers, banking guy, (laughs) and he didn't quantify it. And, um, I remember when we sold Cloud Cloud for $1.6 billion, I remember thinking, "Wow, that wasn't worth the money."

   4. LRLenny Rachitsky20:35

      (laughs)

   5. BHBen Horowitz20:37

      So, so I think if you're doing it for the money, uh, that's a very bad reason, and it'll be extremely difficult to get to an outcome. You really have to have an irrational desire to do something larger than yourself to kind of improve the world in some way, um, that, you know, somehow like that is your purpose, and if, if you don't feel that, uh, then, then you'll never get through it. It just is too many, too many bad things happen along the way.

   6. LRLenny Rachitsky21:08

      So then how do you think of founders that are kind of looking around for ideas, that kind of brainstorm, that look for, you know, market opportunities versus come from, "I have this mission, I gotta do this thing in the world"?

   7. BHBen Horowitz21:17

      If you have a, you know... And my, uh, business partner, Mark, always talks about this. So, like, if you have a product that forces you to build a company, (laughs) um, that is a great case of it, right? Like, okay, you built something, and the world wants it, and you need a company to deliver it. That's gonna... You know you already have the right product. And so, that's very helpful. I think there are cases of people... I think Hewlett-Packard was kind of built that way, that they were like, "Okay, like, we gotta build technology." You know, it was that abstract. "We've got to build some technology for the world," you know? And then they started with, "Well, like, what do you need?" You know, to... (laughs) They, they called it the next bench thing. What does the, uh, engineer sitting next to me need? The next engineer on the bench was sort of kind of how they defined the first set of products. So, it can work the other way, but like I think the thing that is in common is, like, it's just a very abstract idea about, like, you have to build something that's gonna be important, that, like, is gonna... You know, people are gonna like working there, people are gonna benefit from the products. Like, you have to have some, like, weird concept other than, "Oh, this is gonna be successful, and I'm gonna make a lot of money." I think that's... I think you're way better off taking Zuck's offer at Meta and just-

   8. LRLenny Rachitsky22:33

      Mm-hmm.

   9. BHBen Horowitz22:33

      ... doing that. Like, that's a way

5. 22:36 – 24:54

   The Databricks story: thinking bigger

   1. BHBen Horowitz22:36

      better deal.

   2. LRLenny Rachitsky22:37

      Along these lines, something else Shaka suggested I ask you about. Apparently, there's a story where the CEO of Databricks asked you for $200,000 in the early days, and you said no, and it's not because you didn't wanna, you know, invest, and, and it was more about helping him think bigger. Uh, how did... What happened there?

   3. BHBen Horowitz22:54

      Uh, so there were six, six of them. There were six Ph.D. students, and, uh, uh, well, and Ion Stoica, who was their professor, um. And Ion, uh, was a super genius, uh, but, you know, they, (laughs) you know, when I, when I met with them, they were like, "You know, we need to raise $200,000." Uh, and I knew, you know, like I knew at the time that, uh, what they had was this thing called Spark, and they had... You know, the competitor, uh, was something called Hadoop, and Hadoop, you know, had very well-funded companies already running towards it, and Spark was open source. So, like, the clock was ticking, uh, and, you know, and I think they didn't quite know what they had, (laughs) uh, but I... And then there's also a thing always, although I, I wouldn't say Ion has this mentality, but professors in general, like, it's a pretty big win if you start a company and you make $50 million. Like, you're a hero on campus. Like, that's a, that's a pretty cool thing to have done. And so I'm always a little nervous about a kind of a company that comes out of academia thinking too small anyway. And, uh, so I said, "Look, I'm not gonna write you a check for $200,000. I'll write you a check for $10 million," um, 'cause like this company... You need to build a company. You need to really go for it if you're gonna do this. Otherwise, like, you guys should stay in school. (laughs) And, and they were all graduating right, right then. So, that was kind of that. And Ali-Ali actually was VP of engineering at the time, and it was a while, you know, before, uh, I made him CEO, and that, that was very good luck on my part, 'cause I had no idea that they had a guy that good inside the company who could become CEO when I invested like that. That was just, you know (laughs) , God smiled on me and gave me that one.

6. 24:54 – 28:06

   Managerial leverage and CEO psychology

   1. BHBen Horowitz24:54

   2. LRLenny Rachitsky24:54

      Um, so speaking of Ali, I actually asked him what to ask you about, and he immediately shared this story, I don't know if you remember this, in your first one-on-one with him, after you made him CEO, he was struggling with a bunch of low performers 'cause he was coming in to lead the company, and he was trying to turn things around, trying to coach them, trying to level them up, and your advice to him was, uh, quote, "You don't make people great. You find people that make you great, that make the company great, that you learn from, not the other way around." And it's something that he called managerial leverage. What is that all about? What's the lesson there?

   3. BHBen Horowitz25:27

      Oh, yeah. Yeah, so like, uh, understand, he had just become CEO, so I was teaching him. He had been VP of engineering, and CEO is different, and, and I'll get into why and, and what I mean by leverage. So actually wrote a post about this with a Lil Wayne quote, where I, I think the quote was, "The truth is hard to swallow and hard to say too, but I graduated from that bullshit, now I hate school." And, and that, that, that was always my feeling about this particular idea, was, look, if you're a VP of engineering, you can develop people. You can teach them to be better engineers, you can teach them to be better engineering managers, like that's very, like, doable. But if you're a CEO, like what do you know about being CFO? Like, what do you know about being, you know, VP of HR? What do you know about any of these jobs, you know, except maybe VP of engineering, right? Like, and so the idea that you're gonna take somebody who doesn't... who isn't world-class at marketing and make them world-class, and you don't know anything about marketing, is a dumb idea. It just doesn't work. Um, and then the company can't afford for you to be spending time on that, 'cause they need you to make very high quality, fast decisions. They need you to set the direction for the company, and they need you to have a world-class team. And so like, that... And it's a very hard lesson if you've been (laughs) VP of engineering, 'cause if you're a good VP of engineering, you do develop your people, but as a CEO, like it's not like you don't do any of it, um, but it is very, very small, you know, compared to... So I like to make things just very stark, so you, you get what I'm saying. I don't like to hedge it. And then managerial leverage means this. It is very simple. It's, okay, if I have the ideas about what your department should do next, if I, you know, am kind of pushing you to kind of move your organization forward, then that's no leverage. What's leverage is, if you're telling me what you should do and how you can push the company forward, that's leverage. Then I'm getting kind of more than I'd have if you weren't there. Otherwise, I could just manage a fucking team. And that's the point. When you feel like you're not getting leverage when you gotta go say, "Hey, why aren't we doing this? Why aren't we doing that?" That's when you gotta make the change. And by the way, he's unbelievable at that. Um, as good as anybody I've seen. As, as a guy who's not callous as a CEO, he really cares about the people who work for him, he really wants them to have great careers and all that, but he does not hesitate, like if he's losing leverage, he'll make a move.

