How I Built This with Guy Raz - The future of driving is autonomous with Dmitri Dolgov of Waymo

Episode Date: November 23, 2023

Waymo Co-CEO Dmitri Dolgov is convinced that his company’s vehicles are better at driving than any human. Dmitri has spent thousands of hours riding in them, and recently Guy had the chance to try one out as well...This week on How I Built This Lab, Dmitri recounts the decade-plus journey of building Waymo into the world’s first company to operate a fully-autonomous ride hailing service. Plus, how Waymo’s approach differs from Tesla’s, and Dmitri’s take on when we’ll see more AV’s on the roads than human-driven cars (spoiler: sooner than you may think!)This episode was produced by Kerry Thompson with music by Ramtin Arablouei. It was edited by John Isabella with research help from Chris Maccini. Our audio engineer was Neal Rauch. You can follow HIBT on X & Instagram, and email us at [e-mail] Privacy Policy at https://art19.com/privacy and California Privacy Notice at https://art19.com/privacy#do-not-sell-my-info.

This podcast is brought to you by Squarespace. I talk to entrepreneurs all the time who are looking for a way to upgrade their digital footprint. Well, whether you're just starting out or you're scaling your business, Squarespace is the easiest way to build a great website that stands out. It's an all-in-one website platform that gives you everything you need to claim your domain, showcase your products, and get paid. Anyone can use Squarespace's cutting-edge design tools to build an online presence that truly reflects what makes your business special.

There are templates, intuitive drag-and-drop editing, and even an AI-enhanced website builder. Then, Squarespace's built-in analytics tools help you make smarter business decisions. Review website traffic, learn where to focus engagement, and track revenue all in one place. Looking to grow your business, Squarespace even offers fast, easy business financing

through Squarespace capital. Go to Squarespace.com slash built for a free trial. And when you're ready to launch, use offer code built to save 10% off your first purchase of a website or domain. Loans issued by Celtic Bank and serviced by Stripe, all loans subject to credit approval. This show is in partnership with Airbnb. This past summer, I took my family to Vienna, and it was incredible. We spent our days wandering the old streets, stopping for coffee and pastries, visiting museums, and just soaking up the history of one of the most beautiful cities in the world. And one of the things that made the trip so special was the home we booked on Airbnb.

It had tall windows, beautiful old details, and plenty of space for all of us. And being in that home on Airbnb, right in the middle of Vienna, walking distance from so much of the city, made it feel less like a visit and more like we were actually living there. Plus, taking a trip is the perfect time to host your space on Airbnb. Your place with all of its personal touches and its amazing location could make someone else's vacation even better. Your home might be worth more than you think. Find out how much at Airbnb.ca.com slash host. Hello and welcome to how I built this lab. I'm Guy Raz. There's no driver. How crazy is this?

That's crazy. It's their first time ever. You're going to look at this as an adult when all cars are driverless. So you're going to be like, yep, I did this driverless drive with Daddy. So it's probably hard to tell from the sound alone, but what you're hearing is audio from a ride. I recently took in a Waymo Autonomous taxi. My son and I were in San Francisco heading across the city. These taxis are now all over San Francisco and Phoenix. And yes, there is no human driver.

There's no human in the car at all, unless you count me and my son, who were passengers. It's a ghost driver, and if I'm being totally honest, they're amazing. The Waymo taxi is now available 24-7 in San Francisco and Phoenix, and it operates similar to Uber or Lyft. You pull out your phone, you fire up the app, order the car, and it magically appears within minutes, except it's fully autonomous. Waymo was spun out of Google a few years ago, but the project to build autonomous vehicles at Google actually goes back to 2009. One of the engineers on that project was Dmitri Dolgov. He's been with Waymo from the very beginning, and today he's the company's co-CEO. Dmitri was born in Russia, a child of two physics professors.

