Latest / What's Your Problem? / From Desert Robots to Driverless Trucks

Transcript

From Desert Robots to Driverless Trucks · What's Your Problem?

• 0:15
  Speaker 1: Pushkin. In two thousand and three, Chris Armsen was a
• 0:22
  Speaker 1: grad student at Carnegie Mellon University and he was working
• 0:25
  Speaker 1: on a research project in Chile in the Ata Kama Desert.
• 0:29
  Speaker 2: Beautiful place, very desolate, the closest thing we have to
• 0:33
  Speaker 2: Mars in many ways on Earth, which is why you
• 0:36
  Speaker 2: were there. We were building a robot to go look
• 0:39
  Speaker 2: for signs of life, and I was developing the navigation
• 0:41
  Speaker 2: system for it, and my PhD advisor came down and said, Hey,
• 0:45
  Speaker 2: the Defense Departments has this competition. They want to build
• 0:47
  Speaker 2: a robot to drive fifty miles an hour across the
• 0:50
  Speaker 2: desert between Los Angeles Las Vegas. And I thought that
• 0:52
  Speaker 2: just sounded awesome. So we bought a Humvy and we
• 0:57
  Speaker 2: cut the top of it, put a big electronics box
• 0:59
  Speaker 2: in a lot of computers. All of the stuff that's
• 1:01
  Speaker 2: in a self driving car today. The high performs computing,
• 1:05
  Speaker 2: the lasers, the radars, the cameras, the hd map apps
• 1:09
  Speaker 2: were part of this. It's kind of the first time
• 1:11
  Speaker 2: when all this stuff came together into a vehicle, and
• 1:15
  Speaker 2: then we developed it for six or nine months, and
• 1:18
  Speaker 2: then we tried to compete with it. The first darper challenge.
• 1:22
  Speaker 2: The team I was a technical director for we actually
• 1:25
  Speaker 2: had the highest performing thing there. It was supposed to
• 1:28
  Speaker 2: go one hundred and twenty six miles across the desert.
• 1:31
  Speaker 2: It ended up going about seven when it got hung
• 1:35
  Speaker 2: up on a berm and almost literally burst into flames.
• 1:38
  Speaker 1: And you, I mean, nobody won. But if anybody won,
• 1:41
  Speaker 1: you won, right, that was more than nobody. That was
• 1:43
  Speaker 1: farther than Yeah. The key thing is nobody won.
• 1:46
  Speaker 2: If you imagine a marathon where you're supposed to go
• 1:48
  Speaker 2: twenty six miles and the best runner went too right,
• 1:51
  Speaker 2: it's yes.
• 1:52
  Speaker 1: Yes, fair, and then burst into flames and then burst
• 1:55
  Speaker 1: into flames.
• 1:56
  Speaker 2: Yeah.
• 2:03
  Speaker 1: I'm Jacob Goldstein, and this is What's your problem? The
• 2:05
  Speaker 1: show where I talk to people who are trying to
• 2:07
  Speaker 1: make technological progress. My guest today is Chris Ermson. He's
• 2:11
  Speaker 1: the co founder and CEO of Aurora, a company that's
• 2:14
  Speaker 1: trying to bring autonomous driving to commercial trucks. As you
• 2:18
  Speaker 1: just heard, Chris started working on autonomous vehicles more than
• 2:21
  Speaker 1: twenty years ago, and in fact, his career traces the
• 2:25
  Speaker 1: really the entire modern arc of self driving cars, from
• 2:29
  Speaker 1: that first DARPA challenge to a top job at Google's
• 2:33
  Speaker 1: self driving car project, the project that later became Waimo
• 2:36
  Speaker 1: to founding Aurora, the company where he works now. We
• 2:40
  Speaker 1: talked a lot about the problems Chris is trying to
• 2:42
  Speaker 1: solve in autonomous trucking, and we also talked about the
• 2:45
  Speaker 1: future of autonomous driving more generally. But before we got
• 2:49
  Speaker 1: to that, we talked about the earlier work he did,
• 2:52
  Speaker 1: work that really helped get the field, the broader field,
• 2:55
  Speaker 1: to where it is today. So we'll pick up the
• 2:57
  Speaker 1: interview back in two thousand and four at that moment
• 3:01
  Speaker 1: when his team's autonomous car had failed and almost burst
• 3:05
  Speaker 1: into flames.
