Marc Andreessen谈VC未来：a16z会上市吗，以及为什么内省很危险？

Marc, you probably don't know this, but I started this show when I was 18 years old, and you were one of 3 names that I wanted to have on the show back in 2015. I have to admit I've ticked off the other 2, so I'm a bit worried that I'm going to have to stop after doing this show, but I'm so touched that you agreed to join me. Thank you for joining me.

1. Why Introspection is Overrated: The Dangers of Learning from the Past

Good. I'm thrilled to be here.

I was running and listening to every show that you've done before, and you recently said that you don't introspect and that introspection is potentially overrated. I really struggled with this because I thought we learned from mistakes, and I valued experience in that way. Can you help me understand the lack of value placed on introspection? Do we not learn from mistakes?

We do learn from mistakes, but the problem is that learning from mistakes is sometimes good and sometimes bad. If you just talk business for a moment, in the venture mindset, this is a very big problem.

2. Do You Actually Need to Like Founders? The Uncomfortable Answer

There's a founder version of the mistake and a venture version of the mistake. The founder version of the mistake is if a founder starts a company in a category and the founder doesn't work, the founder is then emotionally angry at that category for the rest of his life and will not acknowledge when there's something that's going to work in that category. I've just seen that over and over and over again. That's fine because most founders go on to do other things, and it's generally fine and good. It generally doesn't damage them from a business standpoint.

In venture, the same thing happens. If you invest in a category, or if you invest in a kind of company, or you invest in a kind of founder and it doesn't go well, it's extremely easy to learn from the mistake and to basically say, "All right, I touched that hot stove. I'm never doing it again."

Then the next thing shows up in pattern matches, and it's the thing that you should invest in and have the chance to invest in, but you touch the scalded stove. You know, you're learning from your mistakes. You're doing the responsible thing, and so you don't do it.

I think there's something that's particularly pernicious about learning from your mistakes in venture capital. I think that's also somewhat true about life. You get married multiple times, as they say: It's the triumph of hope over experience. I think probably you want hope to triumph over experience in that domain, and I think there are a lot of other domains of life in which that's probably true.

I totally understand what you're saying, especially when you say it's easy to lose money in a sector and then think the sector by nature is cursed: You can't make money in healthcare, you can't make money in X.

I'm old enough to remember internet search: You can't make money in internet search. The internet search companies in the 1990s did not work out well.

How do you, when you are guiding conversation and guiding partners, ensure that they have a fresh mind with every new company and every new investment and are not plagued by the downsides of having bluntly lost money before?

By the way, another example is AI. AI was a tremendously good way to lose a lot of money in venture capital from 1945 to 2017.

When I was getting my computer science degree in the late ’80s, AI was the one field that you knew would never succeed. There had actually been an investment boom for AI in the ’80s, and it failed. Everybody, including all the computer scientists, was like, "Yeah, this field is dead." That happened about 5 times over the course of AI over the last 80 years. Again, another great example.

I think a couple of things are important in terms of how we run our firm, or how you run a firm like this. As you well know, there are 2 categories of mistakes. There's the mistake of commission and the mistake of omission, or there's the mistake of cost and the mistake of opportunity cost.

The mistake of cost is you invest $10 million in a startup, it fails, and you lose the money. That's bad. The mistake of omission is you don't invest in Google, and you lose $100 billion of opportunity cost. Venture is the most polarized possible economic field in which this is true.

If you're running a bond business or a debt business, where you just can't lose money or the whole thing doesn't work, then obviously you can't run in this kind of model. In that case, you better learn from your mistakes. But in venture, I think you're always much more worried about the mistake of omission than you're worried about the mistake of commission.

To your question, in a lot of ways, that's the key thing that Ben and I do at this point in our lives and in our roles at the firm. We're not micromanaging the investment decisions at the firm. We have spectacular senior partners and junior partners who are doing a great job of that, and we're in the room for it and so forth, but we're generally not advocating for or against a deal.

What we are trying to do is get everybody to constantly have this risk-forward mindset: Worry about the mistake of omission before the mistake of commission. This anti-scalded-stove phenomenon—we routinely remind people, "Yeah, you're emotional about this because of your bad experience 3 years ago, 6 years ago, or 10 years ago. Just let that go."

You no longer have to pay for that sin, and you're completely liberated to let that emotion fade into the distance and be able to focus on the opportunity in front of you.

My biggest regret, or omission experience, is one of your companies, actually. It's ElevenLabs. We could have invested at the seed round, but we would have only got 1%. Naturally, as an emerging manager, I thought it was important to retain the high-ownership model I promised my LPs. How do you reflect or advise me on when to break the rules versus when to maintain doing what I said I would do?

Quite honestly, it's the simplest answer in the world, and it's the hardest answer in the world. It's the answer that I think every great investor ends up resolving to 30 years in.

I had this discussion with Arthur Rock, who's obviously virtually the creator of modern venture capital. He actually wrote a paper on this topic. I'll just give you his conclusion, which is also my conclusion.

Arthur Rock, for people who don't know, invested in Apple and Intel in the seed rounds, among many other great companies. For 30 years, he said that he would have been a better venture investor had he fed all of the business plans and pitch decks straight into the shredder upon receiving them, and if he had spent 100% of his time on the résumé.

I think that's basically right. The great founders will buy you enormous upside that may break rules in all kinds of directions and may break precedent in all kinds of directions. The world's best business plan executed by a mediocre team will almost certainly get lapped by a great team.

Having said that, this sounds easy. Of course, why is it hard? It's somewhat tautological, because we define great founders as the ones that have great outcomes. It's a lot easier after the fact to say, "Oh, yeah, Steve Jobs is a great founder," when you look at the success of Apple.

Nevertheless, I think that is the answer: When you have special people, you should back them almost without consideration of other factors, and when you don't, you shouldn't. At the end of the day, that simply is the core thing.

3. The One Trait Marc Andreessen Looks For in Every Founder

How do you think about detecting greatness in founders when your benchmark is founders at a late stage, when they're great? You obviously have been on the board of Facebook for many years with Zuck and seen him at a later stage as well as at an early stage. I spend my time interviewing public company CEOs all the time. Marc, I'm so used to really fine-tuned delivery. When I meet a seed founder who's rough and unpolished, of course, they don't seem as good. How do you think about that challenge and projecting earlier to see if they're good, given how much time you spend with perfection?

I'll just say, look, I think people have different takes on this. My personal formula is basically as follows: You need high IQ as table stakes. You just need somebody who's incredibly smart.

My basic test is, if I have my notebook open and they're talking, am I writing out lots of notes or not? If I'm writing out lots of notes and I'm learning from them, that indicates their level of intelligence and some of the other attributes that we'll talk about. Clearly, they're very smart.

But I think that's table stakes, because just intelligence isn't enough. There are many people who are very smart who are just grinders, or who have, as they say, the clerk mentality: “Put me in the back office somewhere, doing research or something, and I'm never going to build something.” That's fine, but IQ is not enough.

I think the second thing you really need is what my partner Ben calls courage, which is an absolute determination to succeed, to confront problems directly, and to pound through anything. There are various ways to phrase this, but I think the Navy SEALs have the term “embrace the suck.” [laughter] There's something to that.

I always liked the old Looney Tunes cartoons, and I always like to say I want the founder who leaves a founder-shaped hole in any brick wall he runs into, like a cartoon character.

Yeah. [laughter]

My favorite is when they run off a mountain and keep running. I don't know if you've seen this. Then they're suspended midair for a moment, and they hold up the sign that says, “Oops.” [laughter]

Or my favorite one of those was when one of the little kid characters—some little pig or something—did the same thing. He goes out over the cliff, hangs there, and holds up his sign and says, “I'm in second grade. They haven't taught us about gravity yet.” [laughter] Of course, this also happens in startups.

