The Journal. - A Conversation with Alphabet’s Captain of Moonshots

Episode Date: June 4, 2024

Astro Teller, the CEO of Aphabet’s X, runs a lab dedicated to solving some of the world’s most pressing problems while also coming up with viable businesses. After he spoke at the WSJ’s Future of Everything festival, we sat down with him to talk about A.I., self-driving cars, and the changing economy of tech. Further Watching: -Astro Teller’s Meaning of Life  Further Listening: -Google CEO Sundar Pichai on How AI Could Change Search  -The Future of Self-Driving Cars Is Here  Learn more about your ad choices. Visit megaphone.fm/adchoices

A few weeks ago, we asked you, our listeners, if you had all the money in the world, what problem would you try to solve? And we got a ton of responses. I think the biggest problem that needs to be solved on this planet is the extreme poverty that exists. I'd really solve the freshwater problem. Getting carbon out of the atmosphere. High quality education for everyone. Wipe out student loan

debt. And if I had all the money in the world, I would solve homelessness. We asked you that question because we were about to meet with someone who has access to a lot of money to find solutions for these kinds of problems. My name is Astro Teller. I'm the captain of moonshots at Alphabet's moonshot factory, which is called X. So we asked our listeners, if they had a ton of money, what problem would they want to solve? And they said tuberculosis and homelessness and world peace, hunger, education was a big one. A lot of people said climate change. How do you pick what you pick? There's a lot of problems out there.

We are trying things in education, in poverty, in systemic racism. We're trying to find ways to go after the different problems that build up into climate change, ocean health. There's so many problems in the world, and we're working on any of them where we can find a moonshot. Everything is fair game for us. These moonshots are things like self-driving cars, delivery drones, salt-based energy storage, and of course, AI. But over the past year at X, there have been layoffs, increased scrutiny into the budgets, and more questions about whether or not these moonshots can be profitable. about whether or not these moonshots can be profitable.

Welcome to The Journal, our show about money, business, and power. I'm Kate Leinbaugh. It's Tuesday, June 4th. Coming up on the show, a conversation with Alphabet's captain of moonshots, Astro Teller. Attention all soccer fans. From Orlando to Los Angeles, take to the fields of the USA for your next vacation. Ready to kick off? Discover exciting games and events. Plus, find amazing hidden gems in cities full of adventures, delicious food, and diverse cultures.

You'll love it so much you'll want to extend your stay beyond the matches. Get the ball rolling on your soccer getaway. Head to visittheusa.com. For the past 14 years at X, Astro Teller has been CEO, or as he's known to his co-workers, the captain of moonshots. And do people at X call you captain? Do you insist upon that? I prefer Astro, but sometimes they call me Captain. Captain Astro. No, just Astro, presumably. But some people do refer to me as Captain, but hopefully they mean it in a sweet way. That's cool. I want to pick my title. You should. I totally should. I encourage other people. We have people at X with titles like the Lorax or Lieutenant Troublemaker.

Oh, right. Oh, I, at an Ultimate Frisbee Tournament, was the director of Whimsy and Frivolity. Fantastic. Yes. That's a particularly good one. So, Captain Astro, you are joining us after speaking at the Future of Everything Festival. Castro, you are joining us after speaking at the Future of Everything Festival. We are in a room that's quite loud, and we're going to use AI to get rid of the loud, but some of it will remain.

Awesome. I look forward to listening to it. Thank you for having me. X was started in 2010 as a research lab, and it has derived inspiration from a famous moment in history, back in the 60s. We choose to go to the moon in this decade and do the other things, not because they are easy, but because they are hard. That was President John F. Kennedy, pledging to send an American to the moon. In the tech industry, the phrase moonshot is used to describe similarly ambitious projects. How do you define a moonshot? So a moonshot for us has three basic components. One, there has to be a huge problem with the world

that you can name and you want to solve. If you can't, then it's kind of an academic exercise. Because remember, these things are meant to ultimately become enduring businesses in the world. So you'd better be solving some real, at least aspire to solve some huge problem in the world. Two, there has to be a radical proposed solution, something that's 10 times better than what's out there. Because if what you're proposing is either not a solution or is not radical, then why are we doing it? And then the third is there has to be some kind of core breakthrough technology that both gives us a reason to believe that that science fiction

sounding product or service that you're imagining that could maybe solve that huge problem could actually be built. It doesn't make it super likely, but it's a testable hypothesis. If you can't name some technology, then how do we get started? How do we know in the next six months if we're maybe on the right track? One of X's well-known moonshots is Waymo, the self-driving car company.