• 3:06
  Speaker 2: It was crushing, yeah, right, Like, you know, on the
• 3:09
  Speaker 2: one hand, it was called a Grand challenge, right because
• 3:13
  Speaker 2: you know, a lot of people didn't think it could
• 3:14
  Speaker 2: be solved. It was impossible. Because on the one hand,
• 3:17
  Speaker 2: it's kind of like, okay, it was really hard. On
• 3:20
  Speaker 2: the other hand, you've you know, we're sleeping on the
• 3:23
  Speaker 2: floor in the lab, in the you know, in this
• 3:25
  Speaker 2: garage in the middle of the desert for weeks and
• 3:28
  Speaker 2: months trying to get this thing to go, and then
• 3:30
  Speaker 2: you launch it out there and you have this this
• 3:32
  Speaker 2: epic moment where you're watching this thing that you've built,
• 3:35
  Speaker 2: tear out across the desert and you can just see
• 3:37
  Speaker 2: the top of it, and these helicopters that the military
• 3:41
  Speaker 2: is using to track them as they're out in the
• 3:43
  Speaker 2: desert doing this, and then you hear you know, it's
• 3:47
  Speaker 2: supposed to one hundred and twenty miles, and they're like, oh,
• 3:48
  Speaker 2: it's stuck. And then you hear like billowing smoke and
• 3:52
  Speaker 2: broken and poor thing comes back looking very sad, and
• 3:56
  Speaker 2: you know, it was crushing. But at the same time,
• 4:00
  Speaker 2: we did set records. Right, if you look at the
• 4:02
  Speaker 2: kind of the product of the speed and the distance
• 4:04
  Speaker 2: it went, this was an order of magnitude better than
• 4:07
  Speaker 2: anyone had done before, and so it was a big step.
• 4:10
  Speaker 2: And I think that was partly what encouraged the Defense
• 4:12
  Speaker 2: Department to say, hey, come back again and try it
• 4:15
  Speaker 2: in a year.
• 4:17
  Speaker 1: And then you did, and then you did it again
• 4:19
  Speaker 1: and again. And what year do we go to? Is
• 4:21
  Speaker 1: it two thousand and seven? Is that at the next
• 4:23
  Speaker 1: big moment?
• 4:24
  Speaker 2: Yeah? Two thousand and five was another one of these
• 4:26
  Speaker 2: in the desert. Two thousand and seven they did what
• 4:30
  Speaker 2: they called the Urban Challenge. So two thousand and five
• 4:32
  Speaker 2: we're able to drive it around the desert not hit
• 4:34
  Speaker 2: things too much. Two thousand and seven, they say, okay,
• 4:37
  Speaker 2: that's great, but now let's stay on the right side
• 4:39
  Speaker 2: of the road. Let's move in kind of pseudo urban
• 4:41
  Speaker 2: environments and show that that can work. And that was
• 4:46
  Speaker 2: the last of Grand challenges, because that year we actually
• 4:49
  Speaker 2: accomplished it. We drove sixty miles around this air base
• 4:51
  Speaker 2: and interacted with traffic. They hired a bunch of stunt
• 4:54
  Speaker 2: drivers out of Hollywood to come and drive cars around
• 4:56
  Speaker 2: these robots to make traffic for them, and it was
• 5:00
  Speaker 2: a heck of a day. And then the Defense Department,
• 5:03
  Speaker 2: or at least darpas said, you know, effectively, problem solved.
• 5:07
  Speaker 2: We've got this close enough. Now hand it off to
• 5:09
  Speaker 2: industry or whoever to take it from here.
• 5:12
  Speaker 1: And how did how did your car do?
• 5:14
  Speaker 2: In two thousand and seven, we won, which was nice.
• 5:17
  Speaker 2: It was a good change.
• 5:18
  Speaker 1: So you know, we went and did it catch on fire?
