WEBVTT - From Desert Robots to Driverless Trucks 0:00:15.356 --> 0:00:22.036 Pushkin. In two thousand and three, Chris Armsen was a 0:00:22.076 --> 0:00:25.836 grad student at Carnegie Mellon University and he was working 0:00:25.836 --> 0:00:29.476 on a research project in Chile in the Ata Kama Desert. 0:00:29.836 --> 0:00:33.196 Beautiful place, very desolate, the closest thing we have to 0:00:33.236 --> 0:00:36.156 Mars in many ways on Earth, which is why you 0:00:36.196 --> 0:00:38.996 were there. We were building a robot to go look 0:00:39.036 --> 0:00:41.556 for signs of life, and I was developing the navigation 0:00:41.636 --> 0:00:45.076 system for it, and my PhD advisor came down and said, Hey, 0:00:45.116 --> 0:00:47.436 the Defense Departments has this competition. They want to build 0:00:47.436 --> 0:00:50.076 a robot to drive fifty miles an hour across the 0:00:50.116 --> 0:00:52.596 desert between Los Angeles Las Vegas. And I thought that 0:00:52.676 --> 0:00:56.956 just sounded awesome. So we bought a Humvy and we 0:00:57.036 --> 0:00:59.196 cut the top of it, put a big electronics box 0:00:59.236 --> 0:01:01.756 in a lot of computers. All of the stuff that's 0:01:01.796 --> 0:01:04.756 in a self driving car today. The high performs computing, 0:01:05.356 --> 0:01:09.396 the lasers, the radars, the cameras, the hd map apps 0:01:09.436 --> 0:01:11.716 were part of this. It's kind of the first time 0:01:11.716 --> 0:01:15.676 when all this stuff came together into a vehicle, and 0:01:15.716 --> 0:01:18.876 then we developed it for six or nine months, and 0:01:18.876 --> 0:01:21.596 then we tried to compete with it. The first darper challenge. 0:01:22.516 --> 0:01:25.716 The team I was a technical director for we actually 0:01:25.716 --> 0:01:28.876 had the highest performing thing there. It was supposed to 0:01:28.916 --> 0:01:30.876 go one hundred and twenty six miles across the desert. 0:01:31.036 --> 0:01:35.236 It ended up going about seven when it got hung 0:01:35.316 --> 0:01:38.196 up on a berm and almost literally burst into flames. 0:01:38.876 --> 0:01:41.596 And you, I mean, nobody won. But if anybody won, 0:01:41.676 --> 0:01:43.956 you won, right, that was more than nobody. That was 0:01:43.996 --> 0:01:46.156 farther than Yeah. The key thing is nobody won. 0:01:46.396 --> 0:01:48.796 If you imagine a marathon where you're supposed to go 0:01:48.836 --> 0:01:50.996 twenty six miles and the best runner went too right, 0:01:51.036 --> 0:01:52.516 it's yes. 0:01:52.436 --> 0:01:55.956 Yes, fair, and then burst into flames and then burst 0:01:55.996 --> 0:01:56.516 into flames. 0:01:56.556 --> 0:02:02.996 Yeah. 0:02:03.036 --> 0:02:05.276 I'm Jacob Goldstein, and this is What's your problem? The 0:02:05.316 --> 0:02:07.076 show where I talk to people who are trying to 0:02:07.116 --> 0:02:11.556 make technological progress. My guest today is Chris Ermson. He's 0:02:11.596 --> 0:02:14.636 the co founder and CEO of Aurora, a company that's 0:02:14.636 --> 0:02:18.396 trying to bring autonomous driving to commercial trucks. As you 0:02:18.476 --> 0:02:21.636 just heard, Chris started working on autonomous vehicles more than 0:02:21.716 --> 0:02:25.756 twenty years ago, and in fact, his career traces the 0:02:25.956 --> 0:02:29.596 really the entire modern arc of self driving cars, from 0:02:29.596 --> 0:02:33.196 that first DARPA challenge to a top job at Google's 0:02:33.196 --> 0:02:36.276 self driving car project, the project that later became Waimo 0:02:36.756 --> 0:02:40.236 to founding Aurora, the company where he works now. We 0:02:40.276 --> 0:02:42.556 talked a lot about the problems Chris is trying to 0:02:42.596 --> 0:02:45.956 solve in autonomous trucking, and we also talked about the 0:02:45.996 --> 0:02:49.556 future of autonomous driving more generally. But before we got 0:02:49.596 --> 0:02:52.036 to that, we talked about the earlier work he did, 0:02:52.396 --> 0:02:55.196 work that really helped get the field, the broader field, 0:02:55.276 --> 0:02:57.756 to where it is today. So we'll pick up the 0:02:57.796 --> 0:03:01.316 interview back in two thousand and four at that moment 0:03:01.396 --> 0:03:05.516 when his team's autonomous car had failed and almost burst 0:03:05.516 --> 0:03:06.116 into flames. 