Peter Thiel being contrarian & right, AI, Elon, drug reform, overrated trends & wealth polarization

Today's episode of This Week in Startups is brought to you by Walker Corporate Law, a boutique law firm specializing in the representation of entrepreneurs. Visit them at walkercorporatelaw.com. And by PagerDuty, serving as the hub of your operations, aggregating all of your infrastructure, monitoring tools, and alerting the right people and teams at the right time. Sign up today at pagerduty.com/twist and get a free t-shirt with your first alert.

Hey, everybody. Today, a special treat for you on This Week in Startups. Peter Thiel, who sat down with me at the LAUNCH Festival 2015. 12,000 people in attendance, three days here in San Francisco. And many people told me this was the highlight of the event for them. Peter Thiel, of course, is the venture capitalist and founder behind Palantir, PayPal, and Facebook. He invested in Facebook when it was worth $3 or $4 million. He is extremely candid, honest, and challenging during this interview with me. In fact, multiple times during the interview, he told me I'm asking the wrong question or that my question was inane. I mean, it was challenging for me as an interviewer. And you're going to get to see me go toe-to-toe with Peter Thiel, one of the great, great investors and thinkers in the technology and finance space.

You're going to love it. Get a paper and pen out. There's a lot of knowledge in this episode.

It's what it's all about, man. Money is the root of all evil. Funny how it feeds my people. We ain't gonna live like equals until we get the money, spend the money, and defeat you. Money is the root of all evil. Funny how it feeds my people. We ain't gonna live like equals until we get the money, spend the money, and defeat you.

Good afternoon, everybody. Good afternoon. What a glorious day we have out there in San Francisco. The weather is outstanding, and you're all inside, but you're inside for a good reason. The next speaker is one of the most considered, intelligent, successful minds we have in this generation of Silicon Valley, and I think that's... enough of an intro. And let me just take a quick moment to thank Unilever, Foundry, RingCentral, and Icehouse. Big round of applause for our founders—sponsors, rather. Thank you for that. I always appreciate when you guys put a good hearty welcome. And Peter Thiel, come on out. Please welcome Peter Thiel. Thank you. Oh, a lot of Peter Thiel fans. Peter, have you ever seen 2,000 of your fans in one place?

Is this the largest on the book tour so far?

It's like this is over 2,000. Well, it depends a lot on what context. So, 2,000 people here is better than 2,000 people in Orlando, Florida.

Yes. Yeah, no offense to Orlando. No offense to Orlando at all. But any business plan you get from Florida, you delete immediately, correct? That's what I was told when I started angel investing. If it's from Florida, be very suspect.

Well, it's always dangerous to have absolute rules, because we once went, at Founders Fund, we once went through a list of absolute rules, and there were about 20 or 30 of them, and they add up, and eventually you don't look at anything anymore. So there are always good shortcuts. Too many shortcuts somehow goes wrong.

Would Facebook or Palantir have made it through those 30 rules, I wonder?

You always hope so in retrospect. I think Facebook would have made it through regardless because it was at a point when a company like Facebook today would probably be valued not at $5 million but more like $100 million, having the same metrics when I invested back in 2004. You invested in Facebook at a $5 million valuation? Yes. Yeah, but they had 100,000 users. It was growing really fast. They only needed money to buy more computers because there was so much demand for the service. So that's actually always a pretty good sign.

And Sean Parker introduced you to him, to Zuckerberg?

Sean, Reid had looked at it a little bit, so there were a few different points of contact.

What was the first meeting like?

You know, it was not like a show on Shark Tank. Zuckerberg was 19 years old. He was still pretty introverted. Sean did most of the talking. So, if you base it—people always exaggerate how important these pitch meetings are. Reid Hoffman and I had spent about a year looking at all these social networking sites before that. So we were ready to write the check before we met and it didn't really matter what people were going to say.

And what was it about that business that made you so convinced it would be a game changer? Because I remember at the time, we weren't all convinced. There was something you guys knew that we didn't know.

Well, a few different pieces, but I think there was an intensity of usage that was already very, very big. It was very promising. It had all these network effects, which I think are never to be underestimated. There was something about the college market that was probably being underestimated. I think investors always have a bias to invest in things they themselves use, and they undervalue things they don't use. So there aren't very many investors who are in college. So anything that's just big on college will be somewhat underrated systematically. It was a bunch of different things that added up like that.

And take me to Palantir, because this is another one of the major companies of our generation, of this last wave. What is that? It's worth $10 billion or something in that range? It's something maybe a little north of there. A little north of there. And you were an angel investor-slash-co-founder of this company. Yes. And how much was that company worth when you started the investment?