How did Waymo come up through X? And how did you identify it as a thing that you wanted to work on? So huge problem with the world. More than a million people die in car accidents every year. More than a trillion dollars is wasted every year just sitting in traffic, doing nothing, scraped paint and like body shop work. So that's the huge problem with the world, or those are some of the problems with our current system. Radical proposed solution, it may not seem quite as radical now, but 14 years ago when we first proposed it, the idea that cars were going to drive themselves was pretty radical. So let's take a very small Manhattan project where we get a bunch of

these people together, give them a really hard set of things to try to do. And that will either create evidence that it's too early if they can't do it and then we'll throw it away. Or if they can do it, it will give us evidence to double down on that experiment. So we gave them 10 different hundred mile trips that we wanted them to make each of those 100-mile trips in around the Bay Area at least once with nobody touching the steering wheel. And we're going to have paid drivers sit with their hands right by the steering wheel for all the driving we do out in the world to make sure it's safe so we're not endangering the public, but we can be learning on real city streets. So that's how it started. And when that team accomplished this set of 10 hundred mile trips, we said, oh, wow, OK, maybe this is more ready than we thought. After this testing period, Waymo was eventually spun into its own company under Alphabet in 2016. This is what X calls graduating. Waymo now operates in cities like LA, Phoenix, and San Francisco, where it

offers self-driving rideshares. How do you budget a moonshot? So in the very early days, we have a process. If you're coming up with something, we have lots of labs here. Your time is already paid for. If you need a few tens of thousands of dollars, you know, but that's what you get. You get months, maybe. And actually, first, you should get days, which is, can we kill your idea with a white paper or on a whiteboard? Can we prove that it breaks the law of physics, that it wouldn't net to the positive for the world? If we can convince ourselves just philosophically and abstractly that it's not going to make it, awesome. We high-five each other, we throw your idea away,

and move on. If we can't, sure, here's a few tens of thousands of dollars. Not more than that, what is the cheapest, fastest way we could get a piece of evidence that will either show that you're a bit less crazy than it looked like or a bit more crazy that's less right than we thought and we're happy to pay for that amount of money for turning over that card either way but what about the fact if i may that like you all are a bunch of people located in a particular place, in a particular country on the planet, with a particular worldview. And you're making these decisions. Well, again, I'm not making any of the decisions.

You're, let's say you work here at X. And believe me, there are plenty of Xers who say what you just said. And I say, awesome. Go out in the world. Talk to those people. Either find their ideas. It doesn't, I'm going to give you just as much credit if it's their ideas, if it's yours. Or even if it's your idea, go get contact with the real world as soon as you can. Go do it in

New Zealand. Go do it in Kenya. Go do it in Peru or Chile. Go actually, in a sandboxed, safe way, work on whether you're right with the people who are actually going to be affected by it so we can learn from them. So if it's going to be bad, we can throw it away. Even if it's bad for surprising reasons, we can't see it first. Then we're including the whole world in our process. Who are you responsible to? Is it like Google's board and shareholders? Yeah. I mean, I'm responsible to Sundar and to the Alphabet Board of Directors. They understand that what we're doing is very long-term in nature and that a good bit of what we try won't work out. That's sort of inherent in the process. were just talking about, which is, is the total rewards that are produced over the long run

very big relative to the costs that we're spending? As long as that's true, it's working and we should continue. And then we just sort of refine over time how we get even more audacious and turn up the rigor in the right ways. So we're moonshot and factory. But that moonshot factory is now operating in a tech landscape that is focused on cutting costs. That's next. Your teen requested a ride, but this time, not from you. It's through their Uber Teen account. It's an Uber account that allows your teen to request a ride under your supervision with live trip tracking and highly rated drivers. Add your teen to your Uber account today. This episode is brought to you by New Balance Running.

New Balance believes if you run, you're a runner. Whether you're going for your first ever run around the park or going for your personal best in a marathon. Speed, strength, stamina. Whatever goal you're working toward, New Balance has the running shoes, clothes, and accessories to push your run further and help you run your way. Find yours at newbalance.ca slash running. New Balance. Run your way. Find yours at NewBalance.ca slash running. NewBalance, run your way.

For more than a year, tech companies have been cutting back, laying off workers and focusing on profitability. And as part of this, X has been hit too. Earlier this year, dozens of workers were let go. A few months back, X was part of some layoffs at Alphabet. And in an email to staff, you said part of the goal of the layoff was to spin out more projects as independent companies. Is that a change in your thinking? as independent companies. Is that a change in your thinking? Yeah, it is an opportunity. It's a new pathway. So exactly what I was just describing. Some stuff will still go to Google. I'm sure at least occasionally we will make other bets. We have been putting stuff out into the world. We've maybe set up eight or nine companies on the outside

already, but we're getting better and better at that because more and more of our stuff is in the climate change space. And I see this as a great opportunity for us to get more of the enduring businesses we hope we can create out into the world, supported by market-based capital so they can sort of go run for the races. Is it a retrenchment of Google's ability to fund moonshots? I think it's an amplification of what we're doing so that we can do more of the thing that we're best at, which is originating the moonshots, and let more of the world work with us in amplifying the moonshots. Does the moonshot business still make sense given the sort of retrenchment in tech?