He went to high school in the U.S. and eventually returned to Russia to attend the Moscow Institute of Physics and Technology. Dmitri then came back to the U.S. to get his Ph.D. And in 2006, he took part in a competition called the Urban Challenge, which was sponsored by, by DARPA, the U.S. Defense Agency. The idea was to build an autonomous vehicle that could make it through a series of obstacle courses. The grand challenge was to drive 150 miles through a desert, a completely static environment,

but until then, robots were not capable of doing that. So it took two tries in 2004 and 2005 to accomplish that. So the next step was, let's make it a little bit more interesting. Let's make the environment dynamic, right? Let's add some rules of the road, stop signs, other vehicles. but you know do it in a controlled environment in that MOC city. So that was the challenge. And we built a car to do exactly that.

We equipped it with a bunch of sensors, you know, lighters, lasers, radars, cameras, you know, computer, and then wrote software that would allow it to follow the rule, understand and follow the rules of the road and interact with other, you know, dynamic actors, whether, you know, human or other robots. So I'm looking at one of the vehicles that you helped put together. It's called the Junior. And it's a Volkswagen Passat, I think. And I think came in second place.

This is like 2007. And was the underlying technology basically what we're talking about today? Was it pretty similar? Or was it like a crude version? You know, yes and no. It depends on how you talk about like big car plus sensors, plus computers, plus software. you know, that stays, right? But of course, you know, it's what, almost 18 years and then everything has changed, right? The sensing technology has gone a very, very long way. Computers, you know,

have evolved software, right? All the breakthroughs, especially in AI. So, you know, that led to many, many breakthroughs in the area of software. So now the system we have kind of has the same components, but all of the components and how they work together is, you know, qualitatively, drastically different. Yeah, it's amazing how many, you know, these DARPA challenges, how many sort of self-driving car companies they spawned. I mean, we've talked to Kyle Vogt of Cruz. He was in, he competed in the 2004 and 2005 challenges.

Dave Ferguson of Nuro, which is another company. He was part of the Carnegie Mellon team. Chris Ermson, who also was obviously worked with you at Waymo. he's now with Aurora, was also part of Carnegie Mellon team. It's amazing how many, like you knew all of these people. And this is, as you say, it was a small community of people working on this huge challenge to basically create vehicles that could be fully autonomous. That's right.

That's right. And, of course, I worked very closely for many years with Dave and Chris and they're great friends. So, yeah, it's a lot of innovation and a lot of companies and progress. came out of those early days of the RPA-Gen grand challenges. You and Sebastian Thrun and Chris Irmson, you all went to go work for Google. And you were part of that founding team, which was at the time a secret project, Google's self-driving car project. And at that time, it's amazing to think 2009, because it seems like ancient history now, you were given a charge by Larry Page and Sergey Brin, the co-founders of Google.

You had two years to essentially accomplish two things. Do you remember what they were? Yeah, I remember them very well. The first one was to drive 100,000 miles in autonomous mode. That was way more than orders of magnitude more than what anybody has done at that time. And the second one, and actually that one turned out to be more interesting and much more challenging, was to drive 10 routes. Each one was about 100 miles long. And they were very carefully selected, I think, by Larry and Sergey personally, and they were fairly devious in how they created them to make it interesting.

And the goal was to drive each one from beginning to end in full autonomy, so with no human intervention. which at that time seemed almost impossible. I remember we had a lot of people, even experts in the field, kind of laugh at us when we attempted it. Two years to accomplish this, I think you guys finished it with three months left. Like you actually, I mean,

going into this project, do you remember thinking, we're never going to do this in two years? I was actually, you know, maybe naively, but fairly optimistic. Right. And it was hard. and they're all different.

And yes, there are many moments where we would, you know, finish one route. And we would think ahead of like what would it take to do the next one. And, you know, we would write some software. You know, we would, you know, collect some data and test it. And then we would actually try it and have, then we would, you know, hit a bunch of, oh, crap moments of like, okay, wow, this is way more difficult than we expect it even those early days. So when you did finally accomplish it, did you, what? Were you guys able to have a party and celebrate or was it just like, nice job. Now we got to keep it quiet. Yes, both. We celebrated and, you know, it felt like it was a big accomplishment. But yeah, we kept it quiet.

But, you know, like, that's what made it so incredibly fun. I guess, you know, that phase was one of my favorite phases of the whole project. It's like the early days of a startup, you're up against what might seem like an impossible goal, but it's very, you're clearly defined, right? You have a very clear milestone, a very clear goal. You are singularly focused on it. You have a small team. Everybody's working around the clock 24-7 and, you know, sprinting together. And every day and every hour, you are prototyping, right? You're learning so you can move incredibly fast. And every day of every hour, you're making amazing progress and you're learning your thing. So that was, you know, that was a tool. It was a blast. And every day probably presented a new series of challenges.