Well, we did—we did the first angel round, like, $800,000 valuation. And then, you know, we did $800,000 from there. So—but you don't angel invest anymore. You just do venture capital now. You know, I don't know if there's really sort of a bright line. We do everything from seed to Series A, B, C, D, all the way up the spectrum. I would say my single overarching idea, both in my book, Zero to One, and in terms of investing, is that I believe—I think most people always tell you, most business books, most business advice, always how to be better at competing, how to compete more effectively. And I believe that the key is not to compete at all. The key is to do something that nobody else is doing.

The politically incorrect word for it is that you should have a monopoly. We can debate at what points monopolies are good or bad for society. But from the inside, if you're a founder, early investor, early employee at a company, you always want to aim for monopoly. And so I'm ready to invest at whatever point I'm convinced it's going to be a monopoly.

And Facebook, you felt like this is going to be the dominant social network and there'll be only one major one?

You know, I'd like to say that I had that much figured out. I thought it would be dominant on college campuses and that would be worth a lot more than five million. I didn't really think it was going to be quite as big as it turned out to be.

And there are some pretty big fights going on right now. I can think of Android versus iOS. This seems like a fight worth having. Is that a fight that Google shouldn't be in or Apple shouldn't be in? Too soon to tell?

Well, it's pretty core to Apple. I would argue it's not that core to Google. I mean, Google is fundamentally just a search engine, just a search company. That's where 98% of the revenue comes from. And then I think Google always likes to brand itself as an all-purpose technology company. They're driving self-driving cars. You have Google Glass, phone operating systems, balloons in outer space, but really 98% of it's a search engine. Everything else is probably a distraction.

And the fight with Microsoft you talk about in the book became a huge distraction for Microsoft. Do you think that the apps business, the Gmail apps business, has been a distraction for—and now they have an operating system, Chrome—do you think that business at Google is not worth them spending too much time on? They should just keep with the search?

It's always complicated. I mean, sometimes you do have to fight. Sometimes it's existential and you have no choice. But I think we're always too prone to it—too prone to find conflicts with other people, with other companies. We're too prone to thinking what's valuable is whatever the other person's doing, and so we have to fight them for that. And I think it's almost always a better question to ask, what can I do, what can my company do, what can our company do that nobody else is doing, versus what is somebody else doing that's cool that I want to take away from them? That's, I think, generally a much less constructive approach.

Ah yes, the Walker Corporate Law Group is an amazing group of attorneys with 10 to 20 years of experience and they do fixed fees. What does that mean? That means you don't have to be scared when you open up that PDF from your attorney every month. Oh my God, is it going to be $3,000 or $300 or $30,000 or $15,000? That's agita, that anxiety I have lived with my whole life. But now, fixed fees. What does it mean? I need to incorporate, I need to get a trademark, I need to do an employment agreement, I need to do M&A. They're gonna give you a flat rate and they're gonna say, hey, this is about what it's gonna cost and here you go, we'll just do it for this amount of money. And Scott and his team, they left all those big, big firms because they wanted to do something small and intimate.

And they spend so much time with founders. I've had so many founders from hackathon teams who have IP issues, like, oh, who owns this? Two of the hackathon members don't want to do this anymore. Two of them want to make it into a company. What do we do now? Oh, scary. Scott makes it not scary. You can email him, scott@walkercorporatelaw.com, scott@walkercorporatelaw.com. Walker, W-A-L-K-E-R, obviously. walkercorporatelaw.com, 415-979-9998, 415-979-9998. Sometimes Scott Walker picks up the phone himself, and he often talks to entrepreneurs for hours who come from This Week in Startups or from the Launch Festival. He's a true mensch, and... you know, you can't go wrong with Scott Walker. Everybody I've sent to him has just come back to me and said, NPS score 10.

Very likely to refer a friend to Scott Walker. That's why his business is booming. And he's chosen this walk of life. He only wants to work with entrepreneurs and help them have their vision become reality. And for that, I salute you, Scott Walker and the Walker Corporate Law Group. Email scott@walkercorporatelaw.com. Okay, let's get back to this amazing episode. I think the other core tenet of the book is really, what do you believe that is true that nobody else believes to be true, I guess?

Is that the right way to say it? Yes. It's always a combination of contrarian and true. People always characterize me as contrarian. I think that's misleading. One plus one equals three. That's a contrarian belief. It's not interesting, untrue, won't get you anywhere. So it's always important for it to be something unconventional no one's thought of, but that also has, you know, either is intellectually true or has some merit in the business context.

And you're huge investors in SpaceX, Elon's second company, I guess, in this recent crop. What is it about that business that he got right and sort of how does he run through the Thiel equation as it is?