7. 28:06 – 31:20

   When founders should be replaced as CEOs

   1. BHBen Horowitz28:06

   2. LRLenny Rachitsky28:06

      Kind of going back to the origin story of a16z, something you guys were really big on was helping founders stay CEOs, become great CEOs, not replace them with professional CEOs. I wanna, I wanna flip this question on you. When does it actually make sense to replace a CEO? When are people not gonna make great CEOs?

   3. BHBen Horowitz28:24

      It really, the, the... There's a very consistent thing that happens, which is, um, you know, when somebody doesn't make it, and it kind of starts with confidence, is the way I would put it. So when... Look, if you, you're, you invent a product, you kind of recruit a team, so forth, all of a sudden you're a CEO. Yeah, but you don't run a big organization, you don't know how to do that. Like most founders are like that. Um, and so if you don't know what you're doing, you're gonna make mistakes, and they all make a lot of mistakes. And then when you make those mistakes, they're very expensive, you know? Like they could cost you to do a down round, or they could cost you to lose a company, or they could cost you a customer, or, or you know, you screw up the product. Like they're very high impact, and not just on you, but everybody who you talked into joining you. And so that kind of motion can really cause you to lose confidence, uh, and then if you lose confidence, what happens is you hesitate on the next decision, and, you know, as we talked about, like hesitation is very dangerous, because one, like it, it locks up the company, but even worse, what happens is if you have senior people working for you, they get very nervous and they feel like they need to jump into that void and make the decision for you, and that's when it gets political, like very political, 'cause people are like vying for power inside your little, you know, screwed up company. And so now you've got a political dysfunctional organization and that, um, y- you know, like that's, that's generally where like, okay, the founder probably can't run this thing anymore, is, you know, that's how it happens, so... You know, most of what we do as a firm is to try to help people with that confidence problem, and there's like a whole series of ideas that we have around that. But that's, you know, you kind of have to somehow climb the confidence and the competency curve together, and it's very hard to do. Uh, and yeah, particularly like if you're an engineer and you're used to getting things right, or if you've been a straight-A student or something like that, it's very disconcerting.... is better to have, like, CEOs who are, like, C-minus students, you know. (laughs)

   4. LRLenny Rachitsky30:45

      Why is that?

   5. BHBen Horowitz30:45

      Get a little facetious there. Well, like, it's just good to be used to failing. So I, I think I, I wrote this, but like the, the, the median on the CEO kind of test is like 18. It's not like 90. (laughs) And so you've got to be comfortable getting a lot of D-minuses, 'cause a D-minus is fine. (laughs) You know, as long as you don't get the F, as long as you don't run out of cash, as long as you don't lose the whole thing, you know, okay. Like, you got through it, keep going, you know? And that's, that's a lot of the, the thing that we try to do with CEOs.

8. 31:20 – 37:57

   Normalizing failure for CEOs

   1. LRLenny Rachitsky31:20

      Yeah, it comes back to the... Your... A core, I don't know, message, through your first book is just how much you'll fail and how much you will struggle and how much pain you'll go through as CEO.

   2. BHBen Horowitz31:29

      Yeah, yeah. And you know, like, I mean, a lot of why I wrote that book was just to normalize that.

   3. LRLenny Rachitsky31:36

      Mm-hmm.

   4. BHBen Horowitz31:36

      I think what happens is, you know, particularly, you know, when I wrote it, and I, I think it's come back and been true now, is like the way a narrative gets written on all these successes is like, oh, they came up with a genius idea and then they built this company and they hired all these smart people and it was all great. But, like, that's not at all how it happens. Um, and you know, I've spent enough time with, like, everybody from, like, Mark Zuckerberg to Sam Altman and so forth that, like, they all go through that same thing that you, who has, you know, your struggling company go through. Like, you screw a lot of things up, and they have massive consequences. Uh, but you, you, you have to kind of maintain your confidence.

   5. LRLenny Rachitsky32:19

      Actually, I was at a storytelling event last night and, uh, we were chat-... I was chatting with someone that I ran into there and told her I was chatting with you today, and she said how meaningful, uh, your first book was to her as a founder, exactly as you said, normalizing that it's very hard and painful and this is just the way it is.

   6. BHBen Horowitz32:35

      Yeah, and the feeling... Look, I mean, you know, if you think about organizational design or, you know, uh, goals and objectives, or OKRs, or what- whatever management technique, like, you need, like, a, you know, a basic, like, eighth grade education to, like, do any of that stuff. It's, it's not that complicated. The difficult part is the, the feeling that you have when you have to do it, uh, is very... Uh, like the hard thing about a, a reorg is you're redistributing power, so you're gonna have people really fricking mad at you 'cause somebody's losing power if you do it correctly. Uh, and that person may be, like, a really good employee. Dealing with that is the hard thing. Like, knowing how the organization should work to make communication better is not that complex.

   7. LRLenny Rachitsky33:18

      Yeah, I think about... I was at Airbnb for a long time, I'm just thinking of Brian who, I don't know, even know if he had a job before Airbnb, you know?

   8. BHBen Horowitz33:24

      Oh, yeah. No, no, I spent a lot of time with Brian and he, uh... After COVID, he... It all kind of clicked for him, um, and then he did that, you know, that, that good talk on Founder Mode and so forth. But the reason that was so articulate is 'cause he had screwed every one of those things up. Um, you know, he hired LT and, you know, all this stuff and, you know, these are very senior people and, you know, he didn't... He wanted to defer to them, but you can't defer as the CEO, you know, 'cause you know what Airbnb should be doing. He may know what fucking finance should do, but you know what Airbnb should do and this kind of thing. And, and then, then it gets really wild, uh, when you... Like, you can't defer decisions as the CEO, you gotta, like, understand what people are saying and go, "No, we're gonna do this."

   9. LRLenny Rachitsky34:12

      And this again comes back to the point of you have to go through the struggle and pain and failure to learn those lessons.

   10. BHBen Horowitz34:18

       Yeah, no, like, I mean, it's... They're really hard to learn without kind of doing and without... Often, like, without paying the consequence. And I... You know, like, even I... Like, I make mistakes. I, I was having a conversation with Ali the other day and I was like... He's like, "How's the going been?" And I was like, "Well, you know, like, I'm finally dealing with something that I had put off for, you know, a very long time." (laughs) And he said, "Why'd you put it off?" I said, "'Cause things were going too good." (laughs) I didn't have to deal with it. And he's like, "Yeah." (laughs) Yeah, he said, "I know that." Like, I... And I'd say Ali is, you know, one of the, if not the kind of best private company CEO out there, and he's making mistake and I'm making mistakes so, like, it, it's just tough.

   11. LRLenny Rachitsky35:01

       You said that one of the... Maybe the main reason founders fail as CEOs is they lose confidence, and you had some ideas that you guys have to help founders work through that. Are there a couple you can share how you help?