0:03:06.476 --> 0:03:09.596 It was crushing, yeah, right, Like, you know, on the 0:03:09.636 --> 0:03:12.716 one hand, it was called a Grand challenge, right because 0:03:13.076 --> 0:03:14.396 you know, a lot of people didn't think it could 0:03:14.436 --> 0:03:17.356 be solved. It was impossible. Because on the one hand, 0:03:17.356 --> 0:03:20.516 it's kind of like, okay, it was really hard. On 0:03:20.556 --> 0:03:23.396 the other hand, you've you know, we're sleeping on the 0:03:23.436 --> 0:03:25.356 floor in the lab, in the you know, in this 0:03:25.396 --> 0:03:28.716 garage in the middle of the desert for weeks and 0:03:28.756 --> 0:03:30.356 months trying to get this thing to go, and then 0:03:30.356 --> 0:03:32.756 you launch it out there and you have this this 0:03:32.836 --> 0:03:35.756 epic moment where you're watching this thing that you've built, 0:03:35.836 --> 0:03:37.716 tear out across the desert and you can just see 0:03:37.716 --> 0:03:41.596 the top of it, and these helicopters that the military 0:03:41.676 --> 0:03:43.436 is using to track them as they're out in the 0:03:43.476 --> 0:03:47.196 desert doing this, and then you hear you know, it's 0:03:47.196 --> 0:03:48.916 supposed to one hundred and twenty miles, and they're like, oh, 0:03:48.916 --> 0:03:52.516 it's stuck. And then you hear like billowing smoke and 0:03:52.556 --> 0:03:55.756 broken and poor thing comes back looking very sad, and 0:03:56.236 --> 0:03:59.076 you know, it was crushing. But at the same time, 0:04:00.276 --> 0:04:02.756 we did set records. Right, if you look at the 0:04:02.836 --> 0:04:04.876 kind of the product of the speed and the distance 0:04:04.916 --> 0:04:07.036 it went, this was an order of magnitude better than 0:04:07.076 --> 0:04:10.276 anyone had done before, and so it was a big step. 0:04:10.356 --> 0:04:12.636 And I think that was partly what encouraged the Defense 0:04:12.636 --> 0:04:15.156 Department to say, hey, come back again and try it 0:04:15.156 --> 0:04:15.556 in a year. 0:04:17.196 --> 0:04:19.716 And then you did, and then you did it again 0:04:19.836 --> 0:04:21.476 and again. And what year do we go to? Is 0:04:21.516 --> 0:04:23.996 it two thousand and seven? Is that at the next 0:04:23.996 --> 0:04:24.596 big moment? 0:04:24.836 --> 0:04:26.796 Yeah? Two thousand and five was another one of these 0:04:26.796 --> 0:04:30.116 in the desert. Two thousand and seven they did what 0:04:30.116 --> 0:04:32.676 they called the Urban Challenge. So two thousand and five 0:04:32.676 --> 0:04:34.836 we're able to drive it around the desert not hit 0:04:34.876 --> 0:04:37.636 things too much. Two thousand and seven, they say, okay, 0:04:37.676 --> 0:04:39.556 that's great, but now let's stay on the right side 0:04:39.556 --> 0:04:41.756 of the road. Let's move in kind of pseudo urban 0:04:41.876 --> 0:04:46.516 environments and show that that can work. And that was 0:04:46.556 --> 0:04:48.956 the last of Grand challenges, because that year we actually 0:04:49.116 --> 0:04:51.676 accomplished it. We drove sixty miles around this air base 0:04:51.756 --> 0:04:54.276 and interacted with traffic. They hired a bunch of stunt 0:04:54.316 --> 0:04:56.236 drivers out of Hollywood to come and drive cars around 0:04:56.276 --> 0:05:00.356 these robots to make traffic for them, and it was 0:05:00.356 --> 0:05:03.316 a heck of a day. And then the Defense Department, 0:05:03.396 --> 0:05:06.756 or at least darpas said, you know, effectively, problem solved. 