Do you remember what was one of the hardest challenges that you had to figure out? Like something that just took longer than you thought that you just, it just didn't, it wasn't coming together quickly on working on trying to make this happen. Do you remember something you worked on that you just, it was like such a hard problem to solve? There's a one, I mean, we had we had a number. Some were more fundamentally challenging. Some were kind of even comical subbacks where, you know, you would do a ride. I remember one one of the routes was driving on all of the freeways and crossing all the bridges in the Bay Area. And, you know, it's about 100 miles.

And you go through all of the challenges. And, you know, we're attempting this drive. And, you know, the car is doing a good job. You know, it's handling merges and, you know, you or somebody else was sitting in the drivers behind the wheel just in case. That's right. That's exactly right. exactly right. And that was kind of what made those early days a lot of fun is that if you do everything, right, you would be, you know, putting some hardware in the car, you would be calibrating the

hardware, then, you know, the next hour you're writing some software, you know, whether it's tools or something for the car to actually make decisions. And then, you know, you get in the car and you, you give it a try. So anyways, on this route, we are driving along, yeah, is doing a job. We get almost to the very end. Right. So at this point, you're, you know, holding your breath. You're waiting for, you know, the last, you know, mile or so of that, you know, 100 miles. And the way that particular run was supposed to finish is that we're coming down the Golden Gate Bridge, you know, into the city. And there's a set of Talbos at the, you know, when you go through the bridge. Yes, as you go into the city.

That's right. At the end of the Golden Gate Bridge. And they're narrow because I use them almost every day. That's exactly right. And, you know, the one that our car wanted to go through was closed at the time. And it's just not something that we ever encountered and we thought about. Oh, right. Because it didn't, it couldn't figure out like the X, the red X or the green arrow.

Like it didn't, you couldn't recognize what those meant. That's right. And there was actually a gate. So, you know, it would stop. But it was like, it would not change lanes and pick a different one. Like, oh, you know, my God. Now back to square one. Because probably every, you know, this impossible to imagine.

every eventuality, but something's going to come up. That's exactly. That's exactly right. And we would have to deal with high speed traffic on freeways. We had one route that went and kind of took this windy road from the Bay Area to Highway 1, through the mountains. And we were driving along. And then a bicycle fell off the truck in front of us.

So that's not something that that time the car could deal with. We had another route that went through downtown San Francisco. Francisco and the famous Lombard Street. That's very narrow, very windy, and has some of the most adventurous pedestrians and tourists in the world. So challenges like that, right? So that's what made it so challenging and so interesting that it was the kind of the breadth of the experience. And probably every single ride posed a new series of challenges. Like for example, some lanes are shoulder lanes, but then during heavy traffic, they're open for driving. But the lines are painted on them in such a way that it doesn't seem like a lane. Like a human could figure that

out. But an autonomous vehicle probably at that time was like, wait, this doesn't make sense. The lines don't make sense. They don't align. Maybe the car was confused. That's right. That's right. And that would be not the kind of situation or condition that at that time, we were able to solve robustly. The goal there was to learn and do, like we had to do the route once. Right. So if you fail at something, think you would go, you know, improve the system and you would try it again. So there was, you know, a very well-scoped milestone because, you know, it, a hundred miles, you know, is nothing if you want to build a production system, right? But it, in those early days, it was long enough

that it actually forces you to very deeply think and tackle some of the most fundamental, most important, you know, challenges that exist. We're going to take a quick break, but when we come back, more from Dimitri Dolgob on how the engineers at Waymo decided to go after the big win and create a fully autonomous vehicle. Stay with us. I'm Guy Raz, and you're listening to How I Built This Lab. Hey, Guy Raz and HIBET listeners, my name is Richard Crowdy, and my favorite episode as a business professor is the founding of BET, because it summarizes business schools perfectly in keeping revenues up and keeping costs down.