Let's say, well, it was definitely both Tesla and SpaceX were companies where the competition was really weak. So, you know, you were competing with these decrepit car companies in the U.S. that had barely innovated in 50 years. With space, it was these really bad aerospace, half-government, half-military-industrial complex type entities that were, again, sort of very bad at innovating, very inefficient. So if you're going to start a search company, that would be tough, because Google's a really tough company to go up against. Going up against General Motors, going up against Lockheed, you're at least picking pretty weak competition, and that was an important strategic thing to get right early on.

There were a lot of challenges in building these companies, but that part of the strategy, I think, is very underrated.

The fact is, nobody believed electric cars were going to be viable, or very few did.

I remember a conversation I had with Elon in the summer of 2008, and I asked him, when was the last successful car company started in the U.S.? And the answer was Jeep in 1941. And so it was sort of the last half-full, last quarter-full version was about time after 67 years.

How do you feel about where Tesla's gotten as a company, and where do you think they're going to go?

Well, it's been an incredible success. I would never bet against Elon in anything. So I think that's sort of, you know, hard rule number one. It's probably, you know, if you sort of ask, what is it that the innovation, what's been the innovation in SpaceX or Tesla? There are many different forms innovation takes. The most common one we're used to is sort of, you launch something and then you iteratively improve. It's sort of this continuous improvement model. We occasionally have a modality of innovation where you have a big, brilliant breakthrough, like maybe Bitcoin, where someone worked on it in a closet for 10 years and then released this amazing discovery to the world. But I think there's another modality of innovation that's very underrated, is what I describe as complex coordination, where you just take a lot of different pieces, and the innovation is to combine them in a new way.

So if you ask, what is new about SpaceX rockets or Tesla cars? All the components already existed, at least in the initial designs, and the critical thing was to pull them all together in a new form. And that sort of vertical integration is somewhat capital-intensive, which is why it's quite hard to get it financed, but it's done very little. And then once it's done, it's something that tends to get underrated. I think Apple's iPhone, the first smartphone that really worked, is another example of complex coordination. When people write biographies on Jobs, it's like, 'Well, he was a jerk, and that's the most interesting thing.' What really is interesting is how was he able to motivate all these people to build this completely new product, and it was that you could do something incredible not by inventing anything

specifically that was new, but bringing all the pieces together in just the right way. And that's what the iPhone represented. That's what Tesla represents with cars or SpaceX with rockets.

And what is the, no pun intended, the trajectory of SpaceX, do you think? Because he's now gotten to space many times, gotten to the space station many times, and he's almost landed the reusable rocket. He's tried twice, and it's apparent he's going to get it.

And that's truly innovative. Yes, well, I think having built this vertically integrated rocket company, it's now possible to innovate on that in ways that would be much harder if you had this super-complicated subcontractor, sub-subcontractor system where all the components are bespoke and you can't innovate on the whole thing. So having integrated all the pieces, you can innovate. If they get to reusability, that is an enormous breakthrough because it would cut costs 70, 80%.

Cutting the cost 70, 80% to get to space.

If every time you flew an airplane, you had to throw away the airplane, air travel would be kind of expensive.

Yeah, a magnitude more. Well, let's talk about some of the other people that you had at PayPal, because I think what's fascinating is you were responsible, I believe—you weren't responsible for hiring Elon, you were responsible for merging PayPal and X, but you hired Reid Hoffman, David Sacks, Jeremy Stoppelman, Dave McClure... They got fired. Who else?

We had Chad Hurley from YouTube. We had a slew of talented people.

Was it the experience of working at PayPal—create that group of people? Was it just a moment in time? Or was there something about the selection criteria in which you used when you were—how many years old? How old were you at that time?

You know, I was 31 when I started it, 35 when eBay bought it. The PayPal period was a very compressed four years from start to the eBay, less than four years from when we founded the company to when eBay acquired it. I don't know, I think it was probably a combination of different things. I think it was a relatively entrepreneurial, somewhat chaotic culture. We had a lot of very strong personalities. I think a lot of companies biased towards having people who sort of more just drink the Kool-Aid. And there's sort of pluses and minuses to both. You'll have a more smoothly functioning company. You'll maybe have less dissent when things are going wrong. So we were sort of less smoothly functioning, but a lot of people did feel ownership of the product and would raise their voices quite a bit if they felt things were off track.

