clock menu more-arrow no yes mobile

Filed under:

The way SoftBank invests in startups just doesn’t work, says Khosla Ventures’ Keith Rabois

Rabois talked with Recode’s Kara Swisher about startups, innovation, Facebook, President Trump, and more on a recent Recode Decode.

If you buy something from a Vox link, Vox Media may earn a commission. See our ethics statement.

Khosla Ventures partner Keith Rabois. Steve Jennings / Getty Images for TechCrunch

On a recent episode of Recode Decode, hosted by Kara Swisher, Khosla Ventures partner Keith Rabois joined Kara in studio to talk about a range of topics, including the startup landscape, being a conservative contrarian in tech (who has publicly rejected President Trump), and why he thinks SoftBank’s Vision Fund is prolonging the deaths of doomed companies.

(The interview was recorded on December 26 and published January 5. This week, reports emerged that SoftBank would invest $2 billion in WeWork, rather than an originally planned $16 billion that could have given it a controlling stake in the office-sharing company).

“[Softbank has] deferred some companies’ aspirations of going public,” Rabois said. “I think it’s created a crutch for other companies that really don’t have an economic model that’s working, and it’s created a bank account that people can tap into and not have to solve their business problems.”

“I personally believe that scarcity to capital is a good thing, that desperation breeds innovation, and that you need constraints to actually execute well and innovate,” he added. “If someone gives you this pile of money, I think it creates a lot of excuses and soft thinking.”

He noted that he likes the chances of some of the companies SoftBank has invested in, objecting more to the many billions of dollars the Vision Fund (which is itself heavily funded by Saudi Arabia) has injected into the Valley.

“The mentality of throwing money at companies and making them successful just doesn’t work,” Rabois said. “I’ve never seen real examples of just, you take money and you crown a winner. That’s the philosophy, which I don’t believe that works. I think that’s the whole history of Silicon Valley, is that these upstarts with very limited resources and a bunch of misfits have rearranged every single industry, and they’ve done it over and over again.”

One slightly awkward footnote: In September, SoftBank’s Vision Fund said it had invested $400 million in online real estate-buying marketplace Opendoor, where Rabois is the executive chairman.

You can listen to Recode Decode wherever you get your podcasts, including Apple Podcasts, Spotify, Google Podcasts, Pocket Casts and Overcast.

Below, we’ve shared a lightly edited full transcript of Kara’s conversation with Keith.

Kara Swisher: Hi. I’m Kara Swisher, editor at large of Recode. You may know me as someone who’s always picking a fight on Twitter, and usually with Keith Rabois. Anyway, we’ll do that after this podcast. Okay. In my spare time, though, I talk tech, and you’re listening to Recode Decode from the Vox Media Podcast Network.

Today in the red chair is Keith Rabois, an investor and entrepreneur who I’ve known a very long time and who was on a very early episode of this podcast many, many years ago. He was a co-founder of PayPal and he’s an investment partner at Khosla Ventures. Keith, welcome back to Recode Decode.

Keith Rabois: It’s a pleasure to be here. I was not a co-founder of PayPal.

Oh, okay.

I was an early executive. I wish I was a co-founder.

Early executive? You were one of the mafia, though, right?

Yeah, definitely a part of mafia.

Right. Okay.

Peter and Max were technically co-founders with Elon.

All right. Well, I think of all of you as a big old group, but although, Elon came from

He did. There was a merger in March 2000, right before the internet bubble collapsed.

Yeah. You know who wrote that for the Wall Street Journal, that story? Kara Swisher.

That must have been you. Yeah.

Kara Swisher was. Yeah, indeed. So, anyway, you went on to fame and fortune in many different areas. You now are an investment partner at Khosla Ventures. So, that’s what I want to talk about.

Keith, one of the things I do like about you, we don’t agree on a lot of things, and we definitely argue over it on Twitter, but it’s a very lively discussion, a very civil discussion, for the most part, about various issues around Silicon Valley, around ethics issues, around all kinds of things. So, I want to really have a lively discussion today about things that are going on in Silicon Valley. Why don’t we talk ... Let’s get people familiar with your history. Go through it really quickly.

Sure. So, actually, I started as a litigator on Wall Street and moved out here in 2000 when Peter Thiel became interim CEO of PayPal in September 2000.

You knew him from Stanford, right?

I did. I sort of fortuitously met him my first day of freshman year when he was delivering his alternative newspaper to my dorm room, and I happened to be there and was interested and started chatting with him. In any event, I met Peter almost 30 years ago. When he resumed the reins at PayPal, he recruited me to move back out here, so I joined PayPal, which was a bunch of misfits losing a lot of money. We were burning $10 million a month, actually, at the time, which in 2000, was a lot of money, and were running out of money very fast.

When Peter sort of displaced Elon Musk as CEO and in November 2000, I moved out here to join PayPal, and within a year we were able to turn the company around, make the company actually profitable, file to go public, actually, literally filed the day before 9/11, and still was able to have our IPO in February 2002, subsequently sold the company to eBay after being a public company, and then rejoined Peter to help fund and start new companies in 2003, while the rest of Silicon Valley was dead and thought there was not gonna be another wave of innovation.

They did not.

Particularly, the consumer internet wave was “done.” We found some entrepreneurs that were still interested in starting things, and we were able to help propel them. I wound up joining LinkedIn about a year later. Reid Hoffman, who obviously had worked with us at PayPal — had been my first boss, actually, at PayPal — started LinkedIn, and from the very early days I was interested. I invested at the company, and then later joined full-time and stayed through the Reid regime. He eventually found a successor CEO, and I left to go join Max Levchin at a company called Slide that most people don’t remember.

Aw, Slide. I do.

People do remember Max, fortunately, but don’t remember Slide, which eventually was acquired by Google.

I came in and mocked you all the time. Do you recall that? Do you recall?

That is true. You wrote a great series of posts about it.

It was a gaming company.

Yeah. It was a social gaming company on a platform.

It wasn’t the worst idea. It was just part of an era. It was part of an era.

No. It actually made some sense.


We used a lot of data to program our games. The problem was we lacked a top-down vision, and I think you need both. You need data and a vision, and so we let users sort of vote with their feed, and we optimized around that. We ultimately ...

There was a bunch of companies like you.

Yeah. We were ultimately acquired by Google, but it wasn’t a very successful ride after about three-and-a-half years. I stayed at Google for about two weeks and got persuaded to join this company called Square, which hadn’t launched yet. It was about 17 to 20 employees. Jack Dorsey had founded it and then funded it, and was working on launching it, and I got recruited to join, decided to do that, went through kind of a rocket ride, actually, very reminiscent of my PayPal days, and then eventually joined Vinod, who had been on my board, Vinod Khosla, at Square, joined him as a managing director at Khosla Ventures, and believe this or not, March of 2013 to almost six years ago, which is an eternity.

It’s a long time, right.

Yeah, it’s an eternity.

Yeah. I never thought you’d become a venture capitalist. You were sort of an operator.

Well, I was an operator, but I was always angel investing for about a decade before. From about 2003 to about 2013, I’d probably invested in about 80 companies, and I actually enjoyed the process of meeting with very early-stage entrepreneurs who were talking about the proverbial two kids in a garage, found a lot of really interesting companies before other people knew they existed, developed a bit of a track record and enjoyed that, and sort of always thought that one day, I’d want to be a venture capitalist, and decided finally, in 2013 to take the stress out of my life, which when you’re operating a company, every day, there’s a crisis. Jeff Jordan taught me a long time ago that one percent ...

This is Jeff Jordan from OpenTable. He’s also a venture capitalist now at Andreessen Horowitz.

Venture capitalist, eBay executive who championed, actually, the acquisition of PayPal back in the day, but Jeff had this rule of thumb when he was managing people that about one percent of your employees have a crisis every day, and that could be a personal crisis, a professional crisis, a medical crisis, a family crisis, but when you’re the boss, all those crises are yours.

Yeah. Someone used to call them “big, giant babies” to me, but ...