0:05:07.836 --> 0:05:09.676 We've got this close enough. Now hand it off to 0:05:09.756 --> 0:05:11.636 industry or whoever to take it from here. 0:05:12.516 --> 0:05:14.356 And how did how did your car do? 0:05:14.436 --> 0:05:17.116 In two thousand and seven, we won, which was nice. 0:05:17.156 --> 0:05:17.916 It was a good change. 0:05:18.436 --> 0:05:21.796 So you know, we went and did it catch on fire? 0:05:22.116 --> 0:05:24.996 No fire? This time, it just kind of worked. We 0:05:25.076 --> 0:05:29.436 had a little bit of nerves at the very beginning, 0:05:29.516 --> 0:05:32.636 because you know, our system was pretty reliable, robust, it 0:05:32.716 --> 0:05:35.476 worked all the time. We go and bring it out 0:05:35.516 --> 0:05:38.116 to the start line and we can't get it to 0:05:38.196 --> 0:05:42.236 boot properly because it can't see this GPS satellites that 0:05:42.236 --> 0:05:46.276 it needed, and we never had this problem before. We're like, 0:05:46.756 --> 0:05:48.756 you know, frantically trying to figure out what's going on. 0:05:48.796 --> 0:05:52.196 And then we look over and realized that for race 0:05:52.276 --> 0:05:56.716 day they brought a JumboTron and they've set that up 0:05:56.836 --> 0:05:59.196 right where the robots are at the starting place. And 0:05:59.276 --> 0:06:01.996 it turns out they don't really check these mobile jumbotrons 0:06:02.036 --> 0:06:05.556 for like how much interference they're generating, I guess. So 0:06:05.596 --> 0:06:07.356 we had this panic moment we said, you know, can 0:06:07.396 --> 0:06:09.556 you turn off the jumbo tron? And sure enough they 0:06:09.556 --> 0:06:13.756 turn off that and suddenly everything works again. So you know, 0:06:13.956 --> 0:06:18.156 it was it wasn't without its challenges, so so but it. 0:06:18.156 --> 0:06:19.916 Works right, So it works, and it works in a 0:06:19.996 --> 0:06:24.476 like quasi real environment, a pseudo real environment. So I 0:06:24.516 --> 0:06:26.676 mean when you at that moment, when you were thinking 0:06:26.716 --> 0:06:31.236 about the future, the future of self driving cars, how 0:06:31.836 --> 0:06:33.116 did it look to you? What did you think? 0:06:33.996 --> 0:06:37.476 Yeah, I think at that point was starting to believe 0:06:37.476 --> 0:06:39.796 there was a thing that could happen here, right, and 0:06:40.036 --> 0:06:44.316 probably you know, certainly naively thought oh yeah, we've got 0:06:44.316 --> 0:06:44.876 this right. 0:06:46.076 --> 0:06:49.716 This maybe everybody kind of thought that, right, Like. 0:06:50.036 --> 0:06:53.676 Yeah, I think, And that's been part of what's made 0:06:53.676 --> 0:06:56.396 the journey really fun for me. It's just there's just 0:06:56.476 --> 0:06:59.516 so many things to learn, so many new problems to 0:06:59.516 --> 0:07:03.036 overcome along the way of taking something from a it 0:07:03.196 --> 0:07:05.996 worked on this day in a race, to hey, we 0:07:06.036 --> 0:07:10.156 can go do something transformational that will save lives and 0:07:10.196 --> 0:07:13.156 improve the economy and you know, do all these other 0:07:13.196 --> 0:07:13.756 great things. 0:07:14.876 --> 0:07:17.996 It seems like that the distance between it worked in 0:07:18.036 --> 0:07:22.276 a race and we can do something transformational has been longer, 0:07:22.636 --> 0:07:24.676 certainly than a lot of people thought. Then a lot 0:07:24.676 --> 0:07:26.676 of like smart, well informed people. 0:07:26.436 --> 0:07:32.276 Thought, yeah, me among them, right. And you know, it's 0:07:32.276 --> 0:07:34.116 one of these things where I think anyone who sets 0:07:34.156 --> 0:07:38.516 out and does something interest and exciting, often I think 0:07:38.556 --> 0:07:40.196 if they understood how hard it was going to be 0:07:40.196 --> 0:07:43.516 before they started, may not have actually taken the first step. 0:07:43.556 --> 0:07:47.156 And so I feel a little lucky to have not 0:07:47.236 --> 0:07:49.636 understood that, because it's caused me to kind of stick 0:07:49.676 --> 0:07:51.956 