Certainly, in retrospect, 2002, 2003, 2004 were very good times to start new companies. It was sort of a contrarian time to start an Internet company. Sort of the macro timing was quite good. But I would also say that the lesson, the overarching lesson you learned going through a business like PayPal, which had a lot of challenges but ended up succeeding, the high-level lesson was: you can build a great company, but it's hard. And I think the lesson you learn inside of most startups is very different. If you're at a startup that fails, you will learn the lesson it's impossible to build a great company. You should try something less ambitious the next time around. If you're at a startup where things work too easily,

which I think was the way most people, say, at Microsoft or Google would have experienced it, you learn the lesson that it's actually easy to build a great company. And hard is somehow better than easy or impossible. Because if it's easy, you don't need to work hard. If it's impossible, there's no point in working hard. And so easy and impossible converge to not working hard. Hard tells you you should have to work hard.

And of that whole group, who do you think, like... I mean, they're all kind of your children, so it's a really hard question to ask, but which one impresses you most of that cohort and why? Or which ones impress you a lot?

You know, I don't like ranking people. Look, I think Elon's super impressive because he's done something outside the IT space, right? And we're sort of in a world that's dominated by IT, by innovation in the world of bits, much less in the world of atoms. And I think that is what makes Elon uniquely charismatic because we do have a sense that it would be good if we were doing a little bit more in atoms and not everything in the world of bits. That being said, I also think it's quite hard to do things in the world of atoms rather than bits. On the Founders Fund side, back in 2007, we thought we should start doing more in biotech, more in transportation, all these sort of futuristic sci-fi-like technologies.

Maybe 20% to 25% of the capital has been invested in that. 75% to 80% is still IT. And there are a lot of very deep reasons why those kinds of businesses tend to work better and why it's actually quite rational for most people to focus on those.

And what have you learned about venture capital now that you've been doing it for so long? Fund four, maybe? Fund five.

Fund five now. You know, 10-plus years. I don't know, there's sort of a lot of different lessons. One that's a little bit of a, you know, it's not quite a precise sort of lesson, but is that, is that people, once something works, people often underestimate it. And when things aren't working, they underestimate how much trouble they're in. So there's a way in which, in most areas of investing, momentum is not that good a way to invest. If a stock's gone up, you don't necessarily want to chase it. When I've done the backtesting on our portfolio, I found that every time a company had a big up round led by a smart investor, it was always a good idea to do your pro rata. Flat rounds, down rounds—it was almost always a bad idea.

The steeper the up round, the cheaper it was. The biggest miss of the last decade was not doing our pro rata or, for that matter, the full Series B at Facebook. It was a 12x up round in eight months. It was the steepest up round in that amount of time in any company we've been involved in. And in retrospect, that was perhaps also the cheapest. Cost us billions. The Series B round of Facebook. And I think one of the reasons it was so underpriced was that investors don't want to step up that much, but then even the people on the inside often underestimate how much things change. So once things start working, once you have that sort of momentum, it was still eight or nine people at the company.

They had this office with this horrible graffiti art on the wall. And so even though you had these abstract charts that were going exponential, on the inside, it didn't feel like things were changing that much. And that's why people underestimate it. And so you have these, these subtle but very important points where somehow the leverage shifts, where somehow the dynamics shift very powerfully, and they tend to get very underestimated.

What do you think about Elon's thoughts on artificial intelligence? Larry Page has obviously bought DeepMind.

Were you an investor in that? Yeah, we were actually seed investors in DeepMind.

So what is it about that company that has specifically created such a polarized debate? Because they were working on some things that spooked some people?

Yeah, we had a—we had an LP meeting at Founders Fund, I believe in 2012, where Demis, the DeepMind CEO, spoke. And I remember one of our investors told me afterwards that he felt it was the last chance he had to shoot him and kill him for the sake of humanity.

Which was the first time for a portfolio company LP to do that—to have that thought. I mean, I'm sure venture partners had that thought.

It wasn't like, 'interested in the company.' He gave a very compelling speech. You'd say, 'Wow, it's going to be a great investment.' But the investor reacted in a somewhat different way. I would say, let me see, I think probably, I suspect the fears are a little bit overdone at this point. I think full, strong AI is still, in my judgment, somewhat further away than people think. That being said, I think it would represent a shift that's really drastic for which we don't have a great conception. And so when people talk about computers substituting for human labor or stuff like that, I think that underestimates the strangeness of AI. It would be like if aliens landed on this planet, the first question we wouldn't ask would be, 'What does it mean for the economy?'

The first question would be political: Are they friendly? Are they not friendly?

Right. And you put AI in that latter category.

I think if you've got full-on strong AI, it would be like aliens landing on this planet.