Yeah. So, after 13 years of suffering through a crisis every single day and kind of the sine wave of emotions, I decided to be a venture capitalist, which at least smooths out those waves. It had less adrenaline and longer feedback cycles, but the day-to-day ebbs and flows are no longer your problem.

Why did you pick Khosla? What was the ... There’s a lot of different venture firms, and this is pre-SoftBank, when Andreessen Horowitz had existed. Right? They had ...

Yeah. Andreessen I think was started around 2008 or ‘09.

Right. Yeah.

So, as I mentioned, Vinod Khosla, who founded the firm, was on my board at Square, so I knew him quite well. Also, David Weiden, who’s another general partner at KV today, had been on my board at Slide, so two of my three GPs, really, had been on my board for the last five or six years, and I think that’s a very healthy way to start a venture relationship, because you really need to have a lot of trust and confidence in each other. You don’t see each other that much. We typically see each other only on Mondays, and yet-

Right. So, talk about ... Khosla’s different, because every venture firm is different. Some go for the whole group thing. Some think of other things. You guys are sort of independent players. Right?

We are independent, in many ways. We take feedback from each other and really learn a lot from each other, which is the benefits of being in a partnership. We also, though ... One other distinction that I really thought would be meaningful is we like to take technical risk, so we invest as early as possible. We try to be bold in our bets as early as possible and as impactful as possible, and I wanted to do more technically founded investments.

So, what does that mean to you?

It means that the core risk is the technology. Can this actually happen? Can this be done? So, for example, a very popular topic would be, can you build an autonomous-driving automobile? That’s a technical risk. The market is there. If someone could deliver, for a reasonable price point, an autonomous vehicle, there’s definitely people who will buy that, but the question is, can you do it? How accurately can you do it? How many miles of data do you need? How many years is it gonna take?

What are the perverse side effects of deploying driverless cars? But 40,000 Americans die every year in automotive accidents, and about 40 to 60 percent are completely preventable. It’s because the driver is distracted with his or her phone, with drugs or alcohol, with talking to their passenger, but there’s no doubt that if the technology exists, there’s something to do with it ...

There would be a market. Right.

... where sometimes you’re taking market risk, which is, it’s very easy to see that the technology will work, but no one cares. No one wants to buy it. So, we take a lot of technology risk, and that was different for me, but I was excited to learn about that, and it allowed me to branch out into areas that I hadn’t been involved in — let’s say healthcare investing, which I was really interested in.

The healthcare system’s a complete mess in the US, and probably globally. There hasn’t been a lot of innovation recently, and I thought that I’d want to spend the next decade of my life diving into new areas like that, and as a result, actually, the last six years I’ve spent about maybe 25 percent of my time and money investing in healthcare innovation, which is something I’d never been able to do before.

Mm-hmm. So, you were looking for just technical solutions versus market. You do do investments in both things, but Khosla’s different.

Oh, I do both, but I felt ...

What would you say you sell to investors when you do it?

Yeah, sure. Well, the thing that I actually have done for almost 20 years now is I invest primarily in founders, so I’m really a people-based investor, and so I’m basically assessing the opportunity that the people in front of me have of changing the world. If you think about how ridiculous Silicon Valley is in some ways, on a rational basis, it’s completely ridiculous to wake up in the morning and say, “I’m going to change an entire industry that’s been around for centuries,” but it happens all the time.

People succeed, but it is kind of irrational and kind of ridiculous to start that way. So, my core expertise is once in a while being correct about meeting someone who’s 19 years old or 25 years old or an immigrant that’s never done anything, has no traditional credentials, and saying, “You know what? This person actually may be able to do that, and I’m gonna give them my time, my money, and my credibility, lend them it, and see if they can pull it off.” And I’ve been right somewhere between 30 percent to 40 percent of the time, which is actually pretty good in investing.

So, when you’re thinking about sort of this scene right now, from when you started to today, talk a little bit about the landscape now, because a lot of things have happened. You had Andreessen Horowitz, which sort of burst onto the scene, and Khosla have been around for a long time, but still, you had ... Vinod has a huge personality and also a huge reputation, but it was mostly business as usual until just recently with SoftBank’s entrance with the Vision Fund, the giant, massive fund that they brought in.

Well, I think SoftBank’s had radical implications on the late-stage, the growth round, so to speak. What we do at Khosla is more early-stage, what I think of as traditional venture capital, which is a seed financing or a Series A round, where obviously, SoftBank’s been deploying a lot of capital in big bets, in mostly late-stage investments, and it has had significant impact in the entire ecosystem.

So, talk about that.

I think it’s deferred some companies’ aspirations of going public. I think it’s created a crutch for other companies that really don’t have an economic model that’s working, and it’s created a bank account that people can tap into and not have to solve their business problems.

I personally believe that scarcity to capital is a good thing, that desperation breeds innovation, and that you need constraints to actually execute well and innovate. If someone gives you this pile of money, I think it creates a lot of excuses and soft thinking. To some extent, this is what happened to us at Slide. We raised one of the first monster growth rounds.

What was it? It was enormous, right?

Well, it was basically Fidelity and T. Rowe.

That’s right. It was Henry Ellenbogen.

Yeah. We raised a $50 million round in 2008, and that was kind of unprecedented, actually, for a private company, for public-stage investors, and I think it made us a little sloppy in several ways. We thought we had sort of infinite resources and infinite time, and it didn’t force us to solve our fundamental issues right away.

So, I think that that is likely to happen with a lot of companies that SoftBank invests in. I think they’ve invested in some very good companies that are clearly gonna be successful, and they’ve invested in a lot of companies that probably have some fundamental sort of root-cause problems, and giving them a lot of money isn’t probably going to solve them.

Right. So, with the theory behind it, what do you think of this theory of just throwing hundreds of millions of dollars and those kind ... because it does create, even if it doesn’t affect you, it does create a mentality up and down the system.

I agree with that on two fronts. So, I think the mentality of throwing money at companies and making them successful just doesn’t work. I’ve never seen real examples of just, you take money and you crown a winner. That’s the philosophy, which I don’t believe that works. I think that’s the whole history of Silicon Valley, is that these upstarts with very limited resources and a bunch of misfits have rearranged every single industry, and they’ve done it over and over again.

Every time someone thinks somebody in Silicon Valley is dominant, whether it’s Yahoo, or before that, AOL or whoever, they get upended by a new startup that’s a bunch of your dropouts that have a very limited amount of dollars, and so I just don’t believe that money is the ... It’s not the panacea, and it’s certainly not the keys to the kingdom. I think there are some companies, very selective few, where they have an engine that is polished and ready to rev, and you’re just giving gasoline to an engine that works really well.

Like a Slack, for example. Slack, I mean.

But there’s very few of those.


Those are very rare, and so I don’t think ... If SoftBank makes 50 investments, maybe five at most are those kind of companies. The disciplined companies that really get product-market fit without a lot of capital are almost surely gonna do better than the ones that chase hundreds of millions of dollars.

So, how do you diffuse that in this ... How do you compete in this system when you’re in this system, just in general?

I think sometimes it takes time. It’s hard. I mean, a lot of my job is really explaining things to entrepreneurs. I don’t have any power. All I have is influence, and so I try to explain the trade-offs, and sometimes it just takes a cycle for people to see other people get burned by experience before things make sense to them, even if you can explain things rationally without the emotional context.

So, for example, we grew up in Silicon Valley that went through 2000 to 2002, so to us it’s ...

Right, and then 2008, right?



But 2008 was a blip. It was very short.

Right, but 2002 was ... Yeah.

2000 to 2003 was like a ... It was called like a nuclear winter here, and almost everybody you met didn’t have a job. They were out of work. I think those who went through that searing experience, and you can see it on Twitter, things that someone like Bill Gurley tweets... It just frames your perspective in a way that you realize not everything’s always up and to the right, and that you need to have control of the leverage in your company and control of your destiny so when the world does change, you’re not caught blindsided. But if you’ve never been through that kind of experience ... I work with a lot of young entrepreneurs that are basically post-2008, and they’ve been through ...