Well, you might be asking the wrong person since it's like literally my business. So you say yes. I say yes to that. I mean, it makes sense for two reasons. One, it's the only thing we do, so I'm not sure what the alternative would be. But also, it's working. So I think we need to respond to the times. X should continue to evolve and is continuing to evolve and try to get better. We shouldn't be sort of thoughtless to the fact that the world is changing. But the thing that we're doing is working. That ratio of the reward relative to the costs is plenty good enough to argue to continue it and maybe even to do more of it.

good enough to argue to continue it and maybe even to do more of it. Given the current financial forces in tech, we wanted to know if Astro feels more pressure from Alphabet to make profitable companies or to solve the world's biggest problems. You know, when we started 14 years ago, X was specifically set up to help Alphabet get a lot more surface area with the world. At the time, it was largely a search company. So they now have a lot more surface area with the world. We have five other bets that came from X. They will happily take something once in a while back to Google when it's really important for Google, but they have enough now. And since a lot of what we're doing is in the climate change space, we cannot wait. We need to get out there, market-based capital, people and strategics who understand things like project finance. If we're going to be making things the size of factories to try to help with climate change in various ways, that's not necessarily best done inside of Alphabet. So Alphabet can have the bragging rights and a strong minority ownership in these things.

But setting them up on the outside so that they can sort of go is actually much healthier. To your point, Alphabet gets to help the world in the way that it has aspired to for X. And then sort of get the halo effect from doing that. For sure. get the halo effect from doing that. For sure. Okay, so now let's turn to AI. As I said earlier, like we asked our listeners what would they want to do if they could do moonshots?

Basically, that was the essence of our question. And, you know, they say things like cure cancer, which is something that you also hear in the conversation of like maybe AI can do that. Is that the path of AI? I'm not sure I would say the path of AI is to cure cancer. AI is a lever for our minds. It's going to help us solve a lot of problems that have been very hard to solve before by helping us explore millions or even billions of possibilities

in a way that a human couldn't manually explore all of those things. And in any circumstance, including in drug design, can artificial intelligence supercharge our ability to be innovative in those spaces and to take down some of humanity's biggest problems? Yeah, I'm sure it can. Will AI take over the job of the captain of moonshots? It might elevate me. I hope it elevates me.

My favorite way of thinking about it is the introduction of spreadsheets, like on a computer, destroyed the vocation of bookkeeping. But it massively expanded what it means to be an accountant. So the jobs changed. There was some retraining that had to happen.

But in the end, A, the work was more interesting. There were actually more jobs and we could do more interesting and powerful things with those same people. I would like to think that AI will do that for many, if not all of us. Right. It's like the bookkeepers were bummed, but the rest of the world was okay with it.

Well, maybe, but I'm also suggesting a lot of the bookkeepers became accountants. There was retraining, and I'm not just sweeping that under the rug. There is things that have to be learned, but it is actually true that most of those bookkeepers became accountants. I mean, you're in the technology moonshot business.

Guilty. You're looking for technology. But like, I mean, the other question is, is society organized right? We're into a private ownership society and not a sharing society. Should humans share more? I mean, I would love that. There's a lot. People are used to getting deliveries now. But one of the things that excites me about the future of the drone delivery business is that we could be moving from an ownership society to an access society for a lot of things.

Right now, you have a hammer at home, and I do too. I have two. See? And everyone around you has a hammer. Everyone around me has a hammer. We don't need that many hammers in the world. We could share one ten thousandth of those hammers between us and never even notice it if somehow magically they could just appear when we need them and be taken away essentially for free. Drones, I hope, are an example of how we could move to a place where we're just sharing things more. Like, I have a lot of books in my house, and no one reads them because I get to them very, very seldom. Why can't we share those books better?

Well, we have those little libraries. I dream of a little library, but for tools on your block. We all don't need to have the same tool. But what if we had Harry Potter owls that could just move them back and forth between our... But the idea of drones with hammers like flying around the place is also like we're going to be wearing hard hats. No, you definitely won't. Okay. No.

I think you're going to be pleasantly surprised. I invite you to come and see one of our drones making deliveries. I think you will be somewhere be pleasantly surprised. I invite you to come and see one of our drones making deliveries. I think you will be somewhere between pleasantly surprised and shocked. All right, Captain Astro, thank you so much. Oh, it was an absolute pleasure. Thank you for having me. Such a pleasure. That's all for today, Tuesday, June 4th.

The Journal is a co-production of Spotify and The Wall Street Journal. Thanks to all of you, our listeners, who sent in voice messages and emails. We love hearing from you. If you want to hear more interviews from the Future of Everything Festival, hearing from you. If you want to hear more interviews from the Future of Everything Festival, subscribers can find them at wsj.com slash future24. Thanks for listening. See you tomorrow.