Reid Hoffman told me to build a parachute on the way down. [laughter]

Yeah, exactly. That sounds great. Get out your knitting needle and get going.

Then I think the third thing is sort of IQ plus courage, and then something fundamental. It's drive, ambition. I occasionally quote Nietzsche's will to power. It's this determination, because courage can just be, “I'm going to solve problems,” and I would argue that's not enough. Ben might say that's not what he means, but it's not just solving problems. There's something about ambition.

In the world being what it is, a lot of people express that ambition as ambition to change the world, improve the world, improve humanity, and so on. I think that stuff's all great. I think those founders are great, and I think those missions are often very compelling at attracting lots of smart people. I think that's great, but I do think there's a more fundamental ambition, which is: “I want to build something of my own, and I want to really demonstrate what I can do.” I have a very primal drive to do that.

People cast moral aspersions on it and call it greed or whatever, so people don't want to talk about that. I'm not even talking about the money component. I'm talking about, “I want to build something.”

What I find with that one in particular—and, actually, with numbers 2 and 3—is that you don't necessarily see them on the résumé, but you can generally see them in the background. If you spend enough time with people, you can get a sense of whether their entire life has basically been a sequence of things being handed to them, followed by credential achievement, which is a lot of what we see in the elite workplace.

Or do you have somebody where it's, “When they were 14, they built this; when they were 17, they built that; when they were 20, they did this”? They've always been in that mode, whether that's building a product, a technology, a company, a work of art, or whatever it is—a primal drive to create.

4. Are the Best Founders Broken? What Makes the Best Founders?

Do you find drive through pain to be the most contributing factor to success? The full version of the theory is that all the great founders are broken in some way, right? You get into psychoanalysis quickly, but you get into the broken home, Steve Jobs being adopted, and all these stories. The metaphor is that when the bone breaks, it either doesn't heal, or when it heals, it's stronger. You're trying to get people who are responding to childhood pain through overachievement.

I think what there is to that that's really important—I often talk about it like this—is that you need some reason to get out of bed in the morning. It's not just, “I have a job,” it's not just, “I don't want to embarrass myself,” and it's not just, “I want to be responsible.”

You have to have a primal reason when things are really, really bad—when the shit really hits the fan, you're miserable, you dread checking your email, and you simply do not want to know what the new bad news is because there's so much bad news that you can't even cope with it. You need a very, very, very primal reason to get out of bed and continue to fight that way.

I think there is something about trauma in the background that explains that. Having said that, some of the best founders in history have no trace of trauma in their background that I can tell.

Zuckerberg is one. He grew up in a classic upper-middle-class New Jersey household, was very close with his parents and his family, and, as far as I know, had a perfectly great childhood. Then Bill Gates—his father was a lion of the Seattle establishment, and he went to all the best prep schools and Harvard. Again, as far as I know, he had a perfectly great childhood.

People who knew both of those guys in their teenage years said, “These are driven guys.” It's very core to their origin stories. You just have to be open to the idea that some people are simply born that way.

What's your primal reason today for continuing to build Andreessen Horowitz with the ferocity and ambition that you still have?

Well, of course, that would require introspection. [laughter] I don't know if I'm going to give you a great answer to that.

I'm elegantly backing into it. Okay. [laughter]

What I tell myself is that, at this point, I'm competing with myself. By which I mean, I'm trying to figure out how to be the best possible version of what I can be and what I can do.

The way I think about it is, “How good did you read—was it the book Extreme Ownership? Did you ever come across that?”

I love Jocko Willink. I also listen to his motivational talks when I go to the gym. Brilliant.

100%. You know his thing on extreme ownership, right?

For people who haven't heard his thing on extreme ownership, he's a famous Navy SEAL commander, a very accomplished guy—the kind of guy people would happily follow into battle, or who would be a great CEO or great founder. That kind of personality.

He has this thing called extreme ownership, and he says, “Life just gets a lot simpler if you assume everything is your own fault.” [laughter] Whatever it is—this LP didn't invest, or this founder didn't take my money—it's like, “Oh, okay, it's my fault.” It's not his fault. It's my fault. Clearly, I didn't do a good enough job. Clearly, I can do better.

His argument is that it gets you productively focused on improvement.

And so, I found that—put it this way—when I'm in my own head and I'm mad about somebody doing something that I don't like, the number one stress-relieving thing I can do is say, “Oh, that's my fault.” Right? Because then it gives me ownership of the problem, and it gives me something that I can do. By the way, it also drains away resentment. It means that I'm not resentful and angry at somebody else, because I'm just like, “Okay, I'll just improve myself on that.”

So I operate in that psychology as much as I can. I try to maintain that psychology. By the way, I recommended that book to Ben when it came out. I fell in love with it, and I said to Ben, “We need to send this to all of our founders and teach this.” And he's like, “Marc, you're out of your mind. Our founders already have the problem where they take too much responsibility. They take too much of the weight of the world on themselves. They're already miserable half the time. We don't need to saddle them with more of that.”

But I think there's something very, very powerful in that. It also has the enormous advantage of becoming an intrinsic motivation over an extrinsic motivation. It's not a motivation to put points on a board. It's not a motivation to achieve a certain net worth. It's not a motivation to be in some league table or to win some prize—these sorts of external markers. Because the problem with all the external markers of success is: are you going to get up in the morning when it really, really sucks? The extrinsic motivations don't do that.

You need something intrinsic, and for me, that's the intrinsic motivation of, “I know I can do this better.”

5. Why Everything Being Your Fault Changes Everything

You said that you're competing with yourself. Do you feel you're your best version of yourself today?

I think I'm my best version of myself relative to all my prior versions of myself. But I'm still far short of what I would like to be. I know of many, many areas of improvement, I guess I'll put it that way.

What's the biggest one that you'd like to change?

Oh, I mean, I could—I mean, there's probably a hundred. I'll give you an example. I have a strength and a liability, which is that I get emotional. The advantage of emotion is that when I commit, I deeply commit. I fall in love with things and become incredibly determined, and I'll go to very long lengths out of a sense of emotion or love.

The negative is that I will get emotional. I've spent a lot of time—and people who know me will tell you I've spent a lot of time—trying not to get negatively emotional in meetings.

6. Fame, Criticism & How to Deal with Haters

Do you care what people say about you? It's something I'm trying to work on, but I still desperately care, honestly, Marc, and it desperately upsets me when I read bad things.

I have a bunch of friends in the entertainment business who I look at and say, “There's no way I could possibly do what you do, which is make myself vulnerable on an 80-foot screen that way.” And they're like, “Yeah, that's the hard part.” Then I always ask, “Do you read your own reviews? Do you read what people say about you?” They all basically say the exact same thing: “I tell everybody I don't, and then of course I do.”

It's very hard to avoid that. I do think “don't read the comments” is generally a very good life guideline. By the way, I will say YouTube comments have gotten much better, so maybe your YouTube comments are productive now. But in general, “don't read the comments” is helpful.

I mean, it's really hard. Everybody's human. I think it's really hard when somebody is cursing you out or calling you horrible things. It's very hard for that not to stick. I would say I'm pretty happy not paying attention to that. Are you aware at this point of the meme of retardmaxxing?

Do you know, if I'm totally honest, I've seen it on every comment of our thread where I say, “I've got Marc coming on the show,” and everyone's like, “Ask Marc about retardmaxxing. Ask Marc about retardmaxxing.” Honestly, it's one of those things where I'm just like, “Okay, get back to my normal research, because I presume retardmaxxing is not politically correct and I shouldn't ask about it.” But you brought up retardmaxxing, so no—to me, I don't know what it is.

Well, first of all, retardmaxxing is totally politically correct because we now have 18 other terms that apply to people who are developmentally disabled. And so “retard” has long since come to mean something completely different.