Zero, right.

... 10 years of uninterrupted growth, and this doesn’t make any sense to them. When you talk about, “Well, the world can change. Capital is not always gonna be free or very cheap.”

Like this week.

Yeah, this week is gonna potentially shake things up, but I can see a little bit of nervousness on some entrepreneurial friends of mine’s faces this week, the ones that are in town, and they’re asking questions for the first time, the kind of questions that three years ago and five years ago they were completely ignoring.

So, I think the benefits of some experience are that you can see how the world shifts and that you can be prepared for it. But I think in total, the world of SoftBank goes through an evolution. I think it’s gonna be very hard to teach, in quotes, people the downsides of taking SoftBank capital, which are many.

Right, and we’ll talk about other issues there in the next sections.

Oh, there’s lots of issues there.

... around the Saudis and everything else. So, to finish our segment, what is the state of venture capital right now, as we’re going into 2019? How would you describe it right now?

Interestingly enough, I think in the last month or two it’s really cooled down. I think you can see that certainly Series B and later level, a fair amount of hesitation on pulling the trigger on monster investments. At the seed level, it’s not as obvious how much impact we’ve seen in the last six to eight weeks. It’s also a little difficult because venture capitalists canonically go on vacation in December, and it was very busy, a shockingly busy December, but I think that was the run on the bank sort of experience.

Because? What do you mean?

Everybody, I think, felt a little bit of unease, so every company that could and every founder that could and every VC that could was raising money ahead of the curve before 2019 started. So, typically, December slows down for investors and for companies because they’re sort of derivative from the venture capitalist schedule. I felt like December was the busiest of my venture career by far, and I suspect ...

In terms of investment or raising?



Raising, and I think it was because people felt that the world could be changing and shifting, and they’re all trying to fuel up with money as fast as possible so that they weren’t caught last in the last chair standing kind of problem.

So, we’ll see what happens in January when everybody comes back to work. Everybody’s in the same place, but with the market turbulence, it’s certainly gonna exacerbate and amplify some of those feelings, but whether it’s a real correction or a temporary blip, I don’t think anybody absolutely knows.

As Peter Thiel likes to say, the only thing he really knows is we’re 10 years closer to the end of this bubble than we were, but that no one can predict the timing.

So, when you look out at 2019, are you optimistic, pessimistic, or do you just gonna keep ...

I’m pessimistic at the later-stage version of Silicon Valley.

We’ll talk about IPOs also in the next section.

I think there’s going to be a ... There’s already been a 30 percent correction of public market prices for technology stocks, which is a reasonable amount already, but I suspect there’s a little bit more there.

I’m very optimistic about the innovation in Silicon Valley, that the fundamental building blocks of technology are accelerating. The innovation at the root level is exciting. Every partner meeting on Monday, we typically have companies present, and because we invest early ...

You see early trends.

... we see the future, and the future is great on all kinds of dimensions. We see fascinating innovation every week, so I’m incredibly optimistic, and these companies are gonna bake over the next three to five years, so what happens in January doesn’t really matter to them.

Doesn’t matter, yeah.

It’s like having a baby, in a sense. It takes a couple years to grow up before you even know much about the child.

That is true. I have children. Do you have children, Keith?

Not yet.

It takes a long time. Can you give me a five-minute idea of what that is? What do you mean? What is coming that you find so intriguing?

Sure. So, a couple of areas. In healthcare, the use of data to improve outcomes is really starting to happen, where math actually does a very good job correcting for the errors and mistakes humans make. We’re starting to see the FDA actually approve technologies and products that use math instead of humans, and that’s encouraging a positive feedback loop where people are more willing to invest in companies that are gonna displace human judgment and make either access more affordable or more immediate.

So, diagnostics, for example.

Yeah. Diagnostics are a good example. The treatment’s a little bit more tricky. We don’t do biotech, per se. I mean, the actual drugs we wouldn’t do. There is a nice trend of using software instead of pharma, which is interesting.

To test things.

Where you can produce the same outcome by using an application to retrain your brain in different ways, and the FDA’s also blessed that. So actually, one of the benefits of the Trump administration, insofar as there are any... I think most people would find that the FDA’s been much more both rigorous but also innovation-friendly.

Yeah. I have heard that. Yeah. That’s what a lot of people are telling me. Anne Wojcicki just said that.

It’s leading to positive outcomes. I mean, these are real people that need new treatments, and I think the FDA, the new FDA is very consciously aware that they have to ruthlessly enforce principles.

For safety.

But they also should look for new ways of doing things.

Most people feel the FDA was way behind Europe or any other, which is unusual, we’re being behind Europe in terms of regular ...

Yeah. No, extremely unusual that a US agency is more heavy-handed and more intensively regulating than the equivalent in Europe. That was probably going on for a while, but it has been corrected in the Trump administration, so that’s one of the best areas of the Trump administration, particularly vis-a-vis Silicon Valley.

All right. So, healthcare. What else? What other areas for innovation?

We’ve had some investments in satellites and in rocket companies. One that’s already sort of pretty public is Rocket Lab, which is producing very low-cost rockets that will put satellites into orbit quickly and regularly with the ...

It’s low-hanging satellites, right? I’ve talked to ... Jerry Yang is also in a bunch of those.

Yeah. This one’s actually ... We’ll put it into orbit for less than the cost of SpaceX, and the goal is to actually be putting rockets into space as frequently as we expect airlines to take off. I mean, if you think about when planes were invented, the idea that planes would be flying every hour everywhere was probably pretty crazy. There will be a point probably in your kid’s lifetime where rockets are taking off just like we expect planes to take off, and that will change a lot of things.

The cost of putting a satellite into orbit and the delay associated with putting a satellite in orbit handicaps and hampers innovation, and so when people realize that they’ll be able to put a dedicated launch into orbit for less than $5 million, and that they can bank on the predictability of putting that into orbit and replacing it, there’s gonna be a lot of innovation that can be done in space. So, I think the Space Race is back.

I spent a little bit of time this year reading about Apollo 8. There’s a great new book called Rocket Man. Starting to get back familiar with how much innovation and the kinds of techniques people use and the kind of thinking people use, so I think we’re gonna see a wave of that innovation that we haven’t really ...

Including travel? Is that something you’ve been investing in?

I have not.

This is more deployment of different satellites.

We have looked at ...

That’s a lot of the things, deployment of satellites, or satellites, essentially.

Well, communications and data are incredibly important to everybody, from military applications to civilian applications, so it’s not surprising that there’s ...

So, girding the globe with satellites?

Yeah. That is gonna increase. Transportation, we’ve looked at some transportation innovation. We haven’t really invested in any. There’s a supersonic plane called Boom that a lot of people find very promising, but I haven’t seen ...

Supersonic plane. This is going ... So, to travel from California to New York ... We’ve talked about this with ... The defense agencies were working on some of this, Mach Five. Is that right, Mach Five?

It’s almost surely doable. The questions are really, what’s the cost gonna be? What are the trade-offs in terms of sound and noise disruption? It’s a little bit like … the Concorde was built on technology, really, from the 1950s to ‘60s, and no one’s really rethought the Concorde since then.

They haven’t. The Concorde was such a disaster. Right?

Well, it was fairly advanced for the 1970s.


It just wasn’t perfect for the ‘80s, ‘90s, and let alone 2000s. And so it’s really 40-, 50-year-old technology, which is an eternity in the tech world. So I think there’s gonna be some innovation there, but we haven’t funded any yet.

We have funded autonomous driving startups and the kind of sensors and core technology components that would be very critical in deploying driverless or autonomous vehicles.

Where do you think that timing is, going out?

I actually think it’s sooner rather than later.

Me too.

One of the reasons why is, humans are just not very good at this.


And so if your base ... It depends a little bit, and we’ll talk about this, these are the other policy issues. But it depends if your baseline is absolute perfection or your baseline is how many humans die every day. And to me, if the technology is better than human driving and people die less frequently, we should deploy it as fast as possible.