   12. BHBen Horowitz35:11

       Yeah, yeah. So we do, we do a lot of things on that. So the, kind of, design of the firm is about confidence. So the first thing is, well, what would give you con-... Well, like, if you can get stuff done. So what if I could give you a network that's as good as Bob Iger's network day one, like the day you step into the job? And so, you know, we have 600 people at the firm, and why is that? Well, most of them are building that network for you so you can call any CEO or anybody in Washington or, you know, kind of any executive or that kind of thing, and get them on the phone and they'll talk to you and you can kind of deal with that thing. Uh, and then that just makes you feel like a CEO. And then, you know, we have a lot of people in the firm, like myself, who you can talk to on, like, a CEO-to-CEO basis as opposed to an investor to CEO and just kind of feel that... Early in the firm's days we used to do this thing, I think I'm gonna bring back in some form, this thing called the CEO barbecue. And it was like, you know, a lot of people have these events where, like, you know, they bring in speakers and this and that and the other, and I always felt like those were... One, they were too many days, and then sometimes, you know, the... What they said wasn't really applicable and that kind of thing. And so what I said, "Well, why don't we just have a barbecue?" And, like, I would barbecue. We get everybody in my backyard, um, there was 500 people at the peak, which is why it had to, kind of, stop (laughs) 'cause I couldn't cook that much food around 500 people.

   13. LRLenny Rachitsky36:46

       (laughs)

   14. BHBen Horowitz36:46

       And then, uh, you know, we'd have Larry Page and Mark Zuckerberg and Kanye West. And so you're a CEO in there

   15. LRLenny Rachitsky36:54

       Yeah.

   16. BHBen Horowitz36:54

       ... before f- ... You're like, "Wow," like, "I must be important. I'm here with all these guys, and we're just hanging out, having a drink, eating barbecue." Um, and so then when I go back to my company, like, I feel like I am somebody and, like, okay, I might not be, like, perfect at all this, but I am really a CEO. I was at the CEO barbecue, for crying out loud, and that kind of thing. And so, y- yeah, th- that's ... But it's all ... (laughs) The whole idea was always like, okay, do you feel like you can do it? 'Cause that's, yeah, that's half the battle. Uh, and look, having been in ... And, you know what? Every CEO has been in a position where they feel like, "Well, maybe I shouldn't be the one running this thing. Maybe it's just too big for me." Uh, and that's a bad ... You don't, you don't wanna go there. Uh, and because, as, as we've said, founders can get to the next product.

   17. LRLenny Rachitsky37:48

       (laughs)

   18. BHBen Horowitz37:49

       And that's something that, uh, you know, f- almost no professional CEO is able to do. Like, there have been rare cases, but very rare.

9. 37:57 – 42:31

   Counterintuitive lessons about building companies

   1. BHBen Horowitz37:57

   2. LRLenny Rachitsky37:57

      So clearly you've worked with a lot of companies, a lot of founders. Let me kind of zoom out a little bit and ask you this question. What's the most counterintuitive lesson you've learned about building companies that goes against common startup wisdom?

   3. BHBen Horowitz38:09

      Well, you know, the common startup wisdom keeps changing. (laughs)

   4. LRLenny Rachitsky38:13

      (laughs)

   5. BHBen Horowitz38:14

      You know? Uh, like, one of the early ones that, um, you know, was wrong, and kind of Brian articulated it, and then now I think a little bit of what people have gone to is also wrong, so ... The, the first idea that was wrong was, like, okay, build, uh, a team of, like, senior executives, you know, as soon as you get product-market fit, as fast as possible, and they can scale the thing. And I think that you got to build that team slowly and deliberately, kind of paced to your ability to ma- integrate and then manage them. Because if you bring in a bunch of senior people and you don't know, really, how they map to your company or how that function works or so forth, then you're gonna start deferring. And once you start deferring, it's gonna get out of control very fast, 'cause they're gonna build empires, they're gonna get political, they're gonna do all that kind of thing. So that was bad advice. You kind of have to do it in a measured way. I think that founder mode, I think a lot of people have taken to never hire anybody with experience. A- and that's also bad advice, uh, in that, look, somebody who knows how to do something can really accelerate your thing. So very early on, um, one of the founders, great founder, uh, Oursalan, at Databricks was running sales. And I'm like, "Ali, like, you're gonna have to hire, like, somebody who knows sales," um, because Oursalan's the PhD in computer science. Like, I like that, but, um, that, that's probably not where you're gonna have to start if you're gonna catch these guys before they take spark and, like, use it against you. And, you know, and, and I sat down with Oursalan and explained why. I said, "Look, you know, a lot of what s- sales, there's a lot of knowledge in how to build a worldwide sales organization." I- you need knowledge of customers, territories, t- territory splitting, rep profiles. Like, they- there's just, like, a litany of stuff that you really can only learn by doing, trial and error, and you don't know any.

   6. LRLenny Rachitsky40:17

      (laughs)

   7. BHBen Horowitz40:17

      And so, like, "You're phenomenal." Like, "Let's get you ..." You know, and he's still, he's a very senior executive in the company now. But we need somebody who knows that. And the idea that there are companies that go, "Okay, we're just not gonna hire that 'cause we're in founder mode," that's also a mistake. So the- th- there's a lot of ... It's more subtle than you think, uh, and it's more complex than you think. And so you kind of have to get all the way to the truth. And these little snippets of advice that VCs give 'cause they watch some fucking podcast are all fucking stupid. Like, it's just, y- there's a lot of depth to these things. You have to know the answer to the next question, the next question, and the next question. And it does drive me crazy. Like, one of the funnier things that happened, um, along these lines, just to show you how little you know as an investor about what it means to be CEO, uh, one of the ... We were at a board dinner. One of the CEOs, uh, says to me, he goes ... Or one of my, you know, one of our CEOs says, "Hey," y- you know, "Ben, like, that thing you told me a while ago about don't be CEO at home?" He said, like, "I, I, I yo- I was doing that, and I stopped, and it really helped me." And then the other kind of VC said, "Yeah, you know, you got to unplug sometime." And I said, I was like, "What the fuck are you talking about? He's CEO. He's not unplugging." Like, "He's getting shit all the fucking time." Like, "He's got to deal with that." Like, that was not what I meant. I was like, "You can't go home and boss your family around." That's what I meant.

   8. LRLenny Rachitsky41:44

      (laughs)

   9. BHBen Horowitz41:44

      You know, like, so, so it's ... The, the ... You hear something from, like, somebody who ... But if you haven't done it, you don't even know what that means. And so then you kind of then trying to transfer the advice to the next guy, that's not gonna work. So anyway, so but it's, you know, like, he, he was very innocent. I

   10. NANarrator42:01

       Yeah.

   11. BHBen Horowitz42:02

       I- I just don't want to kind of, uh, speak bad of him, but, like, it ... That's how it sounds, right? But that's not what it is.

   12. LRLenny Rachitsky42:09

       That's an amazing story.

   13. BHBen Horowitz42:10

       Yeah.

   14. LRLenny Rachitsky42:11

       (laughs) So the advice, (laughs) uh, partly here is just don't believe everything you s- you see on Twitter in little sound bites of advice.