Ah yes, PagerDuty is a great product that manages one of the most painful aspects of running a startup, which is when the service goes down, who do you call? Who do you wake up? Which developer has to leave their daughter or son's soccer game? Oh my God, it's brutal. PagerDuty is... a reality that we all have to deal with. Our services are 24/7. If things go down, you're going to need to have a product that manages all the alerts, who's on duty. And you know what? People quit over pager duty. It's a very challenging, controversial, internal debate amongst developers, right? And developers have a lot of choices of where they're going to work. Sysadmins have a lot of choices of where they're going to work.

You want to have tools to make it easy, and PagerDuty is that tool. Just looking at the client list here, Atlassian, HipChat, Pinterest, New Relic, Airbnb, Panasonic, Slack, and Path, and us. I mean, it's an amazing product. You do not want to roll your own monitoring system. Customers love the apps. They have an iOS and an Android app, of course, so you can receive alerts there. And one of my investments was in Chartbeat, a great company. And Justin Lintz over there, the senior operations engineer, said this: 'PagerDuty gives Chartbeat one central place to send critical alerts. We now have a simple, easy process for on-call scheduling.' Hey, here's your call to action. Sign up today for a free 14-day trial at PagerDuty.com/twist, and you'll get a free T-shirt when the first alert comes in.

That's right, a free T-shirt when you get your first alert. Go to PagerDuty.com/twist, PagerDuty.com/twist. This is a fantastic product from a fantastic company that really solves a massive pain point, which is being on pager duty is brutally painful sometimes. And you know what? Things go down. It's very stressful. You need tools to make it less stressful and manage it better. Really, like when things are bad, you need those tools to make things a little bit easier and a little bit more manageable so people don't quit on you and so your service is up and running. Thanks again to @PagerDuty for sponsoring independent media like This Week in Startups. Speaking of which, let's get back to this amazing episode.

How far...

In your estimation, as an insider, do you think we are from strong AI? Is it something we're going to see in our lifetimes? I mean, this self-driving car we keep talking about, it's two or three years away. For the last two or three years, it feels like 10 years away maybe. Where do you think we are with AI and even backing into the self-driving car? I'm curious.

You know, I'm going to disagree with the premise behind that question. I think the premise behind the question is that there's this future that's out there, and there are these things that will happen in the future. It's sort of like the Ray Kurzweil 'The Singularity Is Near', and we have these exponential curves that are just happening. What we should be doing is eating popcorn and watching the movie of the future unfold. And I think it's hard to know what's going to happen, but it's the wrong questions. I don't think the future is fixed. And I think what matters is a question of agency. And so if we decide to work on it, I think we could build it. If we don't work on it, it will never get built.

It's going to get built. It's going to get built. I mean, Larry didn't buy it.

He shut it down.

If everybody thinks it's going to get built, then nobody will feel like they need to do it and then nobody will build it and it won't get built.

Okay. I feel like I'm playing chess against Kasparov.

No, look, it's always an agency question. Okay, but hold on a second. We shouldn't think of the future as this thing that's out there. It's the question that one should ask, especially as an entrepreneur or early founders, what is the future you want to build? And then that's a future you should try to work to make happen.

What if the person who builds that future, however, maybe doesn't have the same regard for humanity or trying things, or what if they're just a little cavalier or immature? What if the person who decides, "I'm going to try strong AI," is a group of Pakistani students who get flipped and start working for ISIS or something like that? It sounds crazy, I know, but so did 9/11. I mean, this technology is not going to be reserved just for Stanford grads. What if it gets co-opted and people start to realize, hmm, cyberterrorism is step one, and maybe some strong AI to disrupt the world is another?

Well, look, I'm not a technological utopian. I don't think technology automatically makes the world a better place. I think there has been a powerful negative story about the world at least since Hiroshima in 1945. Maybe you can go back even further. But I think one of the reasons our culture at large is so anti-technological—the easy way to see that our culture is anti-technological is you just watch all the science fiction films, and they're all about technology that kills people, that doesn't work, that's dystopian, and the future is some combination of Matrix, Terminator, Avatar, Elysium. I watched the "Gravity" movie a year or so ago. You never want to go into outer space. You want to be back on some muddy tropical island somewhere.

I don't like this pessimism about technology, but I respect that there are reasons that people have it, and I think they go back to all the problems of the nuclear age. So I think it's part and parcel of that. I'm somewhat hopeful that the really crazy people are too incompetent to do things, but it's not absolute.

Yeah, the crazy people have their moments, but you're generally right. The criminals tend to get caught because they're stupid. But it's not absolute. But it's not absolute. We can't put that on the absolute Founders Fund list. What about jobs, efficiency, robotics? This is a very interesting area as well. Working less or working what you want to work on could be seen as a positive, right? So having to have less jobs and having more pursuits in a Star Trek kind of way to evoke a more positive and less dystopian, more utopian worldview, it seemed like everybody there was pretty happy with the fact that they didn't have a job, they had a vocation or a hobby. Is that where we're going to head as a society now?