Where some people would say, “Well, you have to prove it’s perfect.” And the same ... We had this debate, if you like to read old-school history around electricity.


Electricity, at first, was a fairly dangerous technology.

Yeah. Homes got blown up and stuff.

Yeah, all the time. And you can read a great book, Historical Fiction by Graham Moore about Tesla versus Edison and Westinghouse. And it was an awesome part of American history.


But if the framework had been, “Is electricity perfectly safe,” we never would have had electric lights.


And so I think what technology ...

And gas was more dangerous.

Well, gas was absolutely more dangerous, once the technology for electricity got to a certain predictability ...


That’s why nobody uses gas anymore for most things. So I absolutely really reject the idea that new technology should be perfect. That’s not the milestone. The milestone is, is it better than what we do every day? And if that’s the case, then autonomous driving is gonna hop in sooner rather than later.


I don’t think humans will stop driving for a while, but there’s probably a point, even in my lifetime, where humans are not allowed to drive.

Well, it’s interesting. Whenever I gave a speech two years ago and I said, “Someday owning a car is gonna be like owning a horse.” You’ll have it at your ranch and you’ll drive it around, or you’ll ... It’ll be an entertainment versus a real transportation.

I agree with that.

And they shouldn’t be in cities. There shouldn’t be cars in cities that aren’t autonomous.

Yeah, I actually funded one company that’s still in quasi-stealth mode, so I won’t talk too much about it, but that’s reinventing the idea of a city predicated on the idea that transportation is completely archaic and is really built on the 19th century. And we’ll have something live, maybe later this year, even.

Yeah, it was interesting, years ago if you remember, when Larry and Sergey from Google were gonna put ski lifts in San Francisco. Do you remember that crazy scheme?


At the time, I was like, “That’s actually the way we’re gonna get around.” We’re not gonna have cars. We’re gonna get around by these ... Whatever they are, vertical lift and take-off. Ski lifts were fine. I was always like, “This is actually how it has to be in cities. Cars have to leave cities.”

Yeah, cars ... I mean, look ...

Non-autonomous cars have to leave cities.

Non-autonomous cars take up ... Just look at the parking implications.


I’ve read studies that between 20 percent and 40 percent of real estate in a city can be allocated to parking. That’s so crazy. Think about how much housing you could build or what else you could do with that real estate, if you didn’t need as many parking spots.

Right, absolutely.

All right, so let’s ... So anything else, any other areas? Autonomous, healthcare space ...

Well, everybody talks about AI, right?


So that’s the other topic, the hyped-up topic. The way I think about AI is maybe a little bit different, which is, I actually believe in AI and I believe there’s a lot of innovation that’s already happened and still accelerating. However, to me, the end goal is that the customer, the normal person on the street, doesn’t even know that technology’s powered by AI.


So for example, I was a lawyer. There’s no reason you couldn’t replace a lot of what lawyers do with AI.

Right, absolutely.

So I actually wanna fund some people doing that. I’m looking into various ways to do that. But I don’t know the end purchaser of the legal services needs to know whether it was a human or a machine creating the brief.


But I want a brief that’s not drafted by a human or that the ratio of drafting of a human to computer is 99 to 1 or something, or 1 to 99. So I think there’s gonna be a lot of technology components under the hood that are powered by AI, but the product looks like a normal product, just costs a lot less and works a lot faster.

So for example, out of Y Combinator, we funded a company that does patent filings by AI.

Right. Which is a rogue thing.

So they use AI to draft their patent filings with some human labor.


But the customer is delighted because instead of going to a patent firm and paying an exorbitant cost to write a patent filing and waiting three to four weeks, we can do it in one to two weeks for a third of the cost.


So you’re gonna see a lot of stuff like that.

Absolutely, absolutely.

And then more people will be able to consume it.

Lawyers, diagnostics, there’s all kinds of things that are higher-level jobs that are ... But there’s still, to me, they’re like manufacturing. They’re like working in a factory, some of these things are.

Well, they are, more than people realize. I think this is the thesis at Khosla that we’ve had for a while. At Khosla Ventures, we’ve been funding, replacing human judgment with AI, not the manual labor. But there’s jobs that allegedly entail human judgment that are a lot more like manufacturing than people realize. I mean, I was a lawyer. A lot of what lawyers do is a hell of a lot like working in a factory.


That’s one of the reasons why I escaped.

Right, right, exactly. All right, so we talked about innovation going on, but how do you assess this year? Let’s talk about some of the ethical issues with Facebook and with Khashoggi and this investment by the Saudis. Do you think this group of people is equipped, with all these exciting things that are ahead, to deal with it? How do you assess sort of the ethical health of Silicon Valley right now?

Well, I would also frame this very differently.


Which is, compared to who and compared to what?

All right. Okay, these are people who did go on and on about how good they are.

Yeah, that’s fair.

This is a group of people ... Wall Street people never said, “We’re not evil.”

No, but people do come to Silicon Valley with a mission.


And the mission is to improve some element of society that’s probably broken or inaccessible. And I think that’s what’s powered the best companies in the history of Silicon Valley forever. So I believe that that’s true and that’s the missionary zeal.


Silicon Valley’s become more important. It’s a bit like the old Steve Jobs quote of the pirates and the Navy: “Everybody roots for the pirates until they’re successful, then you have to act like the Navy.” And Silicon Valley has historically been the pirates, upending and improving archaic industries.

Which is an old meme, let’s be honest.

Yeah. No, but it’s actually quite relevant. I actually think I coined this, in terms of the change. I did an interview with Travis about five years ago.


Yeah, onstage. And as he was talking, it occurred to me that basically what was happening was the technology companies were becoming so important that they were really more like the Navy.


And when you’re the Navy, you have to act like the Navy. You can’t act like the pirate anymore. And then that transformation probably lagged a bit, especially at some companies like Uber, but maybe even across Silicon Valley. But to me, humans are humans, and they’ve been humans ... If you read Shakespeare, there’s nothing that’s happening in Silicon Valley that you can’t find in Shakespeare.


Because humans have always been humans. Or, if you read Sapiens, you get a feel for this, too, going back not just centuries, but eons. And so to me, when you talk about an ethical debate, the question is, are we doing things better than we used to, better than other humans, better than 100 years ago, better than 1,000 years ago? And the answer to most of those questions is yes, yes, and yes.


And so yes, people can improve. Yes, Silicon Valley can improve. But if you compare it on any benchmark and yardstick to humans in history, it’s usually pretty damn good.


So we can talk about this very specifically on any of these topics, but I don’t start with the premise that Silicon Valley should be so bulletproof and so perfect, or it’s fair to criticize Silicon Valley.

Right. Okay, so what do you think of the criticism? It’s mostly focused on Facebook, of course, as you know.

Yeah, which I think, in some sense, is also a little unfair.


I think some things Facebook has done are unique to Facebook, and that’s a legitimate topic to discuss and criticize. I think a lot of things Facebook has done are actually things other technology companies or other non-technology companies have done, and the media’s been fairly inconsistent in singling out Facebook.


Let’s take a classic example. People talk about using certain data to target ads on Facebook. And they can complain about, you can target on this way or that way, and they find nefarious examples of such, allegedly. I would posit that anything you’ve seen on Facebook in terms of targeted ads can be done on steroids, i.e. more targeted with more data that’s more personally sensitive, using direct mail. And direct mail’s been around at least since the 1940s.

Right, right.

But I don’t read that the New York Times is writing exposés on direct mail.

They have written stories on direct mail.

They probably have, but then they drop it.

Yeah. So why do you think this way? Is it the size of it? The massive impact, it’s the numbers, I think, is probably why.

I think it’s jealousy, truthfully.

Oh, interesting.

I actually think a lot of journalists are jealous of tech. They have colleagues that are ... They have comparable people that went to school with them at Harvard, Yale, etc. that went into tech companies that are saying they age and make a lot more money.

Oh, I’m gonna push back on that. I have ... No.