It also shows that a complex company, like a media company, can be founded by an individual that bootstraps it with the proper type of innovation. Bob Johnson is also a legend, and he's hilarious. I've actually been listening to my top 15 to try to come up with a list for my business students. So thank you, Guy Raz, and cheers. If you want to share your favorite episode of How I Built This, record a short voice memo on your phone telling us your name, where you're from, what your favorite episode is, and why?

A lot like the voice memo you just heard. And email it to us at hibt at ID.wonderry.com and we'll share your favorites right here in the ad breaks in future episodes. and thanks so much. We love you guys. You're the best. And now, back to the show. Welcome back to How I Built This Lab. I'm Guy Raz. My guest today is Dimitri Dolgob, the co-CEO of the autonomous vehicle company, Waymo. I guess in those early days, the Google strategy was really on creating driver assist technologies, not necessarily fully autonomous cars, but cars that could make it easier for drivers to navigate. basically which you've got in like a Tesla now or some other cars that have driver assist technologies. But I guess around 2013, there was a conscious decision to pivot the focus because most of your testing was done on freeways.

That's right. But what I think you guys and most people involved in the space realized is that to really create complexity, you've got to test these in cities. They've got to navigate city centers and stoplights and left turns and intersections. And so what happened in 2013 to kind of get you guys to focus, to shift your focus on building a fully autonomous vehicle instead of just driver assist technology? Right. I would say there are a number of things. As you mentioned, the first phase was just learning and understanding the complexity of the problem. Then we said, let's try to build the product.

And that was the driver assist system. We tried it. That was actually reasonably successful. We had ran a pilot where we gave these cars to about 100 Google employees. and they could use them to commute and take them around on their daily trips. So that was around 2011, 2012. So then in 2013, we looked at the whole thing and we made this decision to go after full autonomy. So the reasons were several.

One was what we learned from people using the driver assist system. They would overtrust it. They would put on makeup and text and one guy, actually fell asleep. You know, car handled everything fine, but that does not what we wanted to see. So that was one of the reasons.

Another one was that, you know, we actually made progress on the core, you know, most difficult aspects of this driving task in on surface streets.

So that, you know, gave us a bit more optimism of, you know, going after, like, the big prize. And this also at the same time, the field was moving forward. So if we wanted to really make an impact

in, you know, make, you know, improve transportation, globally, I thought, okay, you know, the field, you know, driver systems, you know, will develop their, you know, other companies. You'll be, you know, pushing in that direction. But let's play to our strength, right? And let's go after the kind of uncompromising, you know, big win and, you know, big unlock in the space, which is full autonomy and actually building, you know, what we now call the waymo driver that's responsible for the whole task of driving

beginning and to end with, you know, no human behind the wheel. What do you remember about the conversations and the ambitions? Was it like, hey, you know, let's build the technology for fully autonomous vehicles that we can then maybe one day license? Or was it, no, let's basically become a car company. Let's like make fully autonomous technology and purpose built vehicles and, you know, and eventually sell these two consumers that they can have for themselves. Was it one of the ones? Was it one of either of those ambitions discussed? I don't think we ever seriously entertained building a car. It just never made sense to me or others to do that. Like we're not a car company. Building cars hard. There's many companies that spend 100 years getting very, very good at it.

We did in the early days of design a low-speed vehicle that we called the firefly because we needed to take that first. That was our zero to one moment of full autonomy. It was a prototype that you showed publicly should display it in 2014. But that wasn't the goal. It wasn't, we're not going to build cars. It was, but we're going to build the technology that can power any car to become autonomous. That's exactly right.

That's exactly right. And it was the evolution of the thing. Then then crystallized in the Waymo mission, which is to build the Waymo driver. And then, you know, over time, time deployed in different products and different commercial applications, you know, whether right healing, trucking, deliveries, and eventually, you know, personally owned vehicles. And that's the, that's the path that we, you know, set for ourselves in the path that we've been on since that pivot in 2013. Now, in some ways, the real work begins because between 2013 and let's say,