   15. BHBen Horowitz42:16

       I- yeah, I mean, like, I- a- and I think actual CEOs know it. And that's kind of-

   16. LRLenny Rachitsky42:20

       Mm-hmm.

   17. BHBen Horowitz42:20

       ... how, like, uh, people in my profession kind of get a bad rep because, like, giving advice for ... th- that's not something that you know but something that you heard is very dangerous, I think.

10. 42:31 – 48:21

    “Good Product Manager/Bad Product Manager”

    1. BHBen Horowitz42:31

    2. LRLenny Rachitsky42:31

       So speaking of advice that, uh, you've shared that might be out of date now, you are famous for writing, uh, one of the most popular pieces of literature for product managers. There's a lot of p- PMs that listen to this podcast called Good Product Manager/Bad Product Manager. And if you actually go to that post today, at the top, you say, "This document was written 15 years ago, and it's probably not relevant today for PMs. I present this merely as an example of a useful training document."Still, people link to it. I actually just link to it as an ex- as just like this is something every PM needs to read. What is that you think people should maybe not take away from it, and what do you think people still should take away from that piece?

    3. BHBen Horowitz43:07

       Yeah, so the reason I wrote it when I wrote it was, um, that I had a lot of product managers. And one thing about product management is, um, it's a job that's completely different at every company, bl- like, and there is no training for it. So, like, everybody kind of figures it out ho- as they go and depending on, like, what's being emphasized, like, they'll get wrapped around the axle on, you know, like, if it's an enterprise company, okay, well, like, you know, pitching to customers or, like, I need to be, like, really good with the press, or I need to be really good at writing, you know, the product requirements document or that kind of thing. And none of th- those are all, like, these tasks, um, but none of those were the job. Uh, and what I was trying to get out in Good Product Manager, Bad Product Manager was the job is fundamentally a leadership job, and it's a tricky leadership job because nobody is actually reporting to you. So it's like this influence, how do I get people to do what I want, you know, even though I'm not paying them, I can't fire them, I can't promote them and so forth. And whi- which is kind of the essence of, like, real leadership because uh, i- if you start to rely on promotion and firing and so forth for authority, then you, you're never gonna be good at being CEO or anything. So I wanted them to get into the mindset of, okay, your actual job is to get a product into market that customers love that's better than anything that anybody else in the world puts in market. Like, that's your job. And so to accomplish that job, you need engineering to understand you with clarity, you need to understand engineering with clarity, you need to have a really good view of the market and the competitors and the technology and so forth, and you need to kind of put that all together and deliver the thing. And all the other things-

    4. LRLenny Rachitsky45:19

       (laughs)

    5. BHBen Horowitz45:19

       ... are tasks that may or m- you may or may not need to do. I don't know if you need to do them. But, like, the thing is, you know, you have to be the leader. You've got to get the thing done and, and so what I think it's still good on is that, like, the, the mindset, (laughs) like, you know, be the leader, uh, I think the details of it, um, you know, of any kind of thing that was kind of task-specific was, like, really for my group at Netscape in, like, 1996-

    6. LRLenny Rachitsky45:54

       (laughs)

    7. BHBen Horowitz45:54

       ... or whenever the hell I wrote it. So it was y- you know, it was a kind of document I wrote out of frustration. Uh, but I- I'm- I am glad that th- that people still like it, uh, and I think leadership in general is under- undervalued, underestimated. Um, it's a- it's the most powerful thing, uh, and, and most of the great, you know, great companies... Jensen is a great example. Like, what a phenomenal leader he is, not just of NVIDIA, but of the whole industry. And he doesn't have authority over the industry, but, like, he drives it forward and that's, uh, and, and that's why the Good Product Manager, Bad Product Manager is so important, 'cause that thing, if you learn how to do it, that's the thing.

    8. LRLenny Rachitsky46:39

       Uh, I didn't realize you wrote that i- initially as just an internal document, and then you made it public.

    9. BHBen Horowitz46:42

       Yeah. (laughs) Uh, it was.

    10. LRLenny Rachitsky46:44

        Mm-hmm.

    11. BHBen Horowitz46:44

        It was kind of before, um, blogging took off-

    12. LRLenny Rachitsky46:47

        Mm-hmm.

    13. BHBen Horowitz46:47

        ... so it was just an internal thing, uh, and I published it later. It was, uh, I was just getting so mad, you know, uh, and by the way, my product management team at the time was, like, very good, (laughs) uh, very talented people. They just didn't th- they just were not getting that concept. So, you know, like David Weiden's at Coastal Ventures, Raghu Raghuram who was, went on to be CEO of VMware, I mean, like, the team was, like, that team. Uh, but they were driving me crazy. W- and so I, I just, I was like, "I can't yell at people anymore. I have to, like, explain to myself." (laughs) And so it's a good thing. If you find yourself yelling at people, like, you probably haven't explained what you want. Is th- was the other big takeaway from that.

    14. LRLenny Rachitsky47:30

        Did you ever think that piece would be so long-lasting and so, I don't know, popular?

    15. BHBen Horowitz47:33

        No, th- oh, I thought, I, I didn't even know, like, I thought it was kind of, like, aggressive when I wrote it. Like, you could tell I was mad 'cause I called it Good Product Manager, Bad Product Manager. It's like bad dog, (laughs) you know? Bad prod- bad, bad product manager, that's th- that w- that was kind of the emotion I had, so. Uh, you know, it's, it's kind of shocking, yeah, some of the things that you write. Uh, I, I would say though that's, you know, and kind of creative, and you probably know this, like, the ideas that you have, um, the things that you write in five minutes end up being much better than the things you write in five weeks, you know? And, uh, I, I find in talking to, you know, uh, musicians or writers or anybo- everybody has that same experience. Like, the thing that you've already synthesized so much that you just have to write it out is, uh, that's the best stuff.

11. 48:21 – 51:16

    Product managers as leaders

    1. BHBen Horowitz48:21

    2. LRLenny Rachitsky48:21

       There's something that you mentioned there in your answer about the PM being the leader. There's always this kind of sense that the PM is not the mini CEO, how dare you call yourself that. I actually think that's exactly what the PM is. They're basically the closest to the CEO. Their kind of job is to think like the CEO within the team.

    3. BHBen Horowitz48:38

       Mm, yeah. People get mad, you know, like, uh, y- because e- e- everybody, you know, like, this is the whole challenge of management in general. Like, people get jealous over stupid shit. But, like, from, from the perspective of the PM, like, it doesn't matter if you write a good spec or you have a good interview or you do this or do that. Like, what matters is at the product launch, and you have to get all the way to there.... and kind of work backwards from that. And you can't do that without leadership because it is about, okay, we want to build a... And, and you're not necessarily the person who comes up with every idea or this or that. I mean, you're just the keeper of the vision. You know, and that's true for CEOs too. Like, you don't want every idea in a company coming from the CEO. Like that's... I think it's a misunderstanding of what a CEO is, is why people don't like that. They don't know what a CEO is. But a CEO isn't the one who has every idea, gives every order, does every... That's not the way it works. The way it works is there's somebody who's got to consolidate, you know, get all the good ideas, prioritize them, decide which good ideas we're gonna do, and then get everybody on the same page that they have very high fidelity and understanding of what that is. And, you know, so that it is a CEO kind of function. Now, it doesn't mean like I'm better than you, it just means that like that's what I'm doing.