Secondly, the biggest issue is gatekeeping. It’s the removal of the gatekeeper role. So when I was growing up, which isn’t that long ago, the media had massive gatekeeping obligations and impact. So if CBS news didn’t wanna cover your story, there really wasn’t a way to communicate directly to the American people. Or, even when I started in Silicon Valley in 2000, basically if the New York Times, Wall Street Journal, Fast Company, Wired, or maybe one or two other publications wouldn’t write about you, there was nothing you could do. That’s changed with a proliferation of Twitter, Reddit, Facebook. There’s so many channels to communicate your ideas that the role of the media has been eviscerated. The media cannot stop ideas from spreading anymore, and I think that frustrates a lot of people in the media and they kinda wanna put the genie back in the bottle. And you see that in a lot of the coverage. It’s either explicit or implicit.

Naval was the first one to ... Naval Ravikant was the first one to really point this out. In a blog post he wrote in 2012, I believe, he basically predicted this whole trend that the media would push back on the lack of gatekeeping that they’ve historically performed.

Well, lack of gatekeeping. What about, you’ve gotta ... Keith, you can’t utterly dismiss some of their sloppy management at Facebook. Over and over again, there’s instances of not being gatekeepers being problematic.

Yeah, that’s fair. But compared to what? I mean, again, the printing ... As Tyler Cowen has written, I don’t know any criticism of Facebook that wouldn’t have been written about the printing press. When the printing press first was ...

Which there was. There was, yes.

Yes, but the printing press was obviously one of the most important things in the history of the human race. And it created revolutions. Before, religious sects could control the dissemination of ideas.

Right. Yes, monks. I’ve heard of it, right.

And after the printing press, they couldn’t. And so there were lots of new religions, so they created revolutions. And that’s exactly what social platforms are doing. They’re creating revolutions. But the printing press was clearly a ...

Now, the printing press was available to lots of people. Facebook is run by one person, if you really ... Pretty much.

Yeah, but they’re ... If not Facebook, there’s Twitter. And if not Twitter, there’s Reddit.

But Facebook, by far, is the biggest.

I don’t believe that, actually.

All right, okay.

I think Twitter has more impact on policy.

Yeah. No, that’s ... I wrote a column about that this week.

Yeah. I actually don’t think Facebook has nearly as much impact as people think. For example, Google’s pretty influential. It’s hard to live without Google. I don’t find it hard to live without Facebook. It doesn’t perform a utilitarian function for me. Twitter does perform a utilitarian function.

For news, newspaper. Right.

It’s like my newspaper. It’s like my New York Times.


When I was growing up, on Sundays I would sit at my table, have a bagel, and consume the New York Times.


Now I wake up in the morning, I don’t have a bagel anymore, I have eggs instead. But I also read Twitter all morning, and then I click through other links that are interesting.

Right. Including New York Times articles, yeah.

So it’s this ... Yeah, absolutely. It’s a reasonable fraction of stuff I may click through. But fundamentally, my go-to constellation is Twitter.

What would you criticize Facebook for?

Well I think they ... So I think their communication strategy has been intellectually bankrupt.


I think they’ve been incredibly defensive. They should be framing some of these debates instead of waiting for the exposé and then reacting to it. I think they’ve struggled with whether they really want to embrace a censorship role or not, which is a complicated topic. I think, for example, if you take their mission seriously, which is to serve all humans and certainly all Americans, they really shouldn’t be censoring content unless libelous, obscene, that’s about it. Pornographic.

Violent, violent. Promoting violence.

Yeah, maybe directly inciting violence, which is the standard from the Supreme Court in 1969, but anything else, I don’t think you can serve the needs ...

And what about their performance around political advertising in the Russian ... Having it ...

Well, let’s talk about that. So there are some laws that apply to the purchase of ads by foreign entities that have been the law of the United States for a while. And Facebook, like any other media, should observe those laws.


So if you can’t buy a TV ad if you’re a Russian entity, you shouldn’t be allowed to buy an ad on Facebook. It should be consistent across channels. But if you can buy a TV ad, then you should absolutely be able to buy the same ad on Facebook.

Sure, yep.

So I think they were negligent in some of that. Now that said, a lot of the complaints ... Because the total expense ...

Was around the content. I understand that advertising wasn’t quite ...

Yeah, the advertising was a small component.

It’s the use of it, it’s the content, it’s the distribution.

But the content, if those people are allowed to publish a book in the United States, why can’t they share the content on Facebook?

It’s amplified, it’s problematic.

Yeah, but ...

I know, but I’m just saying it puts them in the zero seed.

Right, but if I publish a book, I can buy ads for the book. I can run an ad in the New York Times for my book.


I can go on radio and talk about my book, I can do a book tour.

But look at the difference. Say I ... It was interesting, this sort of pushback of Alice Walker at the New York Times, just even mentioning a book, and the New York Times responding, versus Facebook and being used by the ... Without explaining that these were Russians that are trying to influence the election. You know what I mean? There was ... Until lately, that wasn’t the case.

Well, I think that influencing elections is something every government in history has tried to do.

Right, and this just is a great platform to do it.

The American government on down.

Yes, of course.

So I think the biggest issue was the American government was not wary enough about the agenda of certain other countries. Russia, particularly. But also, you could argue, China. And the tech companies, in some ways, were following the lead of their government. I mean, Obama famously ridiculed Mitt Romney for calling out Russia as a potential threat.


And why Facebook should be more prescient than Obama about the threat of Russia, to me, makes no sense. I think Obama was wrong.

Well, a bit more. Yes, absolutely. But they have a lot of information of what’s moving on their system. Or maybe they don’t.

Well, presumably Obama had a lot of information, I hope.

Probably Facebook and ...

From the CIA and other people.

Yes, yes, yes. And definitely, I think people within Facebook said ... We warned them. Certain people said we did warn them. Others say we didn’t warn them enough. You know.

Let me come back to the metaphor of a book, though.


I think if you can publish a book in the United States, then you should be absolutely free to share that content on Facebook or Twitter.

Even if you can hide who you are.

Well, what are the rules around publishing a book? This country was founded on anonymous or pseudonymous pamphlets.

With cool names, like Alexander Hamilton had...

Well, some were true names, some were not.

No, I know that. I know that. I was ...

I mean, it’s classic American history. This is the reason why we teach American history. It’s not just for obscure facts, it’s so that the debates and principles can apply to now, to today’s world. I don’t see why there should be a rule that says you have to publish under your own name.


I think if you buy ads, there’s a legitimate paradigm that’s been established that only some people can buy ads.

So does it worry you, as an American citizen, that this is allowed to happen?

Not really. But here’s why.


So the one study I’ve been waiting for someone to show me is a survey of some type — and you can debate the methodology; put aside the exact methodology for a second — that the voters in 2016, in some net way, were less informed.

They will never be able to figure it out.

Yeah. But there were ... I actually would argue that in 2016, the average voter knew more about society and life than in any time in American history. If you think about how much the average American knew in 1906 ...

Yeah, so they had more information.

Yeah. So show me a study, in some way that there was more confusion, more misrepresentation, more misunderstandings than any time in any other election. I actually think that’s impossible to prove.


Because it’s actually factually wrong.

It’s ... Well, perhaps.

The average American was pretty manipulated by a lot of media.


And actually wasn’t that well educated. I mean, until the 1960s, ‘70s, ‘80s, even the idea of going to college was sort of the preserve of the elite. Now, everybody debates whether everybody should go to college for free.

All right, so we have to finish this section, but what would you imagine will happen to Facebook now?

Well, I think Facebook has generally played some of their political cards pretty well. They’re pretty deft about using their connections through Peter.

Mm-hmm, Peter Thiel, who’s on the board.

Peter on the board, they’re pretty careful. They listen to Joel Kaplan, who’s getting a lot of coverage. I think Google’s a lot less careful about their perception in Washington. And there’s an interesting problem that these tech companies, the major tech companies, particularly Facebook and Google, to some extent, Amazon may have ...

And Twitter.

Which is, the left wants to regulate tech companies.

Yes, they do.

Because the left doesn’t like accumulated power.