Let me explain why I came across this. The internet meme machine is absolutely spectacular. I think the process of cultural evolution of internet memes is absolutely amazing. I think the whole looksmaxxing thing—I don't even know if you're into this looksmax thing—the whole, like, Clavicular thing, all the terms now are mainstreaming.

Obviously, there are many internet meme examples. One of my favorite websites in life is Know Your Meme, just the comprehensive catalog of memes. The cultural evolution of what's happening online, I just think, is incredible and wonderful in so many ways.

Then I got into this dust-up online a couple of weeks ago about introspection, which you mentioned, and a friend of mine sent me this thing. He said, “Oh, here's your answer. You're retardmaxxing.” And I said, “I'm what?” He said, “Watch these videos.”

There's this guy—we could link to him—who's on YouTube and has basically 100 videos on retardmaxxing. He's my new life coach. I haven't met him, but from a distance, he is. It's basically just like, “Retard, go to work, do a good job, come home. It's fine. Start a company. It succeeds, it fails—it's fine. Have too much to eat one night at dinner? It's fine. Go to the gym. Don't cut your reps—it's fine. Ask a girl if she wants to go out with you. If she says no, it's fine.”

It's the simpler form of extreme ownership. Or maybe it's another form of it that says, “I don't need to take all this in on myself. I can just let it go.”

The thing I love about the internet, Marc, is that there is some guy who is doing these retardmaxxing videos who now has Marc Andreessen as one of his biggest fans. You're just like, how great is that?

They're incredible. It's like a 30-minute video about retardmaxxing, and you would think that after the first 2 minutes, he had covered it. But no. By the way, they're all hysterical. They're all absolutely fantastic. It's literally him on his porch in the middle of nowhere with a cigar, and it's a half hour. It's absolutely spectacular.

I do think there's something to that. Back to your original question, in addition to all the emotional pain that life has already put on us—or that we've already put on ourselves—and a lot of it legitimately so, for the things that we actually do to other people and so forth, it's just like, okay, how much are we going to torture ourselves?

There's something about modern culture, modern Western culture or something, where we've become very guilt-oriented and very into self-flagellation, very into the old concept of the hair shirt. We wear these metaphorical garments that are tremendously painful. There's a point at which some of that is helpful to correct bad behaviors, but it's clearly gone way too far. People get way too far down the rabbit hole, and it becomes very disabling. You probably know a lot of people who are like that.

The way I think about it is that what you do, what I do, what venture startups are like—look, these are high-risk operations. Sometimes they go right, but they go wrong in a thousand ways before they go right, and then even then they may not work. The nature of the beast is tremendous variability and pressure.

Another thing I always thought about a lot as a founder, and I really see this now as a VC, is that founders in particular have a very hard time ever finding anybody they can confide in. If, as a founder, you feel like you admit that you have an issue, you're being a bad leader, because you're showing a crack in the armor. If your people pick that up, they're going to lose confidence in you.

Or if word gets around that you're second-guessing yourself, or that your thing isn't going well, or that you don't have total confidence in it, then all of a sudden investors won't want to invest and candidates won't want to join. So there's this need, if you're going to lead one of these things, to do it with such a brave face. I always call it a metaphor: the duck looks totally placid above water, and then it's paddling furiously underwater.

I just think that founders, in particular, have a very hard time finding anybody they can actually confide in. What happens is that I think everybody individually has an inaccurate view of what everybody else is feeling. In practice, everybody's feeling very tense, nervous, anxious, fearful, and so forth, but everybody's pretending they're not feeling that way. Everybody thinks everybody else is doing great, and everybody else thinks they're the only one faking the smiles at the party.

I think it's incredibly important to have an internal psychological mechanism to deal with that and not have it overwhelm you. And, yes, let me just say, at least this week, my nomination is retarding.

You can kill me. You can tell me. I don't want to ask that. I remember I did a show with Orlando Bravo, and it basically turned into a therapy session. He kind of became my adopted father. Good dude. He gave me a lot of advice.

What am I scared of? I'm scared that I'll be someone like Macaulay Culkin. Marc, do you remember Home Alone?

Yeah.

The kid who everyone knew when he was young, and then it's like, “Oh, yeah, what is he doing now?” Oh, kind of no one knows. I have nightmares about being the Macaulay Culkin of venture.

What are you scared of?

To me, you're the great Marc Andreessen of Andreessen Horowitz. You are nothing to be scared of.

I mean, I've been through every version of this. There's a famous F. Scott Fitzgerald line—he was the author of The Great Gatsby—where he said, in the 1920s, “There are no second acts in American lives.” You get one shot, and that's it. Fortunately, I think he was very, very, very deeply wrong about that.

I think he was definitely wrong about that for American lives, and I think he was generally wrong about that for lives, at least elsewhere in the West. Maybe a little bit less so in Europe, but I think still more than not.

The point being, we don't—look, somebody once told me there are 2 great stories: “Oh, the glory of it” and “Oh, the shame of it.” “Oh, the glory of it” is the story of great success. “Oh, the shame of it” is the story of great disaster. But the even better version is “Oh, the glory of it,” followed by “Oh, the shame of it,” followed by “Oh, the glory of it.” The recovery: getting back up on your feet, re-achieving, and rebuilding.

I don't know. I think as long as you're still alive, and as long as you've conducted yourself in a way that you haven't brought some sort of fundamental legal issue on your head, I think generally, at least in our world, second chances are actually available for a lot of people. By the way, of course, a lot of the great success stories have this in their background, including Steve Jobs himself.

7. Is Venture Now Go Big or Go Home? The Real Future of VC

Of course. You mentioned the nature of the beast there being our business. I've been a student of the business, hence reading so much of your writing for years. When we look forward, how do you think about the future of venture? Is it as simple as “go big or go home”? Obviously, we see Andreessen Horowitz so big now.

I believe—and we try to run the firm this way—that the core of the business is a permanent state of affairs. The core of the business is early stage. The core of the business is a founder or a small founding team with a dream and a clean sheet of paper and, ideally, a garage, although these days it's hard to keep the kids in the garage, so maybe they have a house or an office.

They're expensive in Palo Alto. These are not cheap garages.

Palo Alto garages are indeed expensive. You look at a couple of kids, a dream, and a clean sheet of paper; first money in, and then the first 2 years. That is the core of the business. That fundamentally is the core of the business.

A metaphor we use all the time at the firm is that startups are like baking a cake. If you leave the sugar out of the cake—if you bake the cake and leave the sugar out—you can't pour sugar into the cake afterward and fix your mistake. Sugar has to go in the cake.

The first 2 years are when you're baking the cake. That's when you're really figuring out the formula for what you're doing: what the product is, what the company is, what the business is, what the culture is, and who the team is. Those decisions are absolutely fundamental. If you get those right as a founder, the payoff from that will go for decades. If you get those wrong, even if your company succeeds, you're going to live with those sins forever. They're going to extrapolate out.

That's the core of the business. Many great companies later bring in lots of other partners and growth-stage investors, add other people to the team, build boards, and so forth. It's fantastic and it's great, but there really is no substitute for that inception point, that early moment.

There's no substitute for being the investor who does that. As you know, the investor who's engaged with the company at that stage often becomes the key adviser to those founders for the rest of the company's life. You build this incredible emotional bond, and you have complete context on why all the decisions got made. You remember how it first started, and you remember how hard it was.

I think there's just no substitute for the early stage. This is what I always tell our folks: At the end of the day, the early-stage business has to work. If the early-stage business works, we have option value in doing all this other stuff, but that always is the core of the business.

To what extent is the late-stage fund a function of executing on the omissions of the early-stage fund?

I think it's basically in 2 parts. Part of it is, yes, fixing the mistakes of omission—fixing the mistakes and becoming partners later. That can work really well. Those can be very good investments. We do get very close to some of those founders, but again, they always have somebody early on with whom they're very close, so we see the difference there.