2023, right, which we're going to get to because it's incredibly exciting what is going on now in San Francisco and Phoenix and a couple of other places. But that 10-year period, probably all of a sudden you're back to kind of start-up mode because you've got to get these cars, this technology to be absolutely foolproof perfect, but in complex environments like San Francisco, which I think is next to New York is one of the most complex driving environments in the United States. Tell me a little bit about the process that then began. Is that what happened? Was it a shift to like, okay, let's see what these can do in cities? That's exactly right. We didn't start with the full complexity of San Francisco in those early. So we can kind of think

of that 10 years as maybe three phases and that correspond to three generations. of our technology, three generations of the, of our driver. You know, that first one was on that low-speed vehicle, you know, the Firefly, and that's what we called the third generation of our driver. And by the way, just to clarify, the Firefly probably was designed to be like a campus type of vehicle, right, like in a university or an office park, not necessarily in the city. That's exactly right. I mean, it was a low-speed vehicle. You can only move up to 25 miles an hour. So we're thinking,

you know, maybe large retirement communities or campuses. Like a shuttle. That would just, you would get off and on. That's exactly right. And at that first phase and that third generation of our driver, the goal was to actually build something that can take, you know, a fully autonomous trip. And we actually did that in 2015. And, you know, we put a friend of our project, His name is Steve Mann. He happens to be blind. And in 2015, he took the first autonomous ride in Austin, Texas. It was the first public person that's out of Google to take a ride.

I remember this. This is in the YouTube video of it. And yeah, it was in Austin. And yeah, I mean, that was kind of a big deal. Yeah, so it was a huge moment when we're able to do this first ride in 2015. There was a big celebration after that. Yeah, I can imagine. Okay, so at that point, you and your team had built this third-generation version of the Waymo driver

that could actually take a fully autonomous trip, which is a big deal. And you were essentially betting that this vehicle would keep passengers safe on a real city streets, right? Like, what went into the safety design? That's right. That's right. it was also that early phase is what kind of forced us to start thinking about this very fundamental question. I go, what does it mean for a self-driving fully autonomous vehicle to be ready?

I come and set out on this path in 2013. I said, okay, you know, we're going to, you know, we're going to need a car. It needs to have, you know, a bunch of safety systems. Does that exist? No. Okay, well, let's design one and work, you know, with partners to manufacture it. So we put a lot of thought and work into making it. it safe. It had like a foam, you know, front, it had a plexglass window. The sensor

pods were attached with magnets so they, you know, could detach if something were to happen. And then we, the sensors at that time didn't have the level of reliability or capability that we would need, that we would trust, you know, to go to full autonomy. So we build, you know, that generation had our own lighters and, you know, custom sensor suite. And then, of course, the software that we had to build, like that none of that existed. So let's talk about the technology for a moment, because you mentioned LIDAR and radar, and some of these things, we know what they are. LIDAR, for example, uses basically light lasers to measure distance.

And you've got radar technology cameras all over the car. Can you just kind of break down how they work? I mean, a lot of people who drive Teslas, for example, I have one. They, if we, if you use what Tesla calls full self-driving, which is not really full self-driving, but they rely primarily on cameras. They don't use LiDAR or radar technology. Tesla argues that that is the other ones are just redundancies that are unnecessary. Tell me how your technology works and why you think it's better. Well, these sensors are, they kind of have fundamentally different and complementary physical properties. So cameras give you the high resolution and the richness of color.

And how many cameras, by the way, on a Waymo vehicle now? On the current generation, we have, kind of all of them, including internalists, we have 29 cameras. 29 cameras, okay. But they're passive. That's right. Somebody else has to bring the light, whether it's your headlights or the sun. radars and lighters in contrast are active sensors. So they blast their own energy out on the world and then they, you know, get returns and from

that they can make sense of the environment. And they use, you know, different wavelengths. So they can kind of punch through fog or rain much better than a lighter or camera. I see as a human driver, I'm sure you can relate to how difficult it may drive due to drive at night, right? For example, at night, if you have, you know, somebody, you know, an oncoming car, with their headlights on, high beams. It kind of blinds you, right?