Traditional left view. And the right, who should be the natural defender of not using the government and regulation to interfere in private businesses actually feels that these companies are monocultures of liberalism and doesn’t really wanna lift a finger to defend these companies that are sort of left-wing institutions. So the tech companies have this fundamental problem, which is their natural ideological defenders don’t really like the culture and the employee-based views and the contributions, the political contributions of these companies. So they’re not very excited about defending from the left, the regulation.

Yeah, they’re not going to. Yeah.

I mean, I talk to my friends all the time on the right, and they understand the views about why they should be opposed to what the left wants to do, but they have scarce time in their life and they have scarce political capital. They’re like, “Why would I defend Google? Why would I use my capital for that?”

Do you feel like regulation is coming? Antitrust, correct?

Well, antitrust is complicated because ...

It’s changing.

Well, it may change. But the fundamental principle of consumer harm is pretty deeply embedded in antitrust philosophy in the United States, and it’s very difficult, if not impossible, to show consumer harm, but these ...

To prove customer harm. People love Amazon, right?

Yeah. Well, they’re free products.

They’re great.

For the most part.

They work.

So it’s hard to show raising prices and monopoly profits. Secondly, Amazon tends to have a fairly low price, but not below marginal cost, so it’s been traditionally very difficult in the United States to show anti-consumer behavior when that’s true.

Same thing with Google and Facebook.


So what do you imagine the regulation ...

I think the regulation, certainly out of Europe, will continue to increase because they have a different history of antitrust regulation where it’s more of a policy and politics decision. Secondly, I think that until 2020, the administration is gonna be in the hands of mostly people who buy the Borkian view of antitrust policy, which is unique consumer harm. If Trump were to lose and the Democrats to win, then I think you could see a head of the FTC or head of the antitrust division that really tries to rewrite American antitrust doctrine.

Yeah, and that’s been talked about.

Before that, yeah, you see occasional academic journals ... I mean, this is what I used to do for a living, I was a lawyer, I was an antitrust litigator. And the problem you have — and I’ve thought about this a lot and debated it with some of my colleagues on the right — is coming up with a neutral principle. A principle that would apply to tech companies but that wouldn’t completely undermine all of the antitrust history. And it’s very difficult to find.

It is.

I can think of a few, but it’s not easy at all. And at least if I were in government, I’d want a neutral principle that I could defend as ... Would work for every industry.

This is what Makan Delrahim was talking about when I interviewed him, the same thing.

Yeah. No, absolutely. You need neutral principles. And if you don’t have a neutral principle, I don’t think you want to a start a process where you’re just scapegoating people. That’s kind of Nixonian, actually, in some ways. And I think that’s a very bad precedent.

So what do you expect? If you ... Probably ...

I think you’re gonna see a lot more hearings. So certainly the gavel, the fact that the Democrats have control of the gavel is ...

So privacy.

Some of the nuances of being the majority in the House is you can call hearings whenever you want.


Only the majority can schedule hearings, so I think you’re gonna see Sundar onstage a lot more. And maybe Mark onstage and Sheryl onstage.

Any use to those things, or just ...

Well, occasionally those things backfire. So when you’re onstage, one misstatement can really create a firestorm.


And most of these tech leaders have not been under pressure like that. It’s not like doing a journalist interview.


It’s more like doing a TV interview.

And you’re under oath.

And TV interviews are really difficult, as you ... When you interviewed Mark onstage, for example, years ago, it was very difficult for him.

Yeah, he didn’t do very well in the podcast, though, either.

I think podcasts are easier than TV interviews, because it’s your ... Nixon and the famous Kennedy-Nixon problem.

Right, yeah.

Argumentatively, American history was changed.

Yeah, they don’t like to be on TV anyway.

So in any event, I suspect you’re gonna see them onstage a lot. And that will cause someone to screw up, where they say something inappropriate, not ideally framed, that leads to an investigation, which may uncover some evidence. I think, for example, if you were to subpoena Facebook and Google, you would find documents that are not particularly exciting.

Oh, yeah. My god, their emails have gotta be awful.

They have to be awful.

They have to be awful.

Yeah. And they’d be embarrassing.

Even the small ones that’ve come out, I’m like, “Oh.”

Yeah, well, imagine ... You lived through the Microsoft litigation.

Mm-hmm, oh yeah, that’s exactly right. That’s exactly where they’re gonna get them, with the emails. Because they’re so careless and also on their ... They all have these internal message boards.

It’s impossible to police those perfectly.


As someone who used to run companies, you can try to remind people that one day someone may look at this out of context and, “Please be careful what you write,” but you need people to do their jobs too, which is to communicate directly, succinctly, powerfully, off the cuff, on the fly.

Oh no, Slack and the message boards alone ...

Slack, message boards, this stuff’s gonna be very embarrassing. So the question is what’s the legal doctrine and legal hook that allows someone to subpoena this evidence and get their hands on this treasure trove? Once they have the treasure trove, all hell’s gonna break loose because it’s gonna be very embarrassing.

I don’t think it’s gonna be emails, it’s Slack.

You have Slack, what people think of is instantaneous. The more instantaneous something is, the less they edit.

Right, text, oh God. There’s just so much.

It’s gonna be really ... Facebook Messenger.

So let’s move on really quickly to — Messenger, absolutely — Khashoggi. You and I share the same opinion about this in terms of ... So talk about that.

Yes. Well, to me it wasn’t surprising. Saudi Arabia has been ...

A big investor via SoftBank and others.

Saudi Arabia has discriminated against gays, women, Jews, officially, explicitly, for decades. It executes people who are gay. It executes people who are Jewish. So the idea that they would murder someone isn’t surprising to me. To me, it’s surprising that more people weren’t paying attention to the evils of Saudi Arabia for the last 30 years. Now there’s a reason why they’ve been politically helpful with some other battles that the United States has had in the Middle East, so it’s a marriage of convenience. But that doesn’t mean that Silicon Valley should be participating in this.

We talked on your show, your podcast, two-and-a-half years ago — maybe three years ago, actually — that people boycotting Silicon Valley — North Carolina or Georgia — for somewhat mundane legal changes, whether it’s the bathroom labeling or something equivalent, yet they take money from Saudi Arabia, where it’s illegal to be gay. It’s illegal, like a crime. It’s not just like you can’t go to the bathroom that you choose, it’s like “We’re gonna shoot you.” Or if you’re Jewish, you could be discriminated against in any possible amount of ways. So I think people are incredibly hypocritical here.

I am with you on this. So what will happen? Will they keep taking the money?

Over time, probably not. In the short term, yes, because especially as the economy gets softer in Silicon Valley, the desperation for money among late-stage companies is gonna be higher. But a lot of entrepreneurs are definitely having these debates. And some have been hesitant. I know some entrepreneurs who could have taken SoftBank money that have decided ...

SoftBank, half of its fund is from Saudi Arabia, just to be clear.

It’s more than half, it’s more like 90 percent. It’s half if you allow them to do accounting gimmicks.

It’s a lot.

It’s 80-90 percent if you don’t.

It’s the big funder.

There is no Vision Fund without Saudi Arabia.

Is there any clean money? I mean, where it comes from. They were stack ranking at a party I was at recently which is why ...

Yeah, I know, that’s true. You do have this issue of late-stage money rate. A lot of money comes from Russia, which is no longer acceptable. You have money coming from Saudi Arabia, which, at least to me, should be unacceptable. And you have money from China, which has its own set of issues in that actually the money from China may not be coming for other reasons internal to China anyway. So yeah, you can get money from T. Rowe Price or Lee Fixel or Fidelity.

And where do they get the money from?

They get the money from more traditional sources. Doesn’t mean that all Americans are good or all American money is clean, but fundamentally, if you’re taking money from an investor that trying to effectively launder the money, which is what a lot of countries have been doing, and that’s been propping up some of these late-stage financings.

To me, I think that means you go public, and if you go public you take money from normal, everyday Americans. You have the disclosure obligations and the serious scrutiny and accounting requirements that is associated with taking money from normal Americans, but that degree of disclosure and credibility and scrutiny and transparency is actually a good thing. I think it’s a good thing for your company, actually, to be subject to auditors and subject to reporting.