    4. LRLenny Rachitsky50:05

       This episode is brought to you by Miro. Every day new headlines are scaring us about all the ways that AI is coming for our jobs, creating a lot of anxiety and fear. But a recent survey from Miro tells a different story. 76% of people believe that AI can benefit their role. But over 50% of people struggle to know when to use it. Enter Miro's Innovation Workspace, an intelligent platform that brings people and AI together in a shared space to get great work done. Miro has been empowering teams to transform bold ideas into the next big thing for over a decade. Today, they're at the forefront of bringing products to market even faster by unleashing the combined power of AI and human potential. Guests of this podcast often share Miro templates. I use it all the time to brainstorm ideas with my team. Teams especially can work with Miro AI to turn unstructured data like sticky notes or screenshots into usable diagrams, product briefs, data tables and prototypes in minutes. You don't have to be an AI master or to toggle yet another tool. The work you're already doing in Miro's canvas is the prompt. Help your teams get great work done with Miro. Check it out at miro.com/lenny. That's miro.com/lenny.

12. 51:16 – 56:23

    Why a16z invested in Adam Neumann after WeWork

    1. LRLenny Rachitsky51:16

       Let's talk about AI. (laughs) I'm very proud of us. It's been almost an hour. We haven't even mentioned AI.

    2. BHBen Horowitz51:21

       Yeah. (laughs)

    3. LRLenny Rachitsky51:21

       I think that's a record. (laughs) Uh, okay. So I asked Adam Neumann, uh, WeWork founder, now Flow founder, someone you work closely with now, what to ask you about, and he said he has some really interesting insights about how AI is impacting hiring.

    4. BHBen Horowitz51:35

       You know, Adam is probably the single most controversial investment that we ever made. Like, we got called everything from stupid to sexist to racist to this and that for, like, like, literally just funding Adam.

    5. LRLenny Rachitsky51:50

       (laughs)

    6. BHBen Horowitz51:51

       Um, and I think it's gonna end up being one of the best investments we ever made. Um, he's doing a phenomenal job there. But it's, there's an important principle in, in that which kind of we do as a firm which I think is, um, not widely done but I would love it if people copied it, which is... And I, it's something I learned, um, you know, somewhat from Shakha, which is you, you don't judge a person by the worst thing that ever happened to them. Like, we've all had bad things happen to us. We've all made bad decisions. Most of them, they don't make a miniseries about, right?

    7. LRLenny Rachitsky52:34

       (laughs)

    8. BHBen Horowitz52:35

       And so, like, to judge them on that, you wanna judge people on what they do well, not what they screwed up. And, you know, 'cause that's where you see the talent. I mean, if you look at what Adam did well, it's truly spectacular. Uh, you know, like WeWork, everybody knows what WeWork is. Like it's the... Name a more important commercial real estate brand than WeWork. Like, you can't, uh, and so, like, what an accomplishment. And then, you know, and there were so many things that went into that and, and yeah, so many things he did right. And then if you kind of look at, really unravel the things that went wrong, most of it was like a combination of inexperience and nobody around him that would tell him the truth. And so like that... And, you know, maybe he wasn't good at listening to the truth either at the time. Um, but to throw away a guy on that, which is, by the way, the world was so mad at us for not throwing him away, for, for believing in him, um, is just a, it's a big mistake and... Uh, you know, I credit Mark, 'cause Mark is the one who called him up originally and just said, "Hey, Adam, what are you doing?"

    9. LRLenny Rachitsky53:49

       (laughs)

    10. BHBen Horowitz53:49

        Uh, 'cause like, you know, we watched what you did at WeWork and we thought it was pretty impressive. And so I think that, uh, you know, that, that's probably the biggest secret there. It's not really about AI, but I think that, um, like looking... You know, Judge, uh, Al Davis once said, you know, "Coach players on what they can do," and I think that's very true. You know, judge people on what they can do. Um, you know, coach people on what they can do. Like help, take their strengths and use them as opposed to over-focus on their weaknesses and just, you know, handwring about the one fucking thing they don't know how to do. Um, because, look, everybody's uneven.

    11. LRLenny Rachitsky54:31

        Well, it sounds like what you're describing essentially is that the job of an investor is to find an underappreciated asset and invest, or, before, something people don't see.

    12. BHBen Horowitz54:39

        Yeah, and I mean, I think that it's... You know, like, venture capital is really about investing in people, right? Like, you know that you have ideas as an investor, but, like, what you really are ultimately betting on is the entrepreneur and the entrepreneur's idea because the initial idea isn't where they end up usually. It, it changes a lot, um, with, with everybody we invest in.You know, you kind of have to make the judgment on the person. And, you know, how you do that is, uh, is really, really important. One of the things we emphasize inside the firm is like, like we're investing in strength, not lack of weakness. Like, I want to know, like, how good, like, are they world class? Do they have a world-class strength? And, you know, can that beat anybody? And look, they, everybody's flawed. And so like let's help them deal with the flaws-

    13. LRLenny Rachitsky55:34

        Yeah.

    14. BHBen Horowitz55:34

        ... and, you know, surround them with people who can handle that, and put the right person on the board who can talk to them. Like, I go to all Adam's board meeting, even though Mark's on the board, I go to all his board meetings because I'm the one who's good at, like, killing a guy who's that confident, um, when he's like, "Okay, that's not your best idea." Like, that's a good role for me. Uh, but that's how you deal with that. You don't, you know, throw them away and go, "Oh, okay, like, we don't want to be called names, so we're not going to invest in Adam Newman after he built WeWork." That's crazy.

    15. LRLenny Rachitsky56:08

        That's also why you guys invested in Cluey, I imagine. Similar controversy.

    16. BHBen Horowitz56:11

        Yeah, yeah, yeah. Yeah, yeah, no, that's right. Like, I mean, like if you look at what those guys did, that was, like, some, like, high-level marketing genius too. Uh, and, you know, that, that's really something. Plus the product is awesome.

13. 56:23 – 1:02:43

    Is AI in a bubble?

    1. BHBen Horowitz56:23

    2. LRLenny Rachitsky56:23

       I'm gonna bring us back to AI. Something that a lot of people are talking about right now while we're recording this is this potential huge bubble we're in with AI. Sam Altman said, "We're in a big bubble," which is, you know, that's saying a lot. I'm curious just how you think that-

    3. BHBen Horowitz56:35

       So who owns it? What, what is it saying? So you got to, like... First of all, I should qualify this by saying I'm an investor, um, and Sam's a CEO. So CEOs have to have much more purpose when they talk. Investors just have to be entertained.