with it and really enjoy the journey along the way. 0:07:52.436 --> 0:07:55.316 I mean, entrepreneurs basically always say that thing that you 0:07:55.476 --> 0:07:58.156 just said, but in the case of autonomous driving, I 0:07:58.196 --> 0:08:01.756 really believe it. Right, It seems like extra true. Like 0:08:02.076 --> 0:08:04.876 two thousand and seven was basically like the year the 0:08:04.996 --> 0:08:07.636 iPhone came out, right, like, which is to say, in 0:08:08.076 --> 0:08:11.116 technology long time ago, not to mention that there was 0:08:11.196 --> 0:08:14.636 like no computer vision as we know it today, right, 0:08:14.676 --> 0:08:17.556 which has become important, I mean so much. 0:08:18.196 --> 0:08:23.396 No, it is. You know, the waves of technological evolution 0:08:23.436 --> 0:08:26.356 that we've seen since then are profound, and we've been 0:08:26.396 --> 0:08:28.436 able to kind of surf those as we've been building it. 0:08:28.476 --> 0:08:35.716 And it's been Yeah, it is, it's been a long time. 0:08:39.156 --> 0:08:41.036 So then you go to Google, right, you go to 0:08:41.036 --> 0:08:43.196 Google a couple couple of years after that two thousand 0:08:43.236 --> 0:08:46.756 and nine. So you're at Google from twenty nine to 0:08:46.836 --> 0:08:50.276 twenty sixteen, Right, what do you and the team figure out? 0:08:50.356 --> 0:08:52.356 And then yeah, when do you leave? 0:08:52.916 --> 0:08:58.676 So we get there, and we originally want to show 0:08:58.756 --> 0:09:00.156 that we could go from what we've shown at the 0:09:00.236 --> 0:09:04.196 Darker Challenge to something actually worked on public roads, and 0:09:04.276 --> 0:09:07.476 so we had this challenge of driving one hundred thousand 0:09:07.516 --> 0:09:10.836 miles in public roads and miles of really interesting, challenging 0:09:10.876 --> 0:09:13.276 roads in the Bay Area, things like driving down Highway 0:09:13.316 --> 0:09:16.596 wand from San Francisco to La and across all the 0:09:16.636 --> 0:09:19.396 Bay bridges and things like that, with the idea that 0:09:19.436 --> 0:09:22.356 if we could prove we could do these things, then yeah, 0:09:22.396 --> 0:09:25.716 this technology probably has legs. And it was kind of 0:09:25.756 --> 0:09:29.596 a dread pirate Roberts setup right in that, you know, 0:09:29.916 --> 0:09:33.116 we were told that you've got you know, two years 0:09:33.156 --> 0:09:35.836 to do this, and you know then we'll probably fire 0:09:35.876 --> 0:09:38.716 you the day after, right, and you kind of this 0:09:38.876 --> 0:09:41.156 was kind of with us for the time. And so 0:09:41.276 --> 0:09:44.076 we worked with urgency, and it turns out we did 0:09:44.116 --> 0:09:46.836 in eighteen months, and through that we had some really 0:09:46.876 --> 0:09:51.236 interesting technological evolution. You know, we started to see some 0:09:51.276 --> 0:09:55.036 of the you know, early legs of deep learning started 0:09:55.036 --> 0:09:56.196 to show up in that point. 0:09:56.996 --> 0:10:00.556 During that time, so this is like two thousand and ten, 0:10:00.676 --> 0:10:02.396 twenty eleven, yeah. 0:10:02.156 --> 0:10:03.596 Exactly those times, yep. 0:10:03.836 --> 0:10:04.036 Yeah. 0:10:04.956 --> 0:10:10.156 And then but I think almost more importantly really started 0:10:10.236 --> 0:10:12.636 we really started to understand what this would mean to people. 