And where you’re getting the money from.

Well, not just where you’re getting the money from but how you’re performing.

Right, exactly.

I think people are trying to insulate themselves from accountability.

Well that’s true. That’s 100 percent true. Right, I’m gonna ask some ... I asked a question of what I should ask you on Twitter and of course Antonio — he’s such a pain in the ass — Antonio Garcia Martinez, who was also very argumentative person on Twitter, and I enjoy him doing that, “I think we want a video of the probable fistfight.” We don’t fistfight. The gays don’t fistfight, Antonio.

No. Fistfight, that’s so archaic.

So here’s one from Jason Del Rey, actually: “What’s the definition of contrarian in 2019?” Because you’re known as a contrarian.

Well, I like to think from first principles and I like to think for myself, which means occasionally I align with you and occasionally I don’t. I like to be a non-derivative thinker. It’s not always possible. Everybody lives in concentric circles — your professional circles, your social circles — and that obviously influences you.


But I try to have my own perspective. I like to read, for example, in my free time, I read books.


And occasionally it changes my thinking. Yeah, I read old-school books. That’s what I was doing yesterday: Most of yesterday, being Jewish, was not celebrating a holiday, I was reading, I read like three books yesterday.

Which ones?

So I read this really great book on cancer innovation called Breakthrough, which is incredibly impressive in terms of its ability to tell the story of how using your own immune system to fight cancer is better than radiology or chemotherapy. But it had been suppressed by the medical community for a century.


So this has been known since the 1890s that you can, at least in some kinds of cancer, use your own immune system to fight cancer. But a lot of medical people and experts didn’t believe this, and so they actively suppressed this knowledge. And people were dying as they were suppressing this knowledge.

It’s now becoming more mainstream cause there’s now enough data to overcome all the suppression but it reminds you that even “science and medicine” isn’t immune from political fights. And these political fights cost many people their lives because we spent a century not innovating on how to use your immune system to “cure” cancer.

Second one was Mike Ovitz’s autobiography, which I highly recommend. It’s a great tale in many ways.

Why is that? Why do you like it?

So, the thing I liked is, first of all, it’s very candid. It’s intellectually honest, it’s personally honest. I’m kind of an amateur observer of Hollywood and I found it fascinating, how fine the line is between success and failure. How many of these famous movies that won Academy Awards or that I grew up watching and were culture icons were just barely made. Like, how much heroic effort went into packaging them and making them happen, and that the world could be completely different without a couple of very clever maneuvers. And so sometimes you take this success and work some technology just the same way.

That’s a really good point.

You take these successful companies and assume that of course this was gonna succeed. But behind the scenes there was a hell of a lot of work and exploration and tenacity that went into these things.

Oh, especially by him. He was a knuckle-breaker, right?


He’s not well-liked in Hollywood because of that. All right, I’m gonna ask another ... What’s the third one?

So the third one is this new one-volume biography of Churchill. So, I’ve read Churchill biographies before. This may be the best one-volume. I mean obviously, if you really wanna read about Churchill you have to read more than one volume because his history, his life was full of so many ...

Who’s that by? He was ... Twists and turns ...

So many dimensions, yeah. But this is a very good one.

Who’s it by?

Oh man, I’m spacing on the author’s name. It was just published.


It’s doing very well. It’s selling quite well, which is pretty impressive in its own right because it’s still — one volume means 900 pages, you know, in Churchill.


But I really enjoy it so far. I’m only about 100 pages in.

Speaking of contrarians ...

Yes. I mean Churchill changed parties three times, he changed his views on several topics, but on the biggest issues and the biggest moments of his life, he was right.

He was right, okay. So this is another question now. There’s a lot around your politics, because Keith is conservative. Would you call yourself a conservative?

Yeah, I consider myself conservative more than I consider myself a Republican.

Okay, all right. What does that mean?

To me, conservative means there’s right and wrong and that there’s a meritocratic benchmark that one can use to look at ideas.

Okay, now this is a question: “Do you feel loyalty to the USA when so many Libertarians are so anti-government and seemingly pro-Russia?” Okay. And then there’s another one about your friendship with Peter Thiel, who was obviously controversial because of his links with Donald Trump. “Does he think that Thiel is implicated in Cambridge Analytica scandal and will be interviewed by Mueller, go to federal prison, be deported to Germany?”

Twitter’s up to its usual nonsense.

He does have … That’s all right.

I mean, Cambridge Analytica is pretty funny.

Away from Cambridge Analytica, what about your status as a conservative? How do you ...

Well, I think it’s hard to be conservative sometimes, not just in Silicon Valley, because it frustrates me often when conservatives don’t do the hard work of coming up with innovative solutions to real problems. Being conservative doesn’t mean not having a solution, it means having a different solution, often one that’s not based upon government intervention.

So for example, in healthcare. The US government solution to healthcare is very counterproductive in many ways, but it doesn’t excuse conservatives from having better answers. But it sometimes frustrates me that when we’re out of power, that we weren’t developing better answers to a lot of these problems. So I get annoyed once in a while.

Every like 10 years, you see an up-and-coming conservative who has some fresh perspective and ideas and it’s invigorating. Sometimes they do well in the political world, sometimes they get elected, sometimes they get elected and they disappoint you because they’re like any other politician. So it’s like a sine wave of emotions, being a conservative, because you get optimistic about somebody or something.

Who were you optimistic about? Are you optimistic about Trump? Right now?

Oh hell no, I think I called him a sociopath.

Yeah. So you’re not a Republican as the current Republican Party stands.

And I also believe, I think I was the first person perhaps who funded Never Trump.


If you do some research you’ll see that. So no, not the biggest fan.

Do you debate with Peter about that?

Yeah, a little bit. So Peter’s view on Trump ...

He was interested in lack of regulation, like the FDA getting ...

Well, I think some of that area is very much ... Trump’s been actually good at some of that. I think Peter’s view on Trump is pretty simple to describe, which is, if you believe the US was on the wrong trajectory and wrong course, and it was gonna get worse, not better, the idea of rolling the dice with some random person might be better than staying the course.

So, insert a disease.

Yeah, it was a little bit like roll the dice, anything could be better. And you know, to some extent, that’s what Trump’s actually done. On some policies, for example, I believe his policy on China is actually probably more right than wrong. And I think that actually it’s about 50-50 perspective on this in Silicon Valley at the moment. There’s a lot of people who are very liberal who are actually applauding him for challenging China. So I think on some policies he’s been willing to take a perspective.

You know what, sometimes his directionally … absolutely, topically wrong, directionally correct about China but every single thing he says about it is the wrong way.

Well, part of it is I think he has instincts but doesn’t do the work.

I think many of them are racist or ill-informed or something, but usually these opinions are not based in any kind of research whatsoever, it’s just off the top of his fucking head.

Yeah, but his instincts sometimes are right. The foreign policy establishment, for example, can be wrong about a lot of things in the US.


And his willingness to challenge them and not just take for granted what they say ... So, for example, one of the things I applaud him a lot on is moving our embassy in Israel. Very controversial, everybody who’s running for President for 40 years declared that they would do it, and every single one, whether they were Republican or Democrat, wimped out. So it was in the Democrat platform to move the embassy to Jerusalem, it was in the Republican platform since 1980 to move it. But every time someone became President they found every excuse under the book to not move it. He actually promised it and he actually did it.

And it turns out he was right. There weren’t riots, it actually was supporting our ally, who is democratic. People in Israel, by the way, I was in Tel Aviv in August, they love Trump. Even the liberal part of the Israeli population love Trump because they finally found an American President who actually did what he said.

Did the one thing that he said.

He did one thing, one thing he promised.

But you called him a sociopath, what do you imagine is gonna happen right now?

That is a great question. I think he’s running to a box where the people who are most talented really don’t wanna work for him. And that’s a fatal problem with a company ... We deal with this sort of board all the time: We have a CEO who has some talent but it is destructive and all of a sudden that company can’t recruit talented designers, engineers, etc. And the company itself is in jeopardy because of that. I think he has now burned through so much of his administration ...