The other part is doubling down on the companies that are working or growing. Part of that is just economics: If you have the chance to do that as a professional investor, you should do that.

There's another really fundamental reason we decided to go so big in growth. This is less true now, but 10 or 15 years ago, these companies would raise money from venture investors, and then they would get to a certain point and raise money from a completely different kind of investor that was not tech-centric. They would suddenly end up in a situation where they had a conflict between investment mentalities—a fundamental conflict on the cap table over things like level of risk, level of reinvestment, how fast they should go public, whether they should sell the company, and whether they need to replace the founder and bring in a professional CEO.

To the extent that you bring in non-tech-mentality, non—whatever you want to call it—Silicon Valley-mentality growth-stage investors, you set yourself up for a different set of pressures. One of the things we wanted to be able to do was, with our founders who have the chance to build something really great, be their partner across potentially every round that they do.

As a consequence of that, they can preserve our mentality on their cap table for longer and longer and longer. I think that works pretty well.

Is it possible to literally care about a $5 million seed check when you have $15 billion that you raise at once?

Yes.

It is, and the reason for that is twofold. One is just the conceptual reasons that I described, but the other is pure economics. The upside on the $5 million check is every bit as big as the upside on a $500 million growth investment. Right? This is what’s so unusual about venture. If I make a $5 million seed investment and I nail it, I can make $10 billion on that, $100 billion on that. If I make a $500 million growth investment and I nail it, I can make $10 billion or $100 billion on it. It’s the same upside.

8. Does Price Matter Anymore? The Dangerous Truth About Valuations

Do you buy the “entry price doesn’t matter because we’re going to have $100 billion companies” argument? I just see the round inflation across every round. It makes my life harder. With the greatest of respect, large funds make my life harder because you have a different cost of capital. Do you buy the idea that if it’s a $100 billion enterprise, the entry price doesn’t matter, or do you think differently?

Yeah. So, look, the entry price definitely matters, in particular as the company grows in size. By the way, it matters for a couple of reasons, and this is a lesson that gets relearned over and over again and will be learned many times in the future. It’s the old Don Valentine thing, which I do think is correct: more companies die from indigestion than from starvation. Overfunding is actually very dangerous to the operations of a company. This is the one piece of startup advice that I think is tremendously grounded in reality, for which everybody has many examples in the past. No founder ever listens to it.

My track record of ever convincing any founder on this point is zero, but I will keep trying. So, number one—

Because it’s so flattering. “Oh, I want to give you money. Okay, you think I’m brilliant. Oh, great.”

Yeah. Well, then they come up with 18 reasons why. Then I’ll really push them on it, and they’re like, “Well, we’re going to have a lockbox.” It’s like, “No, you’re not.”

I’ve never had the lockbox.

Nobody’s ever seen it.

Nobody’s ever going to do the lockbox. Nobody ever does the lockbox.

So, back to your question. I think high valuation—I’ll come back to high valuation in a second—but I think there’s an actual core, fundamental linked thing that’s very important, which is that the amount of money you raise through overfunding is actually just as dangerous, or more dangerous, than underfunding. Number one. Number two, the problem with these high valuations is: God help you if you need to clear the bar next time and you can’t, right?

9. Why a16z Invested $300M into Adam Neumann

Every round sets a post. It sets a threshold, a hurdle, for being able to raise in the future. This is something that people learn every cycle, for the first time in a hard way. No new investor wants to do a down round in anybody else’s company. If you put the investor hat on and you’re like, “I’m going to go do a down round in this company because I’m going to be the hero and save the company, or whatever, because it raised too high last time and now I’m going to do the rational investment,” everybody’s just going to hate me, right? The employees are going to hate me, the other investors are going to hate me, and the founders are going to end up hating me. Nobody ever does a down round in somebody else’s company. Setting these posts high is intrinsically a problem. Once again, this is advice that people generally completely disregard.

There are problems like that in the system now, and there will be more problems in the future. Having said that, at least on the venture side—growth is a little bit different—I think every time we’ve passed on a promising venture company because of price, I think it’s been a mistake.

Have the best companies been the most expensive?

So, I think the underlying question—and tell me if you agree with this—is the question of diamonds in the rough. Is that right?

Yeah. Whenever I’ve done a good deal, it’s never worked out.

That’s right. Here’s another thing we say in the firm: don’t ever do diamonds in the rough; only do diamonds.

This is another investor-ego thing, I think. You basically say, “Wow, I’m the investor who’s going to go find the thing that nobody else knows about,” right? Another form of this would be, “All these other investors are herd animals. They’re all just copycatting each other, and I’m the one who’s going to be different. I’m going to do the thing nobody else can think of.” By the way, Peter Thiel does that really well. Nobody else does that well.

And you’re probably not Peter Thiel.

And you’ve probably spent a lot of time with Peter.

You and I say, “I am not Peter Thiel.”

And yes, you, the listener, probably are not as well. Maybe I could say this: there’s a tremendous amount of belief that VCs are stupid. VCs are herd animals, blind, consensus-seeking, heat-driven. They only do the obvious thing. You often get this from people: “They don’t appreciate my special thing.”

Having said that, the general pattern is—9 out of 10 times, or even 95 out of 100 times, maybe 99 out of 100 times—that if it’s got merit to be investable for venture, there are a lot of really smart and hungry VCs out there. They’re working extremely hard to sniff these things out. It’s their full-time job, and it’s all they do. So, I think it’s really unusual to have the diamond in the rough.

Usually, if it’s the diamond in the rough, it means one of two things. Number one, it means a company that’s offside for some fundamental reason. It’s in the wrong place, right? Or it’s structured wrong. There’s a reason why it’s a diamond in the rough that actually ends up becoming a big problem. The example people use, which I think is legitimate, is that there was a point when Uber was available for investment by anybody on AngelList.

So, every once in a while, there’s one of those. There’s a reason why, if you just look at the great outcomes in venture over the last 50 years, and rank the outcomes, it’s the same names over and over and over again. It rotates; every decade or so there’s some rotation in the names, but the persistence is incredibly strong.

By the way, that’s the other reason you have the diamond in the rough: you have a founder who fundamentally is just too ornery to do things the obvious way. They’re hyper-disagreeable, and they have all these theories about how venture is terrible and awful and these VCs are all evil. They’re very focused on terms and control and all this stuff, and they kind of alienate people. By the time you meet them, they’ve alienated 6 of the mainstream venture firms, and now they’re the diamond in the rough. Every once in a while, one of those is going to succeed, but I’m not sure I would want that to be my business.

Do you need to like the founders you invest in, Marc?

So, I say no. Opinions vary. I said earlier that I’m emotional, both in good and bad ways. You do end up getting very close to people, and you do end up wanting to have a high level of trust. It certainly helps if you like each other and trust each other and so forth.

On the other hand, some of the best founders in history—and I can give you example after example in the distant past—were not very likable people. The same thing is true of many of the great artists, filmmakers, literary geniuses, philosophers, and, by the way, political leaders. CEOs, too—there are a lot of cases where these people are not likable. I say no because if you’re trying to fulfill your personal emotional needs at work, I think that’s a very fundamental problem, and you shouldn’t try to do that.

It’s the Harry S. Truman quote: “If you need a friend, get a dog.” The point—or another version of this—is that I say, “Do not bring your whole self to work.” Whatever you do, do not bring your whole self to work.

If you show up, you're professional, you're great to deal with, and you're very productive. You're adding value in every engagement that you do. If that's true for you as a VC, and you're working with a founder, and you're never friends, but you have a great working relationship, and the company, in the later years, sells or whatever, and you never talk to each other again, I've seen that work many times, and I think it's totally fine.

10. When Will a16z Go Public?

Marc, do you want to take Andreessen public? It's the question that came up time and time again, but when you look at the machine that's been built, would you like to take it public?