0:10:13.396 --> 0:10:16.156 Right that we you know, as somebody who had worked 0:10:16.156 --> 0:10:19.516 on this because the technology was cool and initially and 0:10:19.676 --> 0:10:22.436 enjoyed the challenge of it, and seeing the benefit to 0:10:22.476 --> 0:10:24.956 the military as we thought about you know, as we 0:10:24.996 --> 0:10:26.556 started to get into the space and started to talk 0:10:26.556 --> 0:10:30.316 to people in the public sector about the number of 0:10:30.316 --> 0:10:33.316 lives lost on America's roads, you know, forty plus thousand 0:10:33.316 --> 0:10:40.476 people every year. The challenge that people, whether because they 0:10:40.716 --> 0:10:44.276 have physical limitation to prevent them from driving or because 0:10:44.316 --> 0:10:46.956 they've aged out no longer feel comfortable driving, the implications 0:10:46.956 --> 0:10:50.396 that has for the quality of life, and how we 0:10:50.476 --> 0:10:54.076 can make that better for folks, you know, the accessibility 0:10:54.116 --> 0:10:57.956 that we could bring to transportation that's kind of lost 0:10:57.996 --> 0:10:59.876 with the public trans system we have in the US. 0:11:00.236 --> 0:11:02.596 All of this was like, oh wow, there is not 0:11:02.716 --> 0:11:05.876 just it's not just a cool, interesting technology. It's not 0:11:05.916 --> 0:11:07.716 that I get to work with good people. It's not 0:11:07.796 --> 0:11:10.436 just that you know, this could be valuable. It's that, like, 0:11:10.476 --> 0:11:14.116 this is socially really important to do as well. And then, 0:11:14.436 --> 0:11:18.196 you know, the course of the program, we got to 0:11:18.236 --> 0:11:20.396 the point where we launched for the first time a 0:11:20.476 --> 0:11:24.276 vehicle operating on public roads with nobody in it, well, 0:11:24.316 --> 0:11:27.236 actually with somebody at it, with Steve Mahon, who is 0:11:27.276 --> 0:11:29.316 a blind gentleman who'd worked with us on some of 0:11:29.316 --> 0:11:31.796 the early kind of concepts and understanding of this, and 0:11:31.836 --> 0:11:34.876 he took a road to Austin, Texas by himself and 0:11:35.236 --> 0:11:41.436 seeing the emotional impact that had on him of you know, 0:11:41.596 --> 0:11:44.796 once again having the freedom to get from me to 0:11:44.836 --> 0:11:47.036 be without having asked somebody else. It was just like, wow, 0:11:47.116 --> 0:11:51.236 this this actually matters. So that was very cool. 0:11:52.676 --> 0:11:56.436 So so you mentioned the machine learning piece starting to 0:11:56.476 --> 0:11:59.116 come in. It's I think I was looking back at 0:11:59.196 --> 0:12:02.196 I think it's twenty twelve that alex net comes along, right, 0:12:02.236 --> 0:12:03.956 the first thing. I think that's right, kind of big 0:12:04.196 --> 0:12:08.396 computer vision breakthrough. And so it is interesting. I mean 0:12:08.436 --> 0:12:10.956 when I I think as a sort of lay person 0:12:11.156 --> 0:12:13.876 about autonomous cars, now, I think of computer vision as 0:12:13.916 --> 0:12:17.236 like central like, how could you even do it without 0:12:17.276 --> 0:12:20.676 that right, without computers being able to quote unquote understand 0:12:20.676 --> 0:12:23.196 what they see. But this is actually emerging just as 0:12:23.196 --> 0:12:26.956 you're building this car, right, this system, So how like 0:12:27.716 --> 0:12:29.676 tell me about that? Tell me sort of building in 0:12:29.756 --> 0:12:33.476 parallel with this technological enabling wave. 0:12:33.996 --> 0:12:36.716 Yeah, there was a bunch of these things. So Moore's 0:12:36.796 --> 0:12:39.276 law was continued in a march along and the amount 0:12:39.276 --> 0:12:42.756 of computation we had in the vehicles improving the cell 0:12:42.756 --> 0:12:46.156 phone was, believe it or not, was a big boon 0:12:46.236 --> 0:12:50.036 to us because it drove the performance of cameras. 0:12:50.596 --> 0:12:54.436 So, oh, interesting thing about like all those selfies everybody 0:12:54.476 --> 0:12:57.636 just wanting to make better cheaper cameras and the processing 0:12:57.716 --> 0:12:59.036 for images, yeah. 0:12:58.876 --> 0:13:01.956 One hundred, but even just the physical imager itself, right, 0:13:02.036 --> 0:13:05.316 went from you know, a one megapixel three twenty two 0:13:05.396 --> 0:13:07.516 forty camera, which is what if you bought a top 0:13:07.596 --> 0:13:11.236 the line you know, Kodak digital camera back in the day, 0:13:11.636 --> 0:13:17.236 was now my iPhone's got forty eight megapixel, right, and so, 0:13:17.516 --> 0:13:22.956 and the quality and color performance and everything that's improved dramatically. 