These are people he hired, by the way. One of the points that’s hard to be forgiving to him — and people should really be attacking him on — is all these people he’s fired for alleged flaws. He hired them all. So someone made the mistake on the way in. If you inherit someone as a CEO, it’s perfectly fine to fire them when you realize they’re not capable of doing the job. But if you’re the one who hired them, too, you can’t be hiring and firing people all day long.

And there’s a lot of them.

It shows you don’t know how to hire.

Right, exactly. So where do you imagine this going?

I think he holds on, and barring a health issue which — he is 72 years old and under stress all the time and doesn’t have all the healthiest habits — I suspect that he runs for re-election. And if he runs for re-election, I think the odds of him winning are actually more than 50 percent. And the reason why is, obviously his favorable/unfavorable rating is poor, it’s one of the worst in American history. But most of the Democrats who are gonna challenge him are very flawed themselves.

It’s been organizational...

So abstractly, you can say, “Oh I can beat Trump,” but when there’s a very tangible, concrete person who has their own character flaws, that’s gonna be very interesting.

It’ll be an interesting organizational challenge for the Democrats, because they managed to pull it off in the midterms, they managed to pull it off in places like Alabama...

But they didn’t have to nominate one person for the midterms.

Right, I know. But in Alabama, for example, there were lots of different places where they’ve won in ...

Some of that was self-inflicted Republican stupidity.

Yes, but self-inflicted is the description of Trump, isn’t it?

Alabama particularly. Oh no, he commits ... The way I describe Trump is he commits the most unforced errors of anybody I’ve ever seen. He finds a way to put defeat in the place of victory. He attacks his own supporters. Democrats ... In some ways, liberals are very lucky because you could have someone with Trump’s views but without his personal character flaws. And that person, if he or she were elected President, would be phenomenally successful. Trump gets in his own way and probably is accomplishing one-third of what he probably could have accomplished.

Same time, it was interesting when I was talking about today, when I was talking about my piece on how he uses Twitter as a government style, they were like, “Well the next President will use it.” Can anyone use it as well? That’s a question of someone who can actually employ it. He employs it really well, not well, I don’t think it’s well, but it’s kinda sociopathic, but it works for him.

It’s inconsistent, so I think he uses it brilliantly sometimes and it backfires on him sometimes.

But in general, I’m saying I don’t know if there’s another candidate who can use it, it can’t be just all of a sudden everyone’s gonna use it well, because everyone can’t.

No, I agree with that. People either, we have these waves, I alluded to the famous Nixon-Kennedy debate where many historians think Nixon lost the election, which was the first Presidential debate televised. Every new technology platform will empower a different type of person. Franklin Roosevelt famously used the radio. Trump is the first Twitter President. There will be a new platform at some point. A different way of sharing information and debating things.

VR? What’s it gonna be?

The person who runs for President and uses that platform natively will probably do very well.

What is it gonna be? VR? AR?

I don’t believe in VR, personally. I don’t think that’s a thing other than maybe gaming. AR possibly but there’ll be some new channel and some political candidate will pioneer how to use it.

There’s a great book, I’m sure you’ve read Neil Postman’s book, Amusing Ourselves to Death, from 1984. I think it perfectly predicted the Trump administration. But the main point of the book and the main arc of the book is that every medium of communication creates a different type of content that succeeds in that medium of communication.


And Twitter is ideally suited for someone like Trump. There’ll be a new platform at some point and there’ll be a different candidate that takes advantage of that.

Who’s the best Democratic candidate to run against him? Off the top of your head. And for Silicon Valley, who’s the best?

The one who would give him the most trouble is Michelle Obama. I don’t think she wants to run, but she would cause him real problems. Most of the other candidates won’t. There’s a couple of things that are implicit in Trump that don’t get enough attention. One is he received a reasonable fraction of the African-American votes for a Republican. And the Democrats cannot win the President’s seat if he gets double-digit African-American votes.

So the basic swing between Obama and Hillary summed up a lot. There’s a lot of complexity. You can talk about anything you want, but fundamentally, it was that he got more African-American voters than she did. Obama got more than she did. And so someone who naturally appeals to the African-American community will do very well and cause him problems.

Secondly, someone who doesn’t need the media attention, but will get it. Meaning someone who’s above a certain status.

I get it, Michelle and Oprah.

Oprah could, I don’t know. I’m more nervous, I mean, Howard Schulz and people like that are probably gonna run. But I think those people who’ve never been tested tend to backfire a lot. It’s very difficult when you’re not used to this level of scrutiny and stress. Meg Whitman, for example, was a horrific political candidate. All of her races, all of the things she’s done in politics, have backfired.

She’s uncomfortable.

It’s just a different skill. Trump you could tell had some skill in marketing from the day he announced his candidacy, when he came down the escalator. I remember sitting at home, it was a Sunday morning and I was watching it here with the guy I was dating at the time and he looked at me and he said, “This guy’s gonna do well.” And I was like, “What are you talking about?” Cause it’s Trump and I grew up in New York City, Trump in the ’80s.

And he said, “You have to watch his speech.” And I watched it and I was like, “You’re right, he’s definitely got a way of appealing to people and framing things that make people nod their head.” And if you have that you can go far in politics, but that’s a skill that most CEOs don’t have.

Nope, you’re 100 percent right. I said the same thing when I was watching him, and I believe I’m the only person who watched the entire seasons of “The Apprentice,” of the people we know.

Oh God, I feel bad for you.

Oh no! I loved it, I’m just saying that ... the gays knew, the gays knew.

Well I used to listen to him on the New York talk shows when I was growing up and one of the things I’d do in high school and junior high school was we’d listen to talk shows on the radio. It was kind of a common way to entertain ourselves, sports radio and other talk shows. And he’d call in all the time. He was a very significant celebrity.


All right, we have to finish up. We didn’t get to talk about Opendoor, but very briefly, where is it headed? Opendoor, which is your real estate company, which is to be able to sell your house.

Yeah, Opendoor is doing phenomenally well.

We’ve got only two minutes.

So, boom. Opendoor allows you to sell your house in 30 seconds online. You put in your address, we give you an offer of what the house is worth, and you get instant liquidity for your house. So instead of spending 84 days selling your house, having strangers come through your house every day, waiting to see if you get a bid, we make everything easy, simple, and immediate. And it’s done phenomenally well. We’re in 18 cities.

And then you have to sell the houses yourself, right?

We have to resell the houses, which is very complicated. We have 1,000 employees now. But we’re in 18 cities and we’ll be in 30 cities by the end of the year.

There’s all these really interesting real estate companies, there’s a whole bunch of new versions.

There’s a lot of innovation in real estate. Construction. We’ve actually funded ...

You talked about this.

... 10 different companies at Khosla in the last two years which are real estate innovation. And so a huge fraction of the American economy ...

How we build houses is antiquated.

It’s totally antiquated. Like walls, think about building walls.

Walls, exactly.

Right now each one’s costing me. There’s no reason it can’t be productized. We’ve funded a company called Vert that’s gonna do that.

Because they’re all pretty much the same size.

They’re all very similar. We’ve funded a company that’s doing 3-D construction of homes called Mighty, Mighty Building, out of Y Combinator. You’re gonna see a lot of innovation on the construction side over the next decade.

How’s your crazy construction fight doing in San Francisco?

It’s actually doing well. It took ...

We won’t get into it.

It took three years, but we’re building the house finally.

All right. I’m looking forward to coming, if you ever invite me.

I will definitely have a party. AC/DC.

Who didn’t you piss off in that neighborhood? Everybody.

No, just one annoyed neighbor. But San Francisco works that way. If one person can put a lot of ...

Oh I love those stories Keith, they make me laugh and laugh.

Anyway, Keith, it was great talking to you. See, no fistfighting, Antonio. We can talk about interesting topics and disagree and it’s easy to do. Anyway, thanks for coming on the show.

This article originally appeared on