    4. LRLenny Rachitsky56:51

       (laughs)

    5. BHBen Horowitz56:53

       Um, so you gotta give Sam credit for, like, what is in his interest to say? Well, like if it's a bubble, then the one thing you should invest in is him.

    6. LRLenny Rachitsky57:04

       (laughs)

    7. BHBen Horowitz57:05

       And not all these guys chasing after him. Um, so that I, I, I would say, like, that's very smart. And then the other thing that's smart about it is there's nothing that you can say to the press that will make them love you more than saying, "All investors and, like, entrepreneurs chasing this are idiots." Like, they love that, 'cause, you know, the, the press are generally haters. And so it's just like red meat for the haters, which is also super clever. So I think whether even he believes that or not, that was a super smart thing to say.

    8. LRLenny Rachitsky57:37

       (laughs)

    9. BHBen Horowitz57:37

       (laughs) So I'll just put it there. Whereas what I'm gonna say, um, won't be as smart, but it will, uh... I, I, I don't really have a- an ax to grind here. I mean, I could have an ax to grind and say, "Okay, like, let's get all the other investors out. I'll say it's a massive bubble." Um...

    10. LRLenny Rachitsky57:54

        (laughs)

    11. BHBen Horowitz57:56

        But what I would say about that is, um... So the first thing, the one thing about bubbles is anytime everybody thinks it's a bubble, it's not a bubble. Because in order for it to bubble, you need capitulation, in that you need kind of everybody to believe it's not a bubble, 'cause then the prices really go out of control. But as long as there's people who think it's a bubble, then it's hard for that to happen. And, uh, it's funny, I had this debate, um, in I- in The Economist, I think with Steve Blank in, like, 2000 and 11 or '12 when everybody thought it was a tech bubble, if you can imagine that, which it absolutely was not.

    12. LRLenny Rachitsky58:42

        (laughs)

    13. BHBen Horowitz58:42

        But because there were like 1,400 articles saying that we were in a tech bubble, and, you know, I mean, you know where prices were then com- compared to where they are now. Um, but I knew, because everybody was saying it was a bubble, it wasn't a bubble. (laughs)

    14. LRLenny Rachitsky58:55

        Mm-hmm.

    15. BHBen Horowitz58:55

        Like, I knew that. The prices were higher, but the reason the prices were higher was we were getting to a global market. AI prices are higher than prior prices, but if you look at the revenue growth and numbers, we've not seen anything like it. The products are work- ChatGP- Sam's product works so amazingly. We'd never seen that before. Like, n- not even Google, not, not anybody. And so, like, that's real. And, you know, we have companies that, you know, went from zero to 800 million in a year and that kind of thing. And so it's not a... I, I would say there's a basis for the prices going up, first of all. Now, will the, the... I think the thing that's right about kind of what Sam is saying is the landscape is early, really early. The technology is very immature. As amazingly as it works, you know, there's a long way to go to improve it. So it's very possible, you know, when you have that much technological change that the positions that these companies have achieved with their high revenue isn't sustainable, in that that, you know, there will be a competitive, uh, change that either kind of lowers prices or a new number one emerges or that kind of thing. And so they... Yeah, that's possible, but I, I, I wouldn't characterize that as being like a financial bubble, uh, in that... Like, like if you go back to the great dot-com bubble that everybody is always waiting for it to happen again, which I was CEO during, the thing that happened there was very different, which is it was the internet, and every smart investor knew that the internet was a big deal. Like, how could you not fucking know that the internet was a... Of course it's a big deal. But, like, if you go back to 1996, at Netscape we had 90% browser share, and we had 50 million users. So there were 55 million people on the internet in total, and half those were on dial-up. So, and then to build a product like Evite, the greeting card company, had 300 engineers. That's how hard it was to build this stuff. And so the math didn't work, and the math didn't work on any of those ideas, and but the investors kept pouring money in, and then eventually, you know, um, everybody went bankrupt because there was no revenue coming in. And, like, when they figured that out, then nobody would invest in anything. And of course, then everybody realized, well, the internet was actually real and Paul Krugman didn't know what he's talking about, and, like, it was gonna be a big thing. And then Facebook and Google and all these things emerged. But the thing that made it a bubble was, like, the unit economics didn't work, the businesses didn't work. Like, these businesses are all working and they're being priced appropriately for how they're growing. So that's not an effect. So the, the thing that you could say is, they're not gonna keep growing like that, or, like, and so forth, and I'm not sure about that. Like, like, the products, like I said, are working so much better than any technology product that we've ever built has worked. Like, it's just mind-blowing how good this stuff is. And so I don't know. I- I... If I had to bet, I would bet not a bubble. Uh, you know, there... I think there'll be some dislocation. I think companies that... I- I think a- always in venture capital, if you've got, like, a run like this, then the great company and the crap company both get funded, but, you know, that's just, th- the... That's just venture capital. That's

14. 1:02:43 – 1:12:51

    The biggest opportunities in AI

    1. BHBen Horowitz1:02:43

       not a bubble.

    2. LRLenny Rachitsky1:02:44

       For founders starting companies these days, when you look into the future of the AI industry, say, in five, 10 years, how do you... How do you think things will play out/where do you think the biggest opportunities remain? Where are you guys looking to invest most?

    3. BHBen Horowitz1:02:57

       In infrastructure, um, I- I think that, you know, like, there is obviously, like, a real estate power cooling play. Um-

    4. LRLenny Rachitsky1:03:06

       Hmm.

    5. BHBen Horowitz1:03:06

       Yeah. I- I think that's a little outside of, you know, kind of hardcore technology investing that we do. But there's another layer which is, like, who can run... You know, take a given open source model, who can run it the cheapest, um, with the, kind of, lowest latency? And that's gonna be extremely valuable, like, whoever has that. Uh, and, you know, Google has been historically very good at that and so forth. Um, and I... Sam is really trying to build that, um, now with Stargate, and so I think that's gonna be a very important layer of value. I think that, you know, on the foundation model side, you'd have to be very selective as an investor. So in order to compete in foundational m- model world, our basic rule of thumb is, you have to be able to, you know, without much product pro- progress, raise at least two billion dollars, 'cause that's basically what it's gonna cost you to train something that gets you competitive enough to make money. Um, and there are just very few founders like that. Uh, so, Ilya (laughs) is one of those. You know, MIRA is one of those. Um, Fei-Fei is one of those. But, like, that's the kind of class of person you need, and there's, you know, whatever, there's certainly less than 10 of those in the world. And so that- that- that's kind of, like, an important area, but a small area, uh, and then I think the application layer is gonna be very, very interesting. And I think that... You know, if you look at Sam, like, he's making most of his money off ChatGPT, almost all his money off ChatGPT now. And ChatGPT is like... You know, I... It looks like it's got a real... Like it or not, it's got a real moat. It's very hard to knock it off its perch. There's a lot... You know, everybody's taking a shot at it. People with great distribution like Google and, you know, Elon and, and Zuckerberg and everybody, and, like, the thing just keeps going like it is. So I think the applications are both more complex and, uh, kind of stickier than people thought they were originally. Like, so the, the thing that people got very wrong is this whole thin wrapper around GPT. Like, that's really wrong. In fact, here's how wrong it is. Back in the '80s, they- that same phrase was used, but it was thin wrapper around an RDBMS, uh, which-

    6. LRLenny Rachitsky1:05:38

       A database.