I mean, automation is getting pretty severe and it's happening at a rapid clip. There's less people working at the Tesla factory, self-driving cars, et cetera. And we already have large unemployment in a lot of places in the world. As the world goes flat and there's less employment, what's the world gonna look like? What do you think are the positives to that, the positive solutions to that?

Well, let's see. I do think people have been too worried about technology replacing jobs for a long time. I mean, you have the Luddites in the 19th century, where you were worried that, you know, there'd be nobody left to work in textile factories and to try to break the machinery. And it turned out that a lot of the automation and productivity gains freed people up to do other, you know, more productive sorts of things. So I think if we had enormous productivity gains, enormous GDP growth—it might not be perfectly evenly distributed, but if you had GDP grow by 4% a year, I think just about everybody would get better off. I think the challenge is actually that in most sectors, we're not having anywhere near those kinds of productivity gains.

There's been a lot of automation in manufacturing, but you have to keep in mind that that's a smaller and smaller part of the economy. So even if we're improving manufacturing at the same rate as we were improving it 100 years ago, let's say we're increasing efficiencies 10% a year, that makes less of a difference if manufacturing's 15% of GDP versus, say, 50% of GDP, as it might have been something like 100 years ago. And what our economy is more dominated by today are service sector jobs, which actually have not changed that much. So people working in restaurants, people working as kindergarten teachers, people working as nurses or medical technicians. And this is actually why I think there's less economic growth than we think, because we've not actually figured out ways to really increase productivity in so many other sectors.

So in general, people always frame the worry about technology: it's happening too quickly, it's too disruptive, it's too dangerous. And what I worry about is much more the specter of stagnation. Things are not changing quickly enough. We don't have enough growth. And then when you don't have enough growth, that's where I think you have real challenges. In a world without growth, everything becomes a zero-sum game. I'll only get ahead by you falling further behind, and that becomes a much nastier world.

When you look at the various sectors you're investing in, which one do you think has the most potential in the coming years? Healthcare, robotics, AI?

You know, I'm always skeptical of sectors and trends. People always ask me, you know, what are some trends that I see happening in the future? And I never like the question because, again, it's like I'm not a prophet. I don't think the future is fixed in that sort of way. But I also think, at this point, I think all trends are overrated. And so, you know, healthcare IT, education software, somewhat overrated. Software pretty badly overrated. If you hear the words 'big data,' 'cloud computing,' you need to run away as fast as you possibly can. You need to just think fraud and run away. And the reason you want to be really careful about these buzzwords: so if I said, you know, 'I'm building a mobile platform for SaaS enterprises to do big data in the cloud'...

And if you have sort of like a proliferation of buzzwords... Can you introduce me to that company? I know you'll invest without asking any questions. If you're investing, yes. I'll have them call you collectively. But all these buzzwords, they're a tell, like in poker, that the company's bluffing, and that it's undifferentiated because we've heard the buzzwords before. And so if you're the nth company in a category that's well-established, that's problematic. You don't want to be the fourth online pet food company, or the, you know, or the 10th thin-film solar panel company, or the 1,000th restaurant. So it's harder to describe because there's no buzzword that's good. Yes. So I would say, yeah, so I think the things that are...

Conversely, the things that are underrated are the ones where there are no buzzwords, where it doesn't actually fit into any pre-existing categories. And then you have to always be open to listening to those. And of course, the challenge is that very often, I think, even the people who are running these companies will describe them in terms of existing categories, because that's so much easier to do. So Google would have been described as a search engine in 1998. People have said, 'Why do we need another search engine? We already have 20 in that category.' And what was the first machine-powered search, and the PageRank algorithm was actually the key differentiating thing. And if you simply labeled it as a search engine, that would have obscured all the key differences.

Or if you described Facebook as a social network circa 2004, even circa 2015, that's a misleading categorization. There have been many social networks before Facebook. Reid Hoffman, my friend who started LinkedIn, started a company called SocialNet back in 1997. They already had social networking in the name of the company. And it was sort of these avatars in cyberspace, and you'd be a cat and I'd be a dog, and we'd have these weird interactions on the internet. And it turned out... So, no different than last week when... People aren't interested in networking in the abstract. They're interested in real identity. And real identity was the key innovation that Facebook had. It was the first one to crack real identity.

So I think often figuring out the correct way to think about things in categories for which we don't even have the proper language is really critical to do.