Yeah. So I think we don't have to confront that question. It's a serious answer: there's, by the way, nothing we're missing today that we could solve by going public. As you know, by the way, that's increasingly true of a lot of the companies that we both invest in.

I would never rule anything out. Ben and I have run public companies before. Ben has specifically been the CEO of a public company before, so we know what that entails.

I'll tell you my funny version of the story. When we first started a16z, we went around and met with a lot of the top VCs at the time. This was in 2008 and 2009, and we pitched them on what we were doing and got a variety of very interesting feedback. Some of them became very helpful to us and really helped us.

One legendary VC told us at the time, "The thing you're going to hate the most about being a VC is the LPs. These LPs are just the worst people in the world." He then gave us what we call the mushroom talk: "You need to treat LPs like mushrooms. You put them in a cardboard box, put the lid on the cardboard box, put the box under the bed, and don't open it for 2 years."

We said, "Okay. That's one mentality." Then we said, "Wait a minute. We've been running public companies for the last 15 years, and we've been dealing with hedge fund managers. Say what you will about LPs, but whatever you think, at least when you walk in the room, you know they're not short your stock." If you want to deal with pain-in-the-ass investors, go public.

Of course, what we found is that our LPs have been incredible. Our LPs have been incredibly supportive and incredible partners. We obviously try to treat them as partners, but it's just been an incredibly productive relationship. As you know, the best LPs understand venture, they understand the time horizon, and they understand the risk aspects that we were talking about earlier. They've given us license to do a tremendous number of things that have been very risky, of which some have worked and some haven't.

It's been an incredibly productive partnership. I go through all that to say that I can imagine venture firms going public. I think you'd have to have a real theory on the value that you would get, and you would have to really sign up for what it takes to run a public company these days. I would just say that public company CEOs have a very hard job.

If you were a betting man, which I guess you might be, who would go public first: Andreessen or General Catalyst?

That's a good question. I haven't actually talked to him about that. He's certainly building a firm that could go public, but I don't know whether he would or not.

How big a check do you have to write as an LP to get in the meeting with Marc Andreessen?

Oh, you—[laughter] I would shut that question down.

Fair enough. It was a press. I was just intrigued.

I will say this: I think it's actually the same answer as your seed question of what you care about in a $5 million investment. There are certain LPs that are really, really smart. Specifically, there are certain LPs that are very influential in the LP community.

They are not necessarily the same LPs as the ones who write the biggest checks. I should probably not get any names, but there are certain LPs I would 100% meet with, independent of check size. Those are the great ones, the really great ones.

What product do you not have in the Andreessen suite today that you would most like to have?

You mean investment product? You mean investment strategy?

Yeah.

The 2 that we've kicked around for a long time are public equity, on the one hand, and credit, on the other hand. I think there are really good reasons to do both, and then there are issues with both, specifically with respect to running them inside a venture firm. We've never hit the catalyst moment where we've pulled the trigger on either one, but those would probably be the 2 nominations.

11. Why Silicon Valley Is More Dominant Than Ever?

If I were asking you about diamonds in the rough, one of which you mentioned earlier, I would say that, well, I'm in Europe, and location can help you find diamonds in the rough. Do you think you have to be in San Francisco today, or Silicon Valley today, if you're building an AI company?

Yeah. Let me start by saying I wish we could decentralize tech. I come across as a Silicon Valley partisan a lot, and a Northern California partisan. I should, by the way, note that I didn't grow up here. I'm an immigrant to the US, to California. By the way, I haven't left.

Gone to Miami. I get lots of questions like, "Why has he moved to Miami?" I'm like, "Has he moved to Miami?" My research tells me no, but—okay, maybe he's done a Sergey.

No, I'm a Californian. I'm very dug in in California. I am not a Silicon Valley partisan in the sense that I think everything should be in Silicon Valley, or that I think it would be good if everything were in Silicon Valley. I don't believe that.

I am a very, very keen—I would say—student of all the issues in Silicon Valley, and I could spend a long time taking you through them. You probably know them all already. Silicon Valley has real issues as a place, including practical issues: cost of living, cost of housing, cost of transportation, commutes, and then, when you get into politics, it's a whole other kind of parade of horribles. There are a lot of issues.

San Francisco proper has a lot of issues. It's a city that 100% does not want to grow. It's a city where voters, on average, do not want business to be there. It's a city that has real issues with quality of life and so forth.

I would love to see the industry spread throughout the US and then spread throughout the world. I would love to see that. I was very optimistic about that happening in 2020 and 2021.

I thought COVID was obviously horrible, but the sudden phenomenon of video conferencing, and then Slack, and then the virtual workplace, and all the hybrid work, and all these new management methods and technologies that were brought to bear to help companies decentralize and run from home—I was blown away in 2020 that the banking system didn't collapse, the stock market didn't collapse, and that it turned out you could just put all these companies online and they could keep running.

The Valley didn't collapse. In fact, a lot of Valley companies grew a lot. So I was very enthusiastic between 2020 and, let's say, 2023, that we had cracked the code on how to finally get away from the geographic constraints of Silicon Valley.

I think in the last 2 years, that process has whiplash-reversed in an incredible way, and I think the tech industry is more centralized in Silicon Valley than it has been in its entire existence. I think it's AI, very specifically.

I think something very close to 100% of the quality AI companies are in California, and specifically within a 20-mile radius of where I'm sitting right now. There are exceptions, and ElevenLabs, of course, is one of the big exceptions, and Black Forest Labs is another. We have a whole bunch that we're very proud of. There are definitely exceptions.

But, man, if you look at just the value-creation numbers, and if you look at the talent base, and if you look at the flow of where people are going, for better or for worse, it's in Northern California. I just think in practice this region is going to be more central in the next decade than it's been in the last 50 years.

You mentioned the multitude of problems that are in the Valley and California more generally. When you look at the state of play in the US today, are you more optimistic today, or are you less optimistic today?

I'm a lot more optimistic than I was 2 years ago.

I’m a lot less optimistic than I was 20 years ago. There is something magical in the American—I don’t know what you want to call it—character, psyche. There’s something in it, and, quite honestly, a lot of it is the inflow of people from all over the world. A lot of that is the great Europeans who have moved here over the last 400 years.

There’s something about having a country that is this big and this powerful and, let’s say, lucky and blessed in its geography, natural resources, size, and scale, that is nevertheless incredibly dynamic and has risk-taking at the core of its DNA. There’s a willingness and a history of throwing the harpoon at really big bets in extremely aggressive ways. There’s just something amazing about that.

You always worry—or at least I always worry—that that’s diminishing. There’s this term, “managerialism,” that I use a lot, but you always worry that everything’s just becoming managerial, bureaucratic, and stale. There are certainly lots of aspects of the US in which that’s true. But when the new thing appears, there’s something in the American character that jumps at it like crazy and throws the harpoon unbelievably hard.

That’s exactly what we’re seeing in AI right now, right? The level—I mean, it’s actually something that I think is really underdiscussed—the level of enthusiasm, capital concentration, and determination on the part of the people involved. I’m sure you saw Elon’s presentation the other night, right? You watch that thing and your jaw is on the floor. I spend all day with incredibly competent, capable people with great ambitions, and I get to work with Elon on some things, but I watch that thing and my jaw is on the floor at the scope of the ambition.

The honest truth is, Elon would say there’s only 1 place in the world where that could be accomplished and achieved: here. There’s only 1 place in the world where Elon would be able to do what Elon has done over the course of the last 30 years. Thank God he came here to do it. The big AI labs are doing absolutely amazing things. What NVIDIA is doing is absolutely amazing. There’s just something to that.