    7. BHBen Horowitz1:05:39

       Yeah, yeah, yeah. So it meant companies like Salesforce were basically just a thin wrapper, and I think that that's kind of a mistake people made. So, like, we're in this company, Coursera, and if you look under the covers in Coursera, they've built, like, 14 different models to really understand, um, how a developer works, like a high-end... Like a, a, a real developer. And they have... That... Those models have tons and tons of interactions with how people talk to their friend, Coursera, about how they should design their program and so forth, and that's, like, real. That's not, that's not just a thin layer on a foundation model, and I think there are many, many applications like that. Uh, and, and- and so I- I think there's gonna be a lot of opportunity at the application layer. There's gonna be some opportunity at the foundation model layer. Of course, you can invest in Sam, you can invest in Anthropic and so forth and as well, um, but there will probably be, like, a very small number of companies at that set, and then a almost unlimited number of companies at the application layer. And then, you know, as the technology advances, we'll, of course, see more things with embodied AI. I mean, already, you know, autonomous cars are working, like, really well now, um, after a long, (laughs) a long, long, long time since I g- I think Sebastian won the, uh, the challenge in 2006 when he drove the self-driving car across the country.Um, and we're- here we are 20 years later and now they're deployed. So, that was a long time. Uh, robots, I think is a harder problem, and self-dr- driving cars. So, we'll see how that goes. Um, but, uh, you know, there- there's certainly a lot in that world as well.

    8. LRLenny Rachitsky1:07:33

       Wow. Okay. There's a lot to this answer.

    9. BHBen Horowitz1:07:36

       Yeah. S- It is.

    10. LRLenny Rachitsky1:07:36

        Something that... No, it was ex- exactly what I was looking for. The cursor example, it's something that comes up a lot on this podcast, in the application layer specifically.

    11. BHBen Horowitz1:07:45

        Mm-hmm.

    12. LRLenny Rachitsky1:07:45

        The, uh, thought that the way to win in this space and to build a moat is, as you said, build your own model/have proprietary data that you build through people using your product. Thoughts on that?

    13. BHBen Horowitz1:07:58

        Yeah. I mean, I think that's- ends up just being what's required. So, it turns out that, um, the universe is long-tailed, is- who- is- is fat tailed, and humans are very fat tailed in terms of human behavior, human conversation and so forth. So, to get to the real meaning of it and to get to the kind of essence of the problem, you know, in any domain, uh, turns out to be, I think, more complex than we thought. And so like, like the early things and, you know, people were running around, uh, saying, "Okay, there's gonna be one big brain to rule them all," and these kinds of things. Uh, that's kinda not played out yet. Uh, and, and in fact like if you look underneath the covers, you have LLMs which have generalized pretty, you know, like in fascinating ways, but they've kind of also asymptoted in that, um, you know, we, we have run out of data for the most part. And so if you look at the GPT-5 LLM compared to the GPT-4 one and how much more it costs to train and so forth, like it's th- you know, i- it's definitely not going linear anymore. On the other hand, the reinforcement learning side has been linear, but it doesn't generalize. So, if you build a great, uh, programming model, it may be an idiot at math.

    14. LRLenny Rachitsky1:09:24

        Mm.

    15. BHBen Horowitz1:09:24

        And so that I think is just very different than what people would have said three years ago. And I think that's kind of, you know, that there's not something that's both scaling and generalizing yet. And, you know, maybe we'll get there. But that certainly opens the door to something that's more user-friendly, that's more effective in any number of domains than just the basic foundation model infrastructure. Now, those models are incredibly important and they're... And I think OpenAI is probably 80% of the revenue in AI or something like that now, like it's-

    16. LRLenny Rachitsky1:10:03

        Mm. Wow.

    17. BHBen Horowitz1:10:03

        ... massive. And so that foundation model is really, really important. Um, and then the, like the, the, the basic consumer app is really, really important, um, that just answers whatever the hell you wanna know.

    18. LRLenny Rachitsky1:10:14

        (laughs)

    19. BHBen Horowitz1:10:15

        Like tho- those things are like very, very real, but I do think, you know, particularly... And then if you get into like enterprise stuff, and then it's no longer internet data, it's their data, that becomes very different and like Databricks is having a lot of success there because... Okay, well once you're inside a company, guess what? Like, you care about access control.

    20. LRLenny Rachitsky1:10:39

        (laughs)

    21. BHBen Horowitz1:10:40

        Um, that's hard with an AI world, you know?

    22. LRLenny Rachitsky1:10:44

        (laughs)

    23. BHBen Horowitz1:10:44

        It gets trained on some stuff. How does it know who has access to that information, who doesn't and so forth. Um, you have semantic issues. So, if you look at an enterprise, f- find 10 enterprises, they all have a different definition of what a customer means. Like you would think customer is a basic thing. Well, is it a department at AT&T? Is it AT&T? Is it like a person at AT&T? Is like... What the hell is a customer? Uh, turns out to be very ve- very meaningful, uh, particularly if you're trying to, um, figure out like important things like churn and this and that and third. So like, that kind of stuff matters. So it- so it- I- I would just say like the- the problem space is m- a lot bigger than you can just attack with a basic foundation model, um, at, you know, currently. You know, maybe that will change. And like if that changes, then certain prices will have, in retrospect, looked way inflated and others will look, you know, too low. But, y- you know, that- that is TBD.

    24. LRLenny Rachitsky1:11:48

        So, a big takeaway from this is that there's still tons of opportunity for founders to start companies building AI products.

    25. BHBen Horowitz1:11:55

        I think so. I mean, you can solve so many... Everything that we couldn't solve with software we can solve now almost, so it's a really big world, uh, (laughs) and it's funny, you know, like, 'cause we're investors in- in Waymo and, uh, one of the things when you get into like what took so long to make Waymos so safe like they are now, it wasn't the things that everybody reported on the podcast, it wasn't sleet and, you know, heavy rain and... It was people. It was like the human who was driving 75 in the 25 zone. It was very hard for the AI to anticipate 'cause it was rare but important, um, and the number of rare important crazy shit that humans do is very high. Uh, and I think that- that goes for all of AI. Um, so to make things work really well, you have to understand this- this very kind of fat tail of human

15. 1:12:51 – 1:18:53

    Why U.S. leadership in AI matters

    1. BHBen Horowitz1:12:51

       behavior.

    2. LRLenny Rachitsky1:12:52

       Along this AI thread, something that is really important to you clearly, something you talk a lot about is- is the US being successful in AI and be- and leading the world in AI. Why is it so important? Why is it something you spend a lot of time on?