Let's talk about entrepreneurs, young entrepreneurs coming into the space. What do you think the keys are for them today in 2015? Because this is a much bigger industry than when you and I came into it. It's a lot more noise. How should they look at the entrepreneurial process and just starting their career? Should they go work somewhere like Palantir or Facebook and get some time under their belt? I know that you have strong feelings about college, or you're rumored to have strong feelings about college. If you had a son or daughter who was 18, what would you advise them to do?

I probably would still go back to college if I had to still do it. It's because I didn't have any ideas of what to do instead. If I had to do it over again, I'd think a lot harder about why I was doing it, not simply go to college as a default. I don't think there's a right time to start a company. I think the critical thing is to have at least the kernel of a really good idea. At this point, that's sort of what I'm, I don't know, if you said what's the formula for a company, it's maybe some combination of: number one, a talented team that works well together, so it's the internal dynamics that matter a lot; number two, a good product/technology; and number three, a good business strategy.

And somehow, you want all three of those to line up. And if you have those, you can have a great company, even if you screw up a lot of stuff. There are all sorts of processes people focus on. And I think the reality is that most of the companies in Silicon Valley are somewhere between really badly managed to badly managed. And so if you look at these companies from an MBA process or from a Harvard Business School thing, there are all these things that are being screwed up in enormous ways. And it often doesn't matter because you have this, you know, this kernel of a good idea. You know, when I was doing, when I was, I was on CNBC back in September as part of this book promotion thing. And they spent sort of all these hours trying to do these gotcha moments to get you to say controversial things.

And they sort of got me to say at one point, you know, that I thought there was a lot of pot smoking going on at Twitter, which was sort of... When I left the stage, when the president of CNBC came up and said I did a great job, they wanted me back on the show, I realized I said a little bit too much.

But the... "You should have just told them, 'I know this because I've smoked a lot of weed with them.' And that would have defused the whole situation, Peter." "I know, I know."

You should have been like, "We have gotten baked at Twitter." But the larger context was it doesn't matter. Twitter had such a great fundamental idea that whether it was very badly managed or somewhat badly managed, or whether you're in favor of smoking pot or against smoking pot, it just didn't matter because the great idea sort of carried the day on so many other things. And then the more pessimistic version is, if you don't have something like that, it's actually really hard to make up for that, even if you get a lot of other details right. Where do you stand on...

...drug reform, legalization of pot, because that seems to be a fascinating issue. In our lifetime, we've seen so many trends where people were against gay marriage and against pot, and boom, they just switched. How do you feel about this?

I'm pretty libertarian on all these issues, so I think—I think it should be legal. I think it will be legalized. I think that, you know, we did our first investment in a marijuana-linked set of companies. "Oh, so you're investing in cannabis?" You know, it was, again, it was sort of a... They were doing things related to it. It was not like direct farming or anything like that. "Making UV lights or something?" It was a holding company that was doing a set of direct investments. And, you know, everyone said, "Oh, this is really crazy and courageous and out of the box," and I actually think it's not even that controversial anymore. I don't think there was anything that unusual about it.

Yeah, it seems to me like it'll be the next coffee, like the next crop. I mean, maybe not consumed that often, but it will be a staple: sugar, cannabis, coffee.

You guys know we put cannabis in the coffee, right? It's going to be a great afternoon. This is a good example. Even if this is a trend that's true, it doesn't actually translate into automatically figuring out the business. No. It's not clear how you make money farming sugar. I guess there's the Fanjul family that sort of controls all the sugar crops in the US. But outside of that, really hard to know how you make money with sugar. And so even if there's a trend towards marijuana legalization, maybe it means there'll be less money in it. Maybe a lot of the money came from the illegality and from the premium people are able to charge because it was actually illegal. We can sometimes identify really big trends.

This is a trend I believe is happening and will continue to happen, but how that actually translates into a successful business is quite tricky. I suspect there are some opportunities. The details are what matters, not necessarily the big trend.

We talked earlier about monopolies, and you had mentioned, 'Hey, we can debate that.' Let's debate it for a second. Microsoft: huge monopoly. Justice Department tried to go after them. Google: huge monopoly. They claim it's not a monopoly. It's obviously a monopoly. And EU: huge problems for Google. America: they seem to have skirted the issue. What do you think the government's involvement should be when something hits 95% or 85%, whatever you would define a monopoly as? What should the role of government be, if anything?

In your worldview? Well, there's some point where I think monopolies are bad. And I think they're bad—they're sort of bad when they become static, when they become like a troll that's collecting a toll at a bridge, or when they're sort of preventing anything from fundamentally changing. I don't know. If you have a monopoly where nothing changes in 100 years, that's probably just a tax and sort of ends up creating artificial scarcity, and that's why we have antitrust laws. On the other hand, we also have IP laws, copyright laws, patent laws that reflect—and these laws are often poorly designed, but they reflect the intuition that we want to give people a reward for innovating and coming up with new things.