It's very hard to see. It doesn't affect, you know, radar or lighter. Or similarly, you know, driving in, you know, dense rain or fog. So this is why, you know, we think kind of using all of the sensors and, you know, fusing them in our, you know, AI and ML models, so that you can kind of extract the best signal and see the world in the best possible way gives us an advantage. And, you know, really, you can build, you know, a prototype or can, you know,

build a driver assist system without needing all of that extra capability, extra redundancy. But if you really want to take the driver out and go for full autonomy, it gives you a boost. All right. So just to clarify, like, basically, you have the first successful ride with a third generation vehicle in 2015. And then in 2016, way much spins out of Google as an independent company. And then the next year, you become the first company to start, like, regularly operating these AVs in Chandler, Arizona. These are even more advanced in the previous version. What was the idea of, like, hey, let's get these in really good shape. And then, you know, we'll turn them into like Uber's.

That's right. That's right. That point, we were, you know, pretty clear that that was going to be our first deployment, our first product. And essentially, that's what we. launched in Chandler. You know, that that's when we created the Waymo One product and the Waymo One application. And in 2018 in Chandler, we started offering the service to external riders. And then in 2020 in Chandler, Arizona, we launched, you know, the first fully autonomous right healing service that was open to the public. Anybody could just download the app and

call a car. That was our fourth generation, you know, Pacifica minivan. The empty car would go show up and take anywhere. So then we made this decision that it was not the best fast forward to kind of incrementally grow and scale that system. We made the decision to make a hop to what we call the fifth generation of our driver. That's on the JLR IPases, Pacificus, and a whole new... The Jaguar. That's right. That's right. And, you know, we said, hey, let's take a big step. It's going to be, you know, a different car. It's going to be a new generation of hardware. It's going to be very different software with big bets on in a state of the art AI and

okay let's go after the full complexity of the full complexity of downtown Phoenix and downtown San Francisco so that's what we were working on you know on that time frame to then on that new generation or the Waymo driver to launch the Waymo One service we're going to take a quick break but when we come back

how Waymo One works today and Dimitri's take on the future of autonomous vehicles stay with us I'm Guy Raz and you're listening to how I built this lab. Welcome back to how I built this lab. I'm Guy Raz. My guest is Dimitri Dolgob, the co-CEO of the autonomous vehicle company, Waymo, which has started to roll out Waymo I, its autonomous ride hailing service in San Francisco.

The city gave you permission to be a ride hailing service for 24 hours a day, and I've used them probably a dozen times now. They're Jaguar SUVs, all electric. and they, driving around the city, and you just order it like an Uber. You just go out and you, you know, go on your app and it comes and then you unlock the door with your phone and you get in. And you hit start and it goes. And I can't even tell you how many times the first question I get from people is, aren't you scared? Aren't you terrified when you get in one of these things?

And so I have my answer. So what's your answer when people say that? to you. Like, this just seems terrifying for a machine to drive you around a city. Like, it seems so scary. Like, you could, you know, you could just go haywire and drive off a bridge or something. What do you, what do you say to people when they ask you that? Oh, you know, I at this point, I think I'm much more anxious about human drivers than I am about the Waymo driver. So I have, you know, full confidence. But of course, you know, I've been in those cars, you know, many, many times, hundreds, thousands times over the years.

But what we see with other people, like, it's very natural to have that anxiety. It's a very different, very new system and new product. But what we very consistently see, and I guess I would love to hear if that matches your experience, but once people get in the car, you just, you know, after a couple of minutes, they get very comfortable and they go back to, you know, doing whatever they want to be doing, like, you know, back on phone and check yeah exactly i know i know i went i took my son in it um and drove across san francisco and yeah within like 30 seconds or 45 seconds the first time ever he's in there he's like back looking at his phone and it's true it's you get in it and then it goes and it's really cool you're looking

out looking at the steering wheel turn for the first 45 seconds and then you're done and then you're just like going back to what you were doing answering emails or whatever but it is it is amazing that it, you know, how sort of bizarrely ordinary it feels after you, after a couple of seconds. Yeah, but that's great. Yeah. I mean, that's, that's incredibly exciting. And I think some people kind of draw this parallel that, you kind of, your brain switches to passenger mode. Yeah.