By the way, I understand why a lot of other parts of the world don’t want that. Young Marc would have been like, “This is crazy. Why doesn’t everybody see this? Why doesn’t everybody do this?” Obviously, it’s not all pure upside. Part of the American character is rougher than that of a lot of other countries and cultures, so there are definitely pros and cons to it.

Do you worry about the inequality that we’re seeing in terms of wealth inequality? It feels to me like it’s greater than it’s ever been. I think we’re seeing wealth created in technology on a larger scale than it’s ever been, obviously. Do you worry about that wealth inequality?

To start with, it’s definitely not greater than it’s ever been. [laughter] We know that because we know history. The natural mode of history for thousands of years was that there was a strong man—we called him the king or the prince or whatever tribal leader—and he had all the stuff. Then there were the serfs, who just worked the fields and didn’t have any stuff. Then, God forbid, typically in human history, there were the slaves, and they also didn’t have any stuff or any rights.

The long-run state of human history has been a much greater, much more profound level of inequality than anything under capitalism. Number 1, I would challenge the premise of the question. Number 2, the debate about inequality is always: Would you rather live in a society that has a faster level of aggregate growth and generally rising standards of living across the board, but with greater inequality? Or would you rather live in a society with a lower standard of living, lower growth—or maybe even no growth or declining growth—in which things are more equal?

I have a lot of European friends who say, “Marc, you don’t understand. For a normal person, living in Spain is much better than living in the US because the baseline is just much more secure.” I buy that, and I think that’s probably true. Having said that, if you want the country that is going to go to the moon and build AI and all the rest of the stuff that is happening here, you’re going to have a dispersion of outcomes.

I think if you look at the economic growth rate itself, it tells you a lot. Just as an example, there are a bunch of European countries that are now either flat or shrinking.

Do you worry about the future of Europe when you look at that flat or shrinking growth rate for many European countries? Do you worry about it?

I am tremendously pro-European. I’m pro-European at my very core. I’m an Anglophile, a Francophile, and a Germanophile. I love all these countries and all these people. Every country in Europe, I think, has made fundamental contributions to civilization. The human capital in Europe is absolutely amazing.

You’ll hate what I’m about to say. [snorts] One of my things at the firm is that we should back every single European founder who moves to the US. We should reflexively say yes, because—

I 100% agree with that. I think the data would agree with that, too.

Exactly. That’s a combination of 2 things: the raw level of talent and, by the way, the great education system and everything else that goes with that, which Europe has a lot of, coupled with the fact that moving to the US indicates a willingness to seek risk and throw things up in the air to go after a greater level of achievement.

I want to see Europe flourish. I would love to see Europe be full-scale, every bit as dynamic and exciting as the US is on all these fronts. I would love to see AI in Europe be a huge thing. I would love to see London—I mean, God willing, we’ll have elements of that, but I would love to see it. Obviously, London has already played a key role with DeepMind, and ElevenLabs is heavily based there now.

If I were to make you head of the EU, Marc, what would you change about Europe? You can change anything. It’s a magic wand to incite growth and ambition in a way that would allow us to seek new levels of achievement.

I honestly think I’ve had this conversation many times. Over the course of 30 years, I’ve been visited by lots of heads of state, senior officials, people working on different kinds of commissions studying this kind of thing, and so forth. Basically, the conversation is always the same. The good news or bad news is that the conversation is always the same: “We really want a Silicon Valley kind of phenomenon in location X.”

Then I say, “Okay, then do A, B, and C. Here are the things that you do to do it.” Then they say, “What if we can’t do those things?”

We couldn’t do that. No, that’s—

No, no, no. Clearly, we can’t do those things, but there must be some other set of things we can do.

Do you have an option B? [laughter]

Exactly. This is the thing: you and I—and I think every one of our listeners—can fill in exactly what A, B, C, D, E, and F are. As you know, the Draghi report just did this, right? He wrote the Draghi report 2 years ago. He studied the issue. Everything’s in there. Just read that report and do those things.

You’ll notice what’s not happening is any of those things. [laughter]

When you think about all the people you’ve met who have been heads of state or in positions of political power, which one most compelled you to feel that you wanted to invest with them or work with them?

In the last 5 years, I would say it’s the heads of state, in particular in the UAE, Saudi Arabia, Qatar, and Kuwait. There is something really special happening in those countries. I find there are a lot of very talented people—politicians, European politicians, heads of state, former heads of state—where, when you get them in private, they know everything.

They know what needs to be done. You just pick a topic, and they know what needs to be done. It's almost like the policy discussions have been had so many times that we kind of know all the answers already. It's just that we either like or don't like the answers, and specifically, of course, we like or don't like the consequences of the answers. We don't like the trade-offs.

I think there are a lot of people who know, okay, there is a formula; there are a set of things to be done. It could be on this question of having a tech industry, or it could be on some other pressing issue—fiscal, whatever the issue is—and they know what it is. Then they explain, “Here's why we can't do that,” and they go back, and then they go out in public and half-pretend that they don't know what the answer is.

I don't know whether they view that as encouraging. I guess the encouraging thing is that I think the intelligence level is probably higher than it looks. The discouraging thing is that the courage part of it is probably not quite there.

And then they get unelected, and then the cycle starts again. Well, so there is this—I mean, just to talk geopolitics the whole time—but there is this really fascinating, as you know, big difference: the American constitutional system versus the European parliamentary system.

There is this thing where, when an American president becomes deeply unpopular, he sinks down to around 40% approval. When a European politician becomes unpopular, he gets down to around 6%.

Yeah.

They start at 40—no, straight to 6. I don't know. I always look at that and I'm like, wow. If you know that your default path is to go from 40 to 6, maybe it's time to try something different.

We did have a prime minister once where the whole nation was betting on whether, I think, a potato would last longer than her in office or not. It was a legitimate prediction marketplace. A lettuce—and, by the way, it was livestreamed. We needed proof of death of the lettuce.

Yes.

So there we go. We brought up the future of Europe and whether you need to be in Silicon Valley because of AI. When I look forward to how this plays out, when you project forward, do the gains in AI look like AWS in terms of infrastructure dominance, or do they look like the internet in terms of application-value dispersion?

Yeah. So, actually, let me give you a broader answer. I think it's a great way to come at it. I think there's actually a broader answer to the original question.

If you're talking about concentration in Silicon Valley, the question that I answered before is about the mainline companies building AI: Google, OpenAI, Anthropic, Meta, xAI, and NVIDIA. So, Silicon Valley, right? That's true for sure. But I think there's a second phase to it, which, again, I'm very excited about. The second phase, which relates to your new question, is that I think the benefits of AI—the power of AI—diffuse out globally to a degree people are really not expecting.

12. "Stop Chasing Diamonds in the Rough": Why Most VCs Get This Completely Wrong

Furthermore, I think that's already happening. This is also an answer, by the way, to your inequality question, because the assumption always is, well, surely the biggest companies in the world will have access to the best technology, or rich people will have access to the best technology, or whatever. It's actually quite striking: If you look at AI, I think it's the most hyper-democratic, small-D democratic technology we've ever seen.

It follows the internet and smartphones in this, which is why I'm pretty confident this is what's going to happen. I think it's already happening, which is that the best AI in the world is the app that you download on your iPhone off the App Store, right? The best AI in the world is OpenAI, or whichever one it is of the 3, 4, 5 that are really in the race. You download that app; that's the best AI, right?

And, by the way, to your inequality point, you're probably going to have to pay $20 for it. Then, if you really use it a lot, you're going to have to pay $200 for it. But the free ones are pretty good now. Google gives away a lot of AI value for free, Microsoft is starting to do that, and others are doing that as well.

The best AI in the world is the consumerized version that's available to everybody. I think there's a part 2 to our earlier conversation, which is that people all over the world—and it's already happening, because these apps are about to cross 1 billion users and they're growing fast—are not that many years away from having AI running on the 5 billion people who have smartphones and internet access.