• 5:22
  Speaker 2: No fire? This time, it just kind of worked. We
• 5:25
  Speaker 2: had a little bit of nerves at the very beginning,
• 5:29
  Speaker 2: because you know, our system was pretty reliable, robust, it
• 5:32
  Speaker 2: worked all the time. We go and bring it out
• 5:35
  Speaker 2: to the start line and we can't get it to
• 5:38
  Speaker 2: boot properly because it can't see this GPS satellites that
• 5:42
  Speaker 2: it needed, and we never had this problem before. We're like,
• 5:46
  Speaker 2: you know, frantically trying to figure out what's going on.
• 5:48
  Speaker 2: And then we look over and realized that for race
• 5:52
  Speaker 2: day they brought a JumboTron and they've set that up
• 5:56
  Speaker 2: right where the robots are at the starting place. And
• 5:59
  Speaker 2: it turns out they don't really check these mobile jumbotrons
• 6:02
  Speaker 2: for like how much interference they're generating, I guess. So
• 6:05
  Speaker 2: we had this panic moment we said, you know, can
• 6:07
  Speaker 2: you turn off the jumbo tron? And sure enough they
• 6:09
  Speaker 2: turn off that and suddenly everything works again. So you know,
• 6:13
  Speaker 2: it was it wasn't without its challenges, so so but it.
• 6:18
  Speaker 1: Works right, So it works, and it works in a
• 6:19
  Speaker 1: like quasi real environment, a pseudo real environment. So I
• 6:24
  Speaker 1: mean when you at that moment, when you were thinking
• 6:26
  Speaker 1: about the future, the future of self driving cars, how
• 6:31
  Speaker 1: did it look to you? What did you think?
• 6:33
  Speaker 2: Yeah, I think at that point was starting to believe
• 6:37
  Speaker 2: there was a thing that could happen here, right, and
• 6:40
  Speaker 2: probably you know, certainly naively thought oh yeah, we've got
• 6:44
  Speaker 2: this right.
• 6:46
  Speaker 1: This maybe everybody kind of thought that, right, Like.
• 6:50
  Speaker 2: Yeah, I think, And that's been part of what's made
• 6:53
  Speaker 2: the journey really fun for me. It's just there's just
• 6:56
  Speaker 2: so many things to learn, so many new problems to
• 6:59
  Speaker 2: overcome along the way of taking something from a it
• 7:03
  Speaker 2: worked on this day in a race, to hey, we
• 7:06
  Speaker 2: can go do something transformational that will save lives and
• 7:10
  Speaker 2: improve the economy and you know, do all these other
• 7:13
  Speaker 2: great things.
• 7:14
  Speaker 1: It seems like that the distance between it worked in
• 7:18
  Speaker 1: a race and we can do something transformational has been longer,
• 7:22
  Speaker 1: certainly than a lot of people thought. Then a lot
• 7:24
  Speaker 1: of like smart, well informed people.
• 7:26
  Speaker 2: Thought, yeah, me among them, right. And you know, it's
• 7:32
  Speaker 2: one of these things where I think anyone who sets
• 7:34
  Speaker 2: out and does something interest and exciting, often I think
• 7:38
  Speaker 2: if they understood how hard it was going to be
• 7:40
  Speaker 2: before they started, may not have actually taken the first step.
• 7:43
  Speaker 2: And so I feel a little lucky to have not
• 7:47
  Speaker 2: understood that, because it's caused me to kind of stick
• 7:49
  Speaker 2: with it and really enjoy the journey along the way.
• 7:52
  Speaker 1: I mean, entrepreneurs basically always say that thing that you
• 7:55
  Speaker 1: just said, but in the case of autonomous driving, I
• 7:58
  Speaker 1: really believe it. Right, It seems like extra true. Like
• 8:02
  Speaker 1: two thousand and seven was basically like the year the
• 8:04
  Speaker 1: iPhone came out, right, like, which is to say, in
• 8:08
  Speaker 1: technology long time ago, not to mention that there was
• 8:11
  Speaker 1: like no computer vision as we know it today, right,
• 8:14
  Speaker 1: which has become important, I mean so much.
• 8:18
  Speaker 2: No, it is. You know, the waves of technological evolution
• 8:23
  Speaker 2: that we've seen since then are profound, and we've been
• 8:26
  Speaker 2: able to kind of surf those as we've been building it.