And yeah, you do get all of these benefits of, you know, privacy, whether you want to have a conversation with somebody in the car that you're in the car with or you want to make a, phone call. And I mean, yeah, that's exactly, you know, why we're so excited about this product. All right. So, Dimitri, in general, right, I'm optimistic when it comes to technology. But I have to also admit that healthy skepticism is important, right? I mean, we've been promised technologies that are going to make our lives better and change the world, only to be, you know, sorely disappointed by them. I think it's really amazing what's happening with a autonomous vehicles. But I guess I wonder, you know, do you understand some of the cause for concern or some of the skepticism around it? Yes, I think it's very natural. That is very

new technology. It's very different. So I think if you look back in history, it's very common that when a new thing comes around, there's skepticism. There are lots of questions. There is excitement, but there's also a lot of sensitivity. But I wonder, I mean, you know, it's one thing to say, hey, you know, don't worry, this is going to be okay. But how can you guarantee or convince people that the technology couldn't be misused, couldn't be, you know, manipulated in a way that endangered? human life.

Well, I guess, you know, we should start, I would start with the status quo, right? We are not okay, right? If you look at just how many lives are lost to the transportation system that we have today. I'm sure you have heard these numbers before, but, you know, well known that in the U.S. alone, more than 40,000 people die every year. I just need to take a step back. This is, it's kind of insane. Yeah.

If we were to invent cars today or transportation system, like no way we would allow this. And I think over time, we just kind of slowly boiled ourselves, society, to, you know, accept that. And I think as a first order impact of this technology, you know, we can do better. We can do much better. And we are, you know, we're seeing that today. We have driven well north of 5 million fully autonomous miles today. And we've shared some data from the safety impact that our cars have. And I think at this point, we have a fairly robust body of evidence that shows that our cars actually have very clear safety benefits where they operate.

How do we think about, and there's no easy answer right now because it's both an ethical and a legal question. but how do we think about liability, right? I mean, if, let's say, I own a car that is fully autonomous with Waymo's technology, let's say in 10 years from now, and I'm in the back and I'm just doing my work, or I'm asleep, you know, which I should rather be doing. I just go to sleep and let it drive me from San Francisco to L.A., which would be great, but it gets in an accident of just a very fluke accident, maybe another car hits it or something. How do we account for liability? I mean, who's responsible? Is the owner of the car? Is Waymo's technology?

Like, how is that going to work? For the actions of, you know, the Waymo driver? Yeah. The responsibility, you know, lies, you know, with Waymo. If it was, you know, the fault of, you know, another actor, another driver, then you kind of follow the established, you know, processes that are, you know, well understood and well studied by, for example, insurance companies. They have decades of experience of kind of evaluating exactly that question.

And this is where that study that Swisserie has done was very encouraging that when they looked at almost 4 million miles of our fully autonomous operation, they found that massive reduction in 100% reduction in the bodily injury claims and 4x reduction. and property damage, right? So that then you can apply, like you can marry the two and you're starting to see the benefits. So if Waymo is, I mean, basically, if in a future scenario, somebody's in a Waymo car, the Waymo vehicle crashes and it's, there's some kind of fluke and the Waymo is responsible, I mean, you guys have to accept that liability.

But I guess in order to get to that position, you have to be rock, solid, confident that that will never happen. Well, that's what we spend, that's one of the hardest questions that we spend, you know, more than a decade working on. We have, we've developed a very robust, multifaceted, readiness and safety framework. And actually, that's something we shared publicly. And, you know, it's that what we see in all of those methodologies as we improve and validate our system, that the end of day is what gives us confidence in the performance of the system. How many cars do you have on the streets of San Francisco now? We have a fleet in San Francisco of about 250 vehicles.

You know, they're not all out at the same time, but it gives you an approximate order of magnitude. And Phoenix? About the same. I want to say about a couple hundred cars as well. So here's the question. I mean, is the part of the business model right now for Waymo to become like a ride hailing service? Like, you know, I'm sure, I know you've got a partnership that you announce with Uber, but I mean, is the, the idea that, you know, in 10 years time, this is going to be the primary, your primary

business or a part of what you do? Give me a sense of, is Waymo going to be a ride-hailing service or is it going to be a technology company that licenses its technology to both railing services and automobile manufacturers and others? Well, we think of ourselves as building a generalizable Waymo driver. And the business model is to deploy it in different product lines, different commercial applications. There's three main ones. Right healing. That's what we call Waymo one. And then there's, trucking and deliveries, moving goods. And then the third one is personally owned vehicles. So that's the long-term vision. We want the driver. There's trillions of miles being traveled, I think there's