    3. BHBen Horowitz1:13:06

       It- it starts with, uh, I think my view of the US, uh, and its role in the world. So, (sighs) it's... My personal view is like it- it's very, very, very important, uh-... m- not for society to be completely fair, 'cause it's not gonna be completely fair or completely equal, 'cause it- we've never had one that's been completely equal. But it's important that everybody have a chance, um, at life. You know, and particularly, uh, y- you know, kind of both culturally, but also just, like, y- you can't advance the world if, you know, you can't tap into all of your resources. And so if you kind of take away motivation and- and these kinds of things, then y- you get into trouble. And if you look at the, kind of, every country today, a- a- and th- th- this is by the way, um ... So you want, like, the right amount of decentralized power. Um, you know, you don't want it to be completely concentrated. Concentrated power makes it very, very difficult for everybody to have a chance. This is the big, kind of lesson of communism over the last 100 years is, it turned out, right ... And, and, you know, we still have politicians selling it this way today. It's like, "Oh, it's power to the people." No, no, no. It's power to you because you're removing all power from the private sector and installing it into the government, and then you're putting yourself in charge of the government. And so I become extremely powerful. And this is why it didn't matter if it was Mao or Pol Pot or Ceausescu or Stalin, everybody died. Because when you give anybody that much power, nobody has a chance. There is no incentive. Uh, there is no carrot, there's only stick, and so you use that stick. And that's just the nature. It's a systems problem, it's not a person problem. It's not Stalin was evil, Ceausescu was evil. It was like, that system is evil. Uh, and that's e- ch- same with fascism, right? Be it Hitler or Mussolini, it doesn't matter. Like, that level of power is evil. Um, and the US does the best job systematically. It's the best system. It's got all kinds of issues, it's got problems, people always try to defeat it. But, you know, one of the things that, um, y- you know, if you look at kind of the Declaration of Independence or the Constitution, like th- the language is very important. You know, it's, "We hold these truths to be self-evident." What does that mean? It means there ... it's not my rule. It's not the president's rule. It's God's r- And so those rules are above the president. And then you work in that context, and that, that distributes power because you're under the law, not under the person. And we see that, you know, now even with Europe where Europe is kind of the leaders are going, "Well, I have a rule. It's you can't say certain things or I'll throw you in jail." And the kind of shield they hide behind is, "Well, like, we have to keep the kids safe." But if you say something that I don't agree with and the kids hear it, they're not safe, right? Like, so you, the- the- the kind of, um, transitive property of bullshit, like, is gonna override. And so it's really important that we have at least, like, one society. And as flawed as we are, as flawed as the US is, it's still the best, and you can see it by the number of new company creations, the number of new ideas that come out of here and so forth. Um, it's really, really important that the US kind of stays important and powerful in the world. And we know from the last century, if you look at the last century, who were the countries that had economic power, military power, cultural power? They were the ones that industrialized. Yeah, and the ones that industrialized first and best, and the ones that didn't became communists, right? Like, you know, Russia, China, like, they, they were slow on industrialization and they fell into this very fucking dangerous, uh, system. Looking forward, that's gonna happen again but it's going to be AI. And so it is kind of fundamentally important not just to, like, America but to humanity that America succeed at that. You know, like, we don't have to be the one winner or this or that, but we do have to be, like, in that tier. Uh, and, you know, as I go around the world and travel, I can't tell you everybody ... And then everybody was getting, by the way, very, very worried about us, uh, earlier. (laughs) Um, and they say, "Look, we need you to succeed. You know, don't destroy the dollar. Don't, like, fall behind in AI. Don't over-regulate it too early. Don't do- do these things, please, 'cause we need you to win. Like, 'cause we're all, like, counting on that." Uh, and, and I think it's, um, it's the most important work that we do. It's why we're so involved in policy and, and, and so forth. Um, I think this is also, you know, very- gonna be very, very true with crypto which ends up being an incredibly important networking technology that complements AI. Uh, and, uh, and so that's, uh, I, yeah, and that, you- you know, that work is I would say beyond for the money. Like, that's c- y- although, like, we will end up making a lot of money (laughs) with the right policies. Like, so I don't wanna, like, seem totally philanthropic on this. But, but it's more important than that. It's, uh, it's certainly more important than us succeeding or anything like that, that- that- that countries succeed.

16. 1:18:53 – 1:23:18

    The Paid in Full Foundation for hip-hop pioneers

    1. BHBen Horowitz1:18:53

    2. LRLenny Rachitsky1:18:53

       Speaking of philanthropic and other passions of yours, something that I don't think most people know about you and I think will give them another insight into how interesting (laughs) you are, you, uh, run an organization called Paid in Full, which, uh, is incredibly cool. Talk about what that's about, why this is so important to you.

    3. BHBen Horowitz1:19:12

       W- we kind of ... Our, our ethos as a firm is, um, kind of, uh, what I would say is something from nothing. Um, you know, like that this is the greatness of entrepreneurship. You start with nothing and then you, you make something really important.And, um, you know, that, that kind of, th- that is also how hip hop started, where you had, like, a bunch of kids who didn't even have instruments. (laughs) Um, and they kind of created something out of nothing, and th- and the people in that world always talk about that. And, you know, one of the really unfortunate things that, uh, happens is the people, kind of who invent the art form, and certainly in the case of hip hop, don't get anywhere near the kind of proportional benefit of their invention, and a lot of the guys, uh, y- you know, people have forgotten about or, you know, are kind of struggling to make ends meet and so forth. So what we created was this thing called the Paid in Full Foundation, named after, uh, the Rakim Eric B. song, which I did call Rakim and ask him for permission to use the name, so we didn't just take it. And what we do is we give essentially pensions to the old rappers, um, that enable them to kind of continue their work, uh, and then we have a big event, go to paidinfullfoundation.org for tickets, uh, which is amazing, um, where they kind of get the award and, uh, you know, and they're celebrated by all their peers and so forth, and it's really phenomenal. But, you know, so some of the awardees have been, uh, Rakim, um, Scarface, uh, from the Geto Boys, um, Roxanne Shante, Grandmaster Caz, Kool Moe Dee. This year we're honoring, uh, George Clinton for being sampled, uh, Kool G Rap and Grand Puba, um, and also, uh, Jalil from, uh, Whodini? Um, and it's just, it's very, I, I can't even describe how high impact it is on these guys. I mean, I think Rakim was touring close to 200 nights a year, and he got his award and came out with his first album in 15 years and is doing amazingly well, and all of a sudden, you know, everybody's going, "Oh, yeah. That's the greatest rapper of all time." It's like they're finally going back and remembering all the things he did. Uh, Roxanne Shante, m- I mean, nobody had mentioned her in years and years and years, and I think six months after we gave her the award, uh, the Grammys gave her a lifetime achievement award, which is amazing, so it's, uh, it's super high impact, it's a great thing, and, you know, as a hip hop fan, it's a dream come true, so it's the only

Episode duration: 1:37:59