• 8:28
  Speaker 2: And it's been Yeah, it is, it's been a long time.
• 8:39
  Speaker 1: So then you go to Google, right, you go to
• 8:41
  Speaker 1: Google a couple couple of years after that two thousand
• 8:43
  Speaker 1: and nine. So you're at Google from twenty nine to
• 8:46
  Speaker 1: twenty sixteen, Right, what do you and the team figure out?
• 8:50
  Speaker 1: And then yeah, when do you leave?
• 8:52
  Speaker 2: So we get there, and we originally want to show
• 8:58
  Speaker 2: that we could go from what we've shown at the
• 9:00
  Speaker 2: Darker Challenge to something actually worked on public roads, and
• 9:04
  Speaker 2: so we had this challenge of driving one hundred thousand
• 9:07
  Speaker 2: miles in public roads and miles of really interesting, challenging
• 9:10
  Speaker 2: roads in the Bay Area, things like driving down Highway
• 9:13
  Speaker 2: wand from San Francisco to La and across all the
• 9:16
  Speaker 2: Bay bridges and things like that, with the idea that
• 9:19
  Speaker 2: if we could prove we could do these things, then yeah,
• 9:22
  Speaker 2: this technology probably has legs. And it was kind of
• 9:25
  Speaker 2: a dread pirate Roberts setup right in that, you know,
• 9:29
  Speaker 2: we were told that you've got you know, two years
• 9:33
  Speaker 2: to do this, and you know then we'll probably fire
• 9:35
  Speaker 2: you the day after, right, and you kind of this
• 9:38
  Speaker 2: was kind of with us for the time. And so
• 9:41
  Speaker 2: we worked with urgency, and it turns out we did
• 9:44
  Speaker 2: in eighteen months, and through that we had some really
• 9:46
  Speaker 2: interesting technological evolution. You know, we started to see some
• 9:51
  Speaker 2: of the you know, early legs of deep learning started
• 9:55
  Speaker 2: to show up in that point.
• 9:56
  Speaker 1: During that time, so this is like two thousand and ten,
• 10:00
  Speaker 1: twenty eleven, yeah.
• 10:02
  Speaker 2: Exactly those times, yep.
• 10:03
  Speaker 1: Yeah.
• 10:04
  Speaker 2: And then but I think almost more importantly really started
• 10:10
  Speaker 2: we really started to understand what this would mean to people.
• 10:13
  Speaker 2: Right that we you know, as somebody who had worked
• 10:16
  Speaker 2: on this because the technology was cool and initially and
• 10:19
  Speaker 2: enjoyed the challenge of it, and seeing the benefit to
• 10:22
  Speaker 2: the military as we thought about you know, as we
• 10:24
  Speaker 2: started to get into the space and started to talk
• 10:26
  Speaker 2: to people in the public sector about the number of
• 10:30
  Speaker 2: lives lost on America's roads, you know, forty plus thousand
• 10:33
  Speaker 2: people every year. The challenge that people, whether because they
• 10:40
  Speaker 2: have physical limitation to prevent them from driving or because
• 10:44
  Speaker 2: they've aged out no longer feel comfortable driving, the implications
• 10:46
  Speaker 2: that has for the quality of life, and how we
• 10:50
  Speaker 2: can make that better for folks, you know, the accessibility
• 10:54
  Speaker 2: that we could bring to transportation that's kind of lost
• 10:57
  Speaker 2: with the public trans system we have in the US.
• 11:00
  Speaker 2: All of this was like, oh wow, there is not
• 11:02
  Speaker 2: just it's not just a cool, interesting technology. It's not
• 11:05
  Speaker 2: that I get to work with good people. It's not
• 11:07
  Speaker 2: just that you know, this could be valuable. It's that, like,
• 11:10
  Speaker 2: this is socially really important to do as well. And then,
• 11:14
  Speaker 2: you know, the course of the program, we got to
• 11:18
  Speaker 2: the point where we launched for the first time a
• 11:20
  Speaker 2: vehicle operating on public roads with nobody in it, well,
• 11:24
  Speaker 2: actually with somebody at it, with Steve Mahon, who is
• 11:27
  Speaker 2: a blind gentleman who'd worked with us on some of
• 11:29
  Speaker 2: the early kind of concepts and understanding of this, and
• 11:31
  Speaker 2: he took a road to Austin, Texas by himself and
• 11:35
  Speaker 2: seeing the emotional impact that had on him of you know,
• 11:41
  Speaker 2: once again having the freedom to get from me to
• 11:44
  Speaker 2: be without having asked somebody else. It was just like, wow,
• 11:47
  Speaker 2: this this actually matters. So that was very cool.