So I think if you come up with something new, if you weren't able to have something like a monopoly in that, you would never be able to monetize your invention, period. If everything was perfectly competitive, you'd never make any profit. So the restaurant industry in San Francisco, perfectly competitive, but it's a terrible idea to open a restaurant. You will most likely go out of business, and if you're extremely good and it's extremely well managed and run, you'll just barely break even because there are enough people who figured out how to do that. So I do think that on some level these monopolies are the reward for innovation. I think the IT industry generally is one that's unusually monopoly-prone.

Because the marginal costs of software tend to be close to zero. So you get incredible economies of scale. You can often get people to adopt something new very quickly. And then the customers are relatively sticky. And so that combination of elements makes software very monopoly-prone, extremely lucrative. And this is why so many talented people are going into it and why there's so much innovation happening in software. There are many other sectors where I think it would be good for society if we had more innovation, but it's hard to get the microeconomics to work. I think it would be good if we had more innovation in clean energy. And the problem was, you come up with a new solar panel design, you're competing with 99 other companies doing something in solar, and then you still have to compete with wind, and then fracking comes out of right field, and cheap Chinese manufacturing out of left field.

You're a minnow in a vast ocean, competition everywhere. And at the end of the day, the business models were really hard to get to work, even though it would be good for society if some of these other sectors were a little bit more monopoly-prone. Too monopoly-prone becomes bad. The question of the government intervening, I think it got mostly wrong with IBM in the '70s. Things were already shifting to software. And the '90s with Microsoft is debatable. You could say the internet was going to happen regardless or not, it's a little bit of a counterfactual history that's hard to know. But I suspect that if the EU goes after Google, this will probably be a sign that maybe the search monopoly is reaching a natural end since the government seemed to time the point at which they intervene when these things are about to naturally die anyway.

Yeah, perfect timing. What are your thoughts on the polarization of wealth in society? I know you're a libertarian, so I don't think that means a free-for-all, but I think it means generally let the marketplace work it out. But it seems like the marketplace has created a pretty big gap, and things like minimum wage don't seem to be keeping up, cost of living. Things have gotten way out of whack in major cities. People are getting left behind. There's a little bit of suffering out there, generally trending in society as less suffering overall.

What are your thoughts on that polarization of wealth? I don't think it's a good thing. So I think it's happened to some extent. You can debate whether it's happened globally. So let's start with the question. You could argue inequality has actually gone down globally over the last 40 years. For sure it's gone down, yeah. And so it's gone up within countries, down globally. I'll grant that it's gone up within countries. I don't think that's a good thing within the U.S. It's much more debatable what the causes are, and then it's very unclear what the remedies are. So it's always three parts. Descriptive, has it happened? Let's say yes. Then, explanatory, why has it happened? Why questions, much harder to answer. And then remedy, what do you do about it? Much harder.

As a libertarian, I would say that it's a lot of the regulations that make things so expensive and unaffordable and that are putting so much pressure on lower-middle-class and poor people. So in a place like San Francisco, I would not blame the tech industry. I would blame much more the sort of weird coalition of urban slumlords and the Board of Supervisors that's refusing to build any more housing in San Francisco. Because I think it's the cost of living, the biggest single item. I don't think San Francisco would be that hard for people to make a living if it was not for the rents. Yeah, it would be fine. Everything else, you could probably get by on maybe not minimum wage, but somewhere above minimum wage you could get by if you had very cheap rents.

I think the affordable housing issue would be the single issue I'd focus on. I think it's always very specific issues you want to talk about, not these issues in the abstract.

Building more housing, people are against it.

Do you think this is gonna get resolved? Well, it's hard, but the zoning laws would be an example of a bad monopoly, where you effectively have a monopoly on land or on the ways you can use land. And monopolies get very bad when they get reinforced by the government, where, you know, I own these property rights, and I'm going to make sure they go up by getting the government to restrict other people's abilities to develop property. And then, cumulatively, it's very good for the existing landowners and property owners and bad for everybody else.

You've been portrayed twice in fiction now. I wonder how you feel about those and have you seen them? I mean, obviously, Silicon Valley and then in the Facebook movie. What did you think of the characterization of yourself as like, 'Oh my God, please don't go to college'? Because I thought it was hilarious.

Any resemblance is purely coincidental, I was told. Yes, so you can't sue for your life rights. It's coincidental. It's totally coincidental. Look, it's vaguely flattering. I think it's always better to be portrayed as eccentric rather than evil, so I'll take eccentric anytime. Yeah.

Ladies and gentlemen, Peter Thiel.