0:13:23.956 --> 0:13:27.276 And then light ar is another thing that has really 0:13:27.836 --> 0:13:31.196 evolved over the course of the last twenty years, going 0:13:31.196 --> 0:13:34.076 from things where we could get a single point along 0:13:34.116 --> 0:13:38.956 the line to now full three D fields of view 0:13:39.276 --> 0:13:43.436 and allows us to combine the camera data with the 0:13:43.476 --> 0:13:46.916 light our data with the computation. And then the final 0:13:46.996 --> 0:13:49.676 big thing that's moved forward is in the sensing side 0:13:49.756 --> 0:13:54.236 is radar and the automotive radar. The quality that you 0:13:54.236 --> 0:13:56.396 can get out of that, the price point that comes 0:13:56.396 --> 0:13:59.516 in at now, you know, when you combine that set 0:13:59.556 --> 0:14:03.276 of advances, there's orders of magnitude improvement in both the 0:14:03.396 --> 0:14:06.156 quality of data we can consume and the amount of 0:14:06.236 --> 0:14:09.756 data that we can perform computation over that really is 0:14:09.876 --> 0:14:12.596 unlocked our ability to understand the world around these vehicles. 0:14:14.156 --> 0:14:21.156 So you leave in twenty sixteen, like, are people getting 0:14:21.236 --> 0:14:21.996 rides in wherever? 0:14:21.996 --> 0:14:22.276 It was? 0:14:22.396 --> 0:14:25.196 Chandler, Arizona yet at that time? Like, is that is 0:14:25.196 --> 0:14:26.516 it out in the world yet when you leave? 0:14:26.556 --> 0:14:29.956 Okay, No, we've been doing demos and we were doing development. 0:14:29.996 --> 0:14:32.036 I think we had vehicles in Chandler at that time, 0:14:32.116 --> 0:14:35.796 and we're starting to operate, but they were still with 0:14:35.916 --> 0:14:38.076 a person behind the steering wheel, and we were still 0:14:38.436 --> 0:14:43.236 we were building towards what ultimately became the initial channel launch. 0:14:44.436 --> 0:14:45.796 Why'd you leave? 0:14:46.116 --> 0:14:49.436 It was time? You know, I had been there. I 0:14:49.476 --> 0:14:54.076 really valued and left the experience, but at some point 0:14:54.156 --> 0:14:56.396 I kind of lost confidence that we were going to 0:14:56.396 --> 0:14:59.116 make the decisions that we needed to make to be successful. 0:14:59.156 --> 0:15:03.756 And as a person leading the organization that you know, 0:15:03.996 --> 0:15:05.956 Google had been incredibly generous to me, give me this 0:15:06.036 --> 0:15:09.156 incredible opportunity, and if I just didn't believe it, it 0:15:09.236 --> 0:15:11.636 wasn't my place to lead the team. And and so 0:15:12.356 --> 0:15:15.156 you know, I'm a big believer that in business, you 0:15:15.156 --> 0:15:17.716 you know, kind of you have three options. You if 0:15:17.716 --> 0:15:19.276 you see something you don't like, you try and fix it. 0:15:19.516 --> 0:15:21.916 If you can't fix it, you get in line. And 0:15:22.156 --> 0:15:23.436 if you can't get in line, you get out of 0:15:23.476 --> 0:15:26.436 the way, and you know, I did what I could 0:15:26.476 --> 0:15:28.396 to try and you know, kind of move things away 0:15:28.396 --> 0:15:30.796 that I thought they needed to go, and ultimately said, 0:15:30.876 --> 0:15:33.756 you know, this company's been incredibly galous to me. Let 0:15:33.756 --> 0:15:36.196 me get out of the way and let it gets 0:15:36.276 --> 0:15:37.116 on its own path. 0:15:38.236 --> 0:15:41.996 When you say you lost confidence that they were going 0:15:42.036 --> 0:15:45.476 to make the decisions they needed to make, I mean, 0:15:46.196 --> 0:15:49.636 obviously we should evaluate decisions based on the information available 0:15:49.636 --> 0:15:52.796 at the time, but it is the case that you know, 0:15:53.356 --> 0:15:55.236 you can get in Awai Mo in San Francisco in 0:15:55.316 --> 0:15:57.796 one minute and have a beautiful ride right now, Like, 0:15:59.196 --> 0:15:59.836 were you wrong? 