I think that's such a hyper-democratization of the technology. The use of AI—the consumer benefit, the business benefit, the economic benefit—I think that has the potential to be decentralized to a radical degree. Let me pause there, and then I'll answer your question of what this means as an investor.

So I guess the question is really: to what extent do we feel it is a just assessment that the models will move into the infra, move into the application layer, and erode value? Anthropic announced a security update. I'm using that as an example because it's ridiculous in my eyes. They announced a security update, and CrowdStrike and Cloudflare tanked 8–9%. Obviously, it's not threatening CrowdStrike and Cloudflare today.

Do you think the core models—OpenAI, for example—will move continuously into the application layer and consume more and more of the value chain?

Yeah. So, a couple of things. One is there's a bigger phenomenon, which is what I was heading toward in my earlier answer. There's a bigger—there's actually an even bigger phenomenon than that. There's actually a paper on this; maybe we could link to it. It uses a term that sounds like “Schumpeterian economics,” after likely Joseph Schumpeter, the economist who developed the theory of creative destruction.

This economist basically goes through and says there's a concept of Schumpeterian economics, Schumpeterian gains, and the idea basically is—and he does this whole analysis for a whole bunch of different technologies—that when there's a new fundamental technology, whether it's electricity, steam power, computers, the internet, smartphones, or AI, what actually ends up happening is that something close to 99% of the economic value arrives in the market not in the form of economic benefit to the companies that make the thing, but rather to the customers.

Economists call this consumer surplus. Consumer surplus is all of the benefit that consumers are getting that they're not actually fully paying for.

The way this analysis basically works is, if you look at the total amount of economic value creation, for example, downstream of the internet, something like 99% of that occurs to the users of the internet, not the companies that built the internet, right? Same thing with the smartphone, right? Who gets the economic value of the smartphone? Everybody in the world who uses a smartphone to become more productive in their life or in their business gets 99% of the value from the smartphone. Apple and Google get 1% of the value from the smartphone.

I think AI is the exact same. I think it's already that way. I think it's going to be exactly the same way. It might even be greater than that. It might be 99.9999% of the value of AI that accrues to the users, not to the companies that make the AI.

I bring that up because I think that's such a larger economic force. That's such a larger amount of value that's just extending out into the world that, like I said, it's almost like dark matter. Everybody's going to experience that in their own life and in their own business, and everything that they build, wherever they are in the world and they're using AI, and nobody's ever really going to tally that up or get credit for it. But if you do the analysis, it's going to turn out that that's overwhelmingly where the gains are.

13. Why Labour Displacement Theory Around AI is Totally Wrong

So, as a consequence, your question is basically a question of then fighting for the 1% that stays captured in the AI industry itself, which is a very important question, of course, and is central to what we—

Well, I guess the question is: does that whole economic theory change when we believe that we will see the labor being eaten? When, actually, software spend is no longer software spend, it moves into human labor spend, in which case the TAMs explode and we have bigger companies than we could ever have, but a Harvey of the world actually eats a large part of legal work and junior lawyers.

My girlfriend's a junior lawyer, so, yeah, don't kill me, but you're gone. Does the TAM explode? And how do we feel about that?

Yeah. So, you have friends, I'm sure, who were great coders before AI and are now using AI for coding. What's the thing that they all report?

They're far more productive.

They couldn’t live without it. And are they working more or fewer hours than before?

More.

More. Yeah. So, this entire labor displacement thing is 100% incorrect. It’s completely wrong. It’s classic zero-sum economics. It’s the lump-of-labor fallacy.

It happens over and over and over again. It’s always been wrong. It’s going to be wrong again.

Leave it for mediocre people. I know that sounds very judgmental and horrible, but most social media managers are crap. Okay, I’m getting in trouble for this—not you. They’re crap. If you get a social media tool that is AI-driven and can replace an average social media manager for AT&T, surely you’d do it.

I don’t say this to be insulting, but it’s the classic Marxist analysis, right? There’s a certain amount of work to be done, and either the machines do it or the humans do it. Surely those jobs go away.

The answer has to be—and this is what technology has always done, and this is what AI is going to do—and this is why I went through the long description that I did of the hyperdemocratization of AI: every single one of those people who’s a social media manager today now has AI. They all have AI. They all have AI, or they’re about to have AI, and they’re going to have it at their fingertips.

If they want to, then anything that they want to do in their life, in their work, in their career, in their profession, in their job for the rest of time, they’re going to be able to use AI to do those things. They’re going to be able to use AI to become a better version of themselves. They’re going to be able to use AI to learn new skills. They’re going to be able to use AI to become more productive at work.

They’re going to use AI to not do a lot of the grunt work they’re doing today, so that they can do higher-value work. Now I’m just talking classical economics, which is the other side from Marxism. Classical economics says that the actual economic function of technology—and this includes AI—is to raise productivity, and specifically to raise the marginal productivity of the individual worker.

Again, this has happened many, many times. You take an individual worker who used to write with pencil and paper, and you give them a typewriter. Then they used to write on a typewriter, and you give them a word processor. Then they used to do hand accounting, and now you give them a spreadsheet.

14. Are Companies 75% Overstaffed? The Most Controversial Take on Hiring

By the way, social media manager—a job that didn’t exist before the internet, right? Technology creates new jobs.

Maybe I’m a European communist, but then why are we seeing layoffs? Why are we seeing layoffs everywhere? Why is every CEO I’m meeting saying, “We’re flat on headcount, or we’re reducing”?

Oh, that’s very easy. Number 1: interest rates. Interest rates were at 0, and then they went from 0 to 5% at record speed, like 3 years ago. Every company had to replan all of its financials. Its cost of capital went up 5 points. They all had to completely replan their financials.

Number 2: they all overhired during COVID. The hiring binge the companies went on during COVID was wild. It was the combination of the 2. It was interest rates going to 0 during COVID, and then it was the complete loss of discipline at all these companies when they went virtual, when employees just became an icon on a screen, and they just said, “Hire tons more of them.”

What you have happening right now is that essentially every large company is overstaffed. We could debate how much, but it’s at least overstaffed by 25%. I think most large companies are overstaffed by 50%. I think a lot of them are overstaffed by 75%.

Now they all have the silver-bullet excuse, right? “It’s AI,” right? I know this for a fact because, number 1, I talk to them. But number 2, I know this for a fact because AI, until literally December, was not actually good enough to do any of the jobs that they’re actually cutting. It just can’t have been AI.

The other thing is that people look at the hiring rate for new hires, and they look at the spike in how hard it is for new college graduates to get new jobs. Again, people peg that on AI. But I think that’s actually 2 things.

Number 1, of course the companies that overhired and overinvested and have to bring down their spend and their headcount obviously aren’t going to hire very many people. That’s part of it.

The other is that one might make the observation that maybe the skill set of a lot of college graduates over the last decade doesn’t necessarily match the job market, right? That’s a very uncomfortable conversation for people to have. But I think that also has an effect. If you talk to any employer, they’ll immediately tell you that.

Final one before we do a quickfire. You are probably the best copywriter of our time: “It’s Time to Build,” “American Dynamism,” “Software Is Eating the World.” I picture you in this musky room in the American countryside, billowing out as you come up with these titles. What is your copywriting process?

It’s the culmination of raw frustration.

The romanticism of my imagination, but keep going.

It’s Mount Etna. It’s the Mount Etna exploding phenomenon. It’s basically always when I literally can’t take it anymore. I just think that fundamentally, people are thinking the wrong thing.

It’s always the old joke: somebody’s saying something wrong on the internet. It’s that, extrapolated up. It’s when I think there’s a fundamental misperception in the world, and it’s just not correct. Then, of course, I have a sufficient ego to be able to say I can correct that.