• 11:52
  Speaker 1: So so you mentioned the machine learning piece starting to
• 11:56
  Speaker 1: come in. It's I think I was looking back at
• 11:59
  Speaker 1: I think it's twenty twelve that alex net comes along, right,
• 12:02
  Speaker 1: the first thing. I think that's right, kind of big
• 12:04
  Speaker 1: computer vision breakthrough. And so it is interesting. I mean
• 12:08
  Speaker 1: when I I think as a sort of lay person
• 12:11
  Speaker 1: about autonomous cars, now, I think of computer vision as
• 12:13
  Speaker 1: like central like, how could you even do it without
• 12:17
  Speaker 1: that right, without computers being able to quote unquote understand
• 12:20
  Speaker 1: what they see. But this is actually emerging just as
• 12:23
  Speaker 1: you're building this car, right, this system, So how like
• 12:27
  Speaker 1: tell me about that? Tell me sort of building in
• 12:29
  Speaker 1: parallel with this technological enabling wave.
• 12:33
  Speaker 2: Yeah, there was a bunch of these things. So Moore's
• 12:36
  Speaker 2: law was continued in a march along and the amount
• 12:39
  Speaker 2: of computation we had in the vehicles improving the cell
• 12:42
  Speaker 2: phone was, believe it or not, was a big boon
• 12:46
  Speaker 2: to us because it drove the performance of cameras.
• 12:50
  Speaker 1: So, oh, interesting thing about like all those selfies everybody
• 12:54
  Speaker 1: just wanting to make better cheaper cameras and the processing
• 12:57
  Speaker 1: for images, yeah.
• 12:58
  Speaker 2: One hundred, but even just the physical imager itself, right,
• 13:02
  Speaker 2: went from you know, a one megapixel three twenty two
• 13:05
  Speaker 2: forty camera, which is what if you bought a top
• 13:07
  Speaker 2: the line you know, Kodak digital camera back in the day,
• 13:11
  Speaker 2: was now my iPhone's got forty eight megapixel, right, and so,
• 13:17
  Speaker 2: and the quality and color performance and everything that's improved dramatically.
• 13:23
  Speaker 2: And then light ar is another thing that has really
• 13:27
  Speaker 2: evolved over the course of the last twenty years, going
• 13:31
  Speaker 2: from things where we could get a single point along
• 13:34
  Speaker 2: the line to now full three D fields of view
• 13:39
  Speaker 2: and allows us to combine the camera data with the
• 13:43
  Speaker 2: light our data with the computation. And then the final
• 13:46
  Speaker 2: big thing that's moved forward is in the sensing side
• 13:49
  Speaker 2: is radar and the automotive radar. The quality that you
• 13:54
  Speaker 2: can get out of that, the price point that comes
• 13:56
  Speaker 2: in at now, you know, when you combine that set
• 13:59
  Speaker 2: of advances, there's orders of magnitude improvement in both the
• 14:03
  Speaker 2: quality of data we can consume and the amount of
• 14:06
  Speaker 2: data that we can perform computation over that really is
• 14:09
  Speaker 2: unlocked our ability to understand the world around these vehicles.
• 14:14
  Speaker 1: So you leave in twenty sixteen, like, are people getting
• 14:21
  Speaker 1: rides in wherever?
• 14:21
  Speaker 2: It was?
• 14:22
  Speaker 1: Chandler, Arizona yet at that time? Like, is that is
• 14:25
  Speaker 1: it out in the world yet when you leave?
• 14:26
  Speaker 2: Okay, No, we've been doing demos and we were doing development.