0:16:00.476 --> 0:16:05.476 I apparently yes, right, Like you know, like you think, 0:16:05.476 --> 0:16:07.436 there's no way to judge it other than they they've 0:16:07.516 --> 0:16:12.596 ultimately been successful. You know, maybe would have taken a different, 0:16:12.596 --> 0:16:16.076 shorter path, but you know, I think I think it's 0:16:16.116 --> 0:16:18.396 it's yeah. I don't know if you're a parent, but 0:16:18.396 --> 0:16:19.836 one of the things you hope as a parent is 0:16:19.876 --> 0:16:25.316 that you you you embody your kids with things and 0:16:25.316 --> 0:16:27.836 then they go off from the world and succeed. And 0:16:27.876 --> 0:16:30.196 the same is true when you lead teams or build something, 0:16:30.316 --> 0:16:31.956 is that you know you do your part and you 0:16:31.996 --> 0:16:34.876 hope it outlives you and that you know, outperforms what 0:16:34.916 --> 0:16:37.716 you might have hoped for it. And I think it's 0:16:37.716 --> 0:16:39.556 been awesome to see way most succeed. 0:16:40.236 --> 0:16:46.396 So you started in twenty seventeen, right, uh, And as 0:16:46.436 --> 0:16:49.116 I understand it, you didn't start focused on trucks, Like, 0:16:49.156 --> 0:16:51.276 tell me about what you're thinking when you start the company? 0:16:51.716 --> 0:16:54.316 Yeah, how you get to trucks? So you know, I 0:16:54.356 --> 0:16:57.316 didn't leave Google to start the company. I left Goog 0:16:57.316 --> 0:16:58.796 because it was time, and I then spent some time 0:16:58.796 --> 0:17:00.436 trying to figure out what to do and kind of 0:17:00.436 --> 0:17:03.196 came to the conclusion that, you know, it was worth 0:17:03.316 --> 0:17:05.196 taking another shot at this and trying to build something. 0:17:06.436 --> 0:17:08.996 And so found two great co founders in Sterling and True, 0:17:09.316 --> 0:17:12.156 and we wanted to build a driver and we wanted 0:17:12.156 --> 0:17:15.316 to apply that driver and passenger cars and trucks and 0:17:15.316 --> 0:17:17.956 wherever it could go. And we thought at the time, 0:17:18.396 --> 0:17:20.236 we didn't see how it's going to be possible to 0:17:20.316 --> 0:17:23.716 do trucking because based on the experience I had and 0:17:23.756 --> 0:17:25.316 others had, we looked at how far you had to 0:17:25.316 --> 0:17:27.556 see down the road to do that safely, and we 0:17:27.596 --> 0:17:29.996 came to the conclusion you just couldn't do that with 0:17:30.036 --> 0:17:32.436 the technology that exists all the time, and you have. 0:17:32.436 --> 0:17:34.436 To see farther just because it takes longer for a 0:17:34.436 --> 0:17:35.356 big truck to stop. 0:17:35.396 --> 0:17:38.836 This is physics. There's just way more kinetic energy. Right, 0:17:40.276 --> 0:17:42.396 They're heavier, and they're moving at seventy miles an hour 0:17:42.436 --> 0:17:44.316 instead of in a city you're moving at fifteen miles 0:17:44.316 --> 0:17:46.836 an hour. So the combination of the speed and the 0:17:46.876 --> 0:17:49.516 way it means that it just takes longer to stop 0:17:49.596 --> 0:17:54.276 farther distance down the road. And then you add to that, 0:17:54.636 --> 0:17:56.596 it's much easier drive around a light vehicle than a 0:17:56.596 --> 0:18:00.236 big truck, so you don't need a special license. They're 0:18:00.356 --> 0:18:04.436 much smaller, you don't need special thing maintenance programs and whatnot. 0:18:04.476 --> 0:18:07.476 And so we're like, okay, let's focus on light vehicles. 