Usually, that’s it. The actual drafting in every case has been 2 hours. It’s just, “Rip it and go.” But it’s because I spent the preceding 2 years getting increasingly frustrated.

I don’t know about you, but do you have an internal monologue? Do you talk to yourself in your head?

Yeah. Are you kidding me? All the time, especially when I run.

Exactly. Right. So, what happens is—I’m sure you’re probably like this, too—I’m just arguing with myself all the time. By the time I write, I’ve been arguing with myself in my own head for 2 years, trying to figure out what the good arguments are, and then it all just kind of comes together. I just drop it on the page.

I asked Doug Leone this, but I’m intrigued because you have the same challenge. The weight of your voice is so significant. How do you ensure that people will fight back when the weight is as great as it is?

Number 1, let’s say it’s nice. There’s an upside to it. I don’t want to lose the upside. I do like the upside to it.

But, yeah, look, the very specific form of that is—I think there are actually maybe 2 problems. There’s the giving-advice part, and then there’s the just-asking-questions part, which is also a problem, because people will interpret the questions as advice or directives.

The way I think about it is, if I’m dealing with one of my partners at the firm, or if I’m dealing with one of our portfolio CEOs, I actually have to be really careful to say, “Look, I don’t know what the right thing to do here is. I don’t have the information that you have. I don’t believe I can dictate what this is.”

I used to use an example. Do you remember the concept of an in-flight magazine? Does that ring a bell?

No, but tell me.

In the old days, before phones and tablets, when you took an airline flight, there would be a magazine from the airline in the pocket of the seat. That’s what everybody would sit there and read if they didn’t bring anything. It would be the Southwest Airlines in-flight magazine.

The pejorative was, like, “in-flight venture.” It was basically board members who gave advice by way of in-flight magazine. They flew in for the board meeting, read the magazine, and the magazine said, “Java is going to be a big thing,” so they said, “What’s our Java strategy?” Or the same thing for every other new thing that came along.

Maybe the current version of that is whatever I read on X yesterday, whatever I saw on a YouTube video, or whatever I read in the newspaper. You do have to be really, really careful, I think. As you get more senior in this field, you have to be really careful both in your firm and with founders.

God forbid telling them what to do. A: telling them what to do. B: suggesting what they do, which is sort of the same as telling them what to do, which is dangerous. And then C: even just asking questions becomes very dangerous, because they interpret the questions.

And so you just—I think you just have to acknowledge that up front and bend over backwards and say, “Look, this is genuinely not what I’m trying to do. I’m just going to ask questions,” and do that.

Generally, the way that plays out at our firm is, as I said earlier, Ben and I almost never weigh in directly on an investment that one of our partners is working on. The reason is that we don’t want that warping effect to take place, and specifically because we know we lack the knowledge to be able to do that.

In particular, maybe obviously, doing that in public is especially dangerous, right? If there are other people around, then there’s perceived social pressure. So if we’re going to have a difficult conversation with somebody or really question something, we have to take it one-on-one and be very careful about how often we do that.

We’re going to do a quick-fire round, and we’re going to start with an easy one. Adam Neumann and likely Flow was a controversial deal. Why did you do it? What was the thinking behind it?

At the height of the WeWork meltdown, when it was in the newspaper every single day and reaching its endpoint, I talked to a friend of mine who is one of the legends of the real estate world. I won’t name him, but he’s a very credible, very famous real estate guy.

He said, “Look, whatever people say about this whole thing, there are only 2 people in the history of the world who have built compelling brands where people care about the brand, care about the name on the building, for commercial real estate. In the history of the entire world, 1 of them is the president of the United States, and the other is Adam Neumann.”

He said people need to understand that, yes, this one is going sideways now, but this guy is a generational or all-time talent in that industry at doing that. Of course, it’s not just the brand, but the value proposition—the thing that’s underneath that.

That really stuck with me, because it was up against the absolute wall of negativity at the time, when people were tripping all over themselves to say the worst possible things they could about the guy.

Then we got to know him after that and, as you know, became thoroughly convinced and reinforced our view that he was a generational talent. I think we feel very strongly that that is the case. We’re very happy with that investment.

What was the most controversial deal, or most disagreed-upon deal, internally, from your memory?

I don’t think we have individual deals that are really controversial internally, because the deal we make with all of our investing partners is that they all get to take risks. They all get to go out on a limb and do the things that other people are going to think are dumb, so they don’t generally backbite each other on that.

I don’t think it’s really that as much. The bigger issue, I think—and I put this more on Ben and me than anybody else—is, “What are the kinds of investments that we do? What sectors are in and out of the strike zone?”

I’ll give you an example. The most straightforward example is the deal we didn’t do that we should have done: a likely Anduril Series A. It was just obvious that it was going to be special. Palmer—we had worked with Palmer at Oculus—and his colleagues were clearly very capable. It was just kind of obvious that there was something very special there.

But the politics and the cultural elements of it at the time when it first came around scared us off in a way that I very much regret. You’ll notice that we are now extremely enthusiastic investors in defense tech and in things involving law enforcement, national security, and public safety. We would 100% not make that same mistake again.

I think it has to do with us. It’s risk-taking at the conceptual level, beyond the level of an individual company. As I said, it’s generally been Ben and me when we’ve screwed that up.

You sit down with your kids and can tell them 1 thing that you think would make them the most proud about what you’ve done. What would that 1 thing be?

It’s the impact on the world, in the form of what I described earlier: the economic idea of consumer surplus. Conceptually, it’s just, “Wow, stuff that I worked on or built or helped build is really all over the world, and people all over the world are using it, and it’s been tremendously—on net—tremendously beneficial.”

I think that’s 1. The other thing that rises in importance over time is just the number of people that hopefully I’ve been able to have a positive impact on—the number of people I’ve been able to help or support, help get through hard times, or teach different things to, who have gone on to be very successful.

As time passes, I think it’s more of that second category.

15. First Meeting with Mark Zuckerberg

Penultimate one: What was the most memorable first founder meeting you’ve ever had? Not the best founder or anything like that—just the most memorable first founder meeting.

My first meeting with Mark Zuckerberg. It was amazing. Mark was 19 or whatever, and it was Mark and Sean Parker. I knew Sean a little bit, but not well, and I’d never met Mark before.

Sean talked the entire time. Sean literally talked the entire time. It was just talking a mile a minute—every idea. It was absolutely amazing. Mark didn’t talk, and Sean and I basically talked the whole time.

Mark sat and listened, and I walked away and thought, “Wow, that was really weird.” I thought, “One of 2 things has happened here. Either he’s completely unsuited for the job because he literally doesn’t talk, or he’s listening to and absorbing everything that people are saying around him, and he’s going to be on a vertical learning curve like crazy because he doesn’t have the ego need to just say things. He can just absorb.”

Of course, it turned out to be number 2, which is—and I’ve talked about this before—he’s just on this incredible learning curve and has been his entire life, in the most amazing way.

I would say that one. I’ve never told that story before, but that was memorable.

I love that.

The second meeting, I got him to talk. By the way, everything Sean said was right, and it was all genius.

I would have loved to have seen that final one. You’ve been an incredible entrepreneur, you’ve been a great investor, and you’re also an amazing firm builder. If I were to push you—and one of the greatest investors suggested this one, but I can’t tell you who it was—if I were to push you on which one you’d most like to be remembered for in history, what would it be?

What were my choices?

Entrepreneur.

Firm builder or investor?

Firm builder or investor.

Yeah, I think entrepreneur. Ben and I are lucky in that we’ve been able to—you know, a16z itself has been an entrepreneurial project. If I could choose, that would definitely be the one.

Marc, I cannot thank you enough for doing this. As I said, I’ve wanted to do this for 10 years. Thank you so much for joining me.

Awesome. Thank you. I really enjoyed it. The questions are fantastic, and you’ve been doing an incredible job. I also really appreciate the chance.