almost three trillion in the U.S. alone, you know, much more across the world. So we want to have, you know, positive impact on, you know, some meaningful fraction of all of those miles. So, you know, the first business and the first product is Waymo 1 and, you know, right hailing. But we're exploring, you know, different, you know, other different partnerships. I just mentioned that the Uber partnership that we just launched and where it is. And that partnership, by the way, is, what is that going to look like? It's going to be the Uber app will also hail Waymo cars? That's right. That's right. You can use the Uber app and get a fully autonomous vehicle. This is not a money-making operation for Waymo right now.

I mean, the ride-hailing service is still in its infancy, but there's been billions of dollars invested into Waymo. And I have to imagine ride-haling is not where you're going to make your money. It's going to be from selling this technology. So tell me, just from the business perspective, when do you see a path to profitability? Well, you know, a challenge a little bit that, you know, right healing is a massive opportunity. Yeah.

It is, you know, a very big market today. But, you know, it's growing. There's expectations that is going to be significantly bigger by, you know, the end of the decade. But if you, on top of that, if you factor in, you know, the benefits and the, you know, positive economics that fully autonomous vehicles can bring to the table, there's potential for that to, you know, expand quite a bit. So, you know, we were very, you know, laser focused on that as our primary business line. You know, beyond that, you know, we want to pursue trucking and deliveries and then eventually personally owned vehicles.

But right healing, like, I would not dismiss that at all. So probably unrealistic to say that within 10 years, ordinary people could buy a fully autonomous vehicle for themselves, but probably not unrealistic to say that in 10 years from now, in most major urban centers in the U.S., there will be autonomous taxis available for anybody to use. I definitely agree with the latter, and I would not dismiss the former 10 years. This is a reasonably long time, and things can happen nonlinearly. So, a final question for you. So, I mean, if the future is autonomous, and I think it is, I think I really do am convinced.

I think anybody who uses one of these taxis will see it. It's so clear, at least to me, you know, you go in it and it, it's a clean car. It's a very good driver. It's a defensive driver, but it's also not overly defensive, so it's not timid. It's like a very good taxi driver, better. So that, I think, is the future. So, Dimitri, I know that, you know, there are plenty of people who love driving, right? love the experience of controlling their car.

And those people will continue to want to have that ability, and they will. But in your view, are we looking at a future where most people are going to be driven by their cars? In the long-term future, I think, yes, I think that that's where, you know, we're heading. Maybe, you know, taking your car in the future to a racetrack and driving it manually, that's going to be the and of the novel and an exciting thing, rather than the mundane, kind of boring task of driving, commuting. And if you imagine a future where, you know, a large fraction of your cars in the road are, you know, fully autonomous or at least smart enough, then you can start doing things where you're optimizing, you know, more globally. Like, they can coordinate, you know, their speed. You can, you know,

connect them to smarter infrastructure and actually overall increase the throughput of your roads and could you know kind of increase the throughput of your transportation system and you know farther out in the future if you look uh there today personally owned vehicles um are can sit around for 90 of their you know the lifetime right you know you take them you know home or work and then you park it and then you know nine out of 10 hours is just sitting there so if that changes right if you You no longer have to have your car just sitting there for you. It just opens up. There's a lot more space that can be used in cities for other more interesting things.

Dimitri, thank you so much. That was a pleasure. Thank you, Guy. That's Dimitri Dolgav, co-CEO of Waymo. Hey, thanks so much for listening to the show this week. Please make sure to click the follow button on your podcast app so you never miss a new episode of the show. And as always, it's free.

This episode was produced by Carrie Thompson with editing by John Isabella. research help from Chris Messini. Our music was composed by Ramtin Arableu. Our audio engineer was Neil Rauch. Our production team at How I Built This includes Alex Chung, Carla Estevez, Casey Herman, Chris Messini, Elaine Coates, J.C. Howard, Malia Agadello, Neva Grant, Sam Paulson, and Catherine Seifer. I'm Guy Raz, and you've been listening to How I Built This Lab.