0:18:07.676 --> 0:18:10.596 We see there's a real opportunity there. If we can 0:18:10.756 --> 0:18:16.036 crack the can you see far enough problem, then trucking 0:18:16.076 --> 0:18:18.716 to be a great application to go to. Yeah. Yeah uh. 0:18:18.876 --> 0:18:20.516 And so part of what I spent my time doing 0:18:20.596 --> 0:18:22.876 during the first couple of years of the company was looking, 0:18:22.916 --> 0:18:25.316 you know, basically turning over every rock we could with 0:18:25.316 --> 0:18:28.236 with Bart, one of our early team members, to find 0:18:28.236 --> 0:18:30.596 a technology that would allow us to see for far 0:18:30.676 --> 0:18:35.196 enough that we could actually do trucking. And ultimately we 0:18:35.236 --> 0:18:38.876 found that in this little company in Montana called Blackmore 0:18:39.796 --> 0:18:42.356 that had built this really special kind of laser range 0:18:42.356 --> 0:18:45.796 finder that because of the way it did measurements and 0:18:46.036 --> 0:18:53.356 what's called frequency modulated continuous wave it it could see 0:18:53.356 --> 0:18:55.596 basically twice as far as the rest of the light 0:18:55.636 --> 0:18:59.116 our stuff could see. Uh, And so we're like, huh, 0:18:59.236 --> 0:19:02.916 we now we we have some really interesting driving capability. 0:19:02.916 --> 0:19:06.796 We've now got this pseudomagical light our sensor that can 0:19:06.836 --> 0:19:10.396 see far enough trucking for like a really great application 0:19:10.476 --> 0:19:11.116 to go take on. 0:19:12.156 --> 0:19:13.436 Let's talk about a few of the things you had 0:19:13.436 --> 0:19:15.516 to figure out. So you decide you're going to do trucking. Yeah, 0:19:15.556 --> 0:19:19.116 and presumably by trucking, Like I mean, I know where 0:19:19.156 --> 0:19:22.556 you are now, Like, so presumably by trucking. You don't 0:19:22.596 --> 0:19:26.276 mean like the bread truck that goes to the little 0:19:26.396 --> 0:19:28.516 market by my house. You mean the great, big truck 0:19:28.556 --> 0:19:29.676 that goes on the freeway. 0:19:30.076 --> 0:19:33.316 Yeah, that's where we thought there was the biggest opportunity 0:19:33.636 --> 0:19:36.876 was working what we call class eight tractor trailers, big 0:19:36.996 --> 0:19:40.196 you know, semi trucks. You know, they drive a lot 0:19:40.196 --> 0:19:43.156 of miles. We don't have enough people who want to 0:19:43.196 --> 0:19:47.796 do that job, and yet it's absolutely essential to American 0:19:47.956 --> 0:19:52.356 and worldwide way of life, and there's a real need. 0:19:53.156 --> 0:19:55.436 Right as we've dug into it, we realized that there's 0:19:55.756 --> 0:19:58.316 five thousand people killed every year in heavy truck accidents, 0:19:58.556 --> 0:20:01.316 five hundred thousand people injured. And then we can move 0:20:01.356 --> 0:20:05.436 goods more efficiently, which you know is important, particularly in 0:20:05.676 --> 0:20:07.956 the you know, this age of e commerce where people 0:20:07.956 --> 0:20:11.676 expect things next day or you know, be able to 0:20:11.996 --> 0:20:15.356 move those goods efficiently the customers is really valuable. 0:20:17.556 --> 0:20:21.556 Okay, this April, if I've got my dates right, you 0:20:21.676 --> 0:20:26.196 actually have a truck drive for real stuff on the 0:20:26.236 --> 0:20:28.676 freeway without someone behind the wheel, right. 0:20:28.876 --> 0:20:31.876 That's exactly right. It was awesome, right. We've as a 0:20:31.916 --> 0:20:34.756 company been building towards that for almost eight and a 0:20:34.756 --> 0:20:38.196 half years, eight and fourty years, and you know, we 0:20:38.356 --> 0:20:42.396 take doing these things safely, very seriously, and really a 0:20:42.436 --> 0:20:44.596 lot of the last couple of years has been making 0:20:44.596 --> 0:20:48.036 sure we could have confidence that when we let the 0:20:48.036 --> 0:20:50.996 thing go, that it was actually going to be safe. 0:20:51.036 --> 0:20:52.476 And then we got to a point where we're like, yeah, 0:20:52.556 --> 0:20:54.556 this is safe. We can go send this out in 0:20:54.596 --> 0:20:57.956 the daylight and dry weather and have it drive back 0:20:57.956 --> 0:21:00.556 and forth between Dallas and Houston. And I had the 0:21:00.596 --> 0:21:05.876