• 14:29
  Speaker 2: I think we had vehicles in Chandler at that time,
• 14:32
  Speaker 2: and we're starting to operate, but they were still with
• 14:35
  Speaker 2: a person behind the steering wheel, and we were still
• 14:38
  Speaker 2: we were building towards what ultimately became the initial channel launch.
• 14:44
  Speaker 1: Why'd you leave?
• 14:46
  Speaker 2: It was time? You know, I had been there. I
• 14:49
  Speaker 2: really valued and left the experience, but at some point
• 14:54
  Speaker 2: I kind of lost confidence that we were going to
• 14:56
  Speaker 2: make the decisions that we needed to make to be successful.
• 14:59
  Speaker 2: And as a person leading the organization that you know,
• 15:03
  Speaker 2: Google had been incredibly generous to me, give me this
• 15:06
  Speaker 2: incredible opportunity, and if I just didn't believe it, it
• 15:09
  Speaker 2: wasn't my place to lead the team. And and so
• 15:12
  Speaker 2: you know, I'm a big believer that in business, you
• 15:15
  Speaker 2: you know, kind of you have three options. You if
• 15:17
  Speaker 2: you see something you don't like, you try and fix it.
• 15:19
  Speaker 2: If you can't fix it, you get in line. And
• 15:22
  Speaker 2: if you can't get in line, you get out of
• 15:23
  Speaker 2: the way, and you know, I did what I could
• 15:26
  Speaker 2: to try and you know, kind of move things away
• 15:28
  Speaker 2: that I thought they needed to go, and ultimately said,
• 15:30
  Speaker 2: you know, this company's been incredibly galous to me. Let
• 15:33
  Speaker 2: me get out of the way and let it gets
• 15:36
  Speaker 2: on its own path.
• 15:38
  Speaker 1: When you say you lost confidence that they were going
• 15:42
  Speaker 1: to make the decisions they needed to make, I mean,
• 15:46
  Speaker 1: obviously we should evaluate decisions based on the information available
• 15:49
  Speaker 1: at the time, but it is the case that you know,
• 15:53
  Speaker 1: you can get in Awai Mo in San Francisco in
• 15:55
  Speaker 1: one minute and have a beautiful ride right now, Like,
• 15:59
  Speaker 1: were you wrong?
• 16:00
  Speaker 2: I apparently yes, right, Like you know, like you think,
• 16:05
  Speaker 2: there's no way to judge it other than they they've
• 16:07
  Speaker 2: ultimately been successful. You know, maybe would have taken a different,
• 16:12
  Speaker 2: shorter path, but you know, I think I think it's
• 16:16
  Speaker 2: it's yeah. I don't know if you're a parent, but
• 16:18
  Speaker 2: one of the things you hope as a parent is
• 16:19
  Speaker 2: that you you you embody your kids with things and
• 16:25
  Speaker 2: then they go off from the world and succeed. And
• 16:27
  Speaker 2: the same is true when you lead teams or build something,
• 16:30
  Speaker 2: is that you know you do your part and you
• 16:31
  Speaker 2: hope it outlives you and that you know, outperforms what
• 16:34
  Speaker 2: you might have hoped for it. And I think it's
• 16:37
  Speaker 2: been awesome to see way most succeed.
• 16:40
  Speaker 1: So you started in twenty seventeen, right, uh, And as
• 16:46
  Speaker 1: I understand it, you didn't start focused on trucks, Like,
• 16:49
  Speaker 1: tell me about what you're thinking when you start the company?
• 16:51
  Speaker 2: Yeah, how you get to trucks? So you know, I
• 16:54
  Speaker 2: didn't leave Google to start the company. I left Goog
• 16:57
  Speaker 2: because it was time, and I then spent some time
• 16:58
  Speaker 2: trying to figure out what to do and kind of
• 17:00
  Speaker 2: came to the conclusion that, you know, it was worth
• 17:03
  Speaker 2: taking another shot at this and trying to build something.
• 17:06
  Speaker 2: And so found two great co founders in Sterling and True,
• 17:09
  Speaker 2: and we wanted to build a driver and we wanted
• 17:12
  Speaker 2: to apply that driver and passenger cars and trucks and
• 17:15
  Speaker 2: wherever it could go. And we thought at the time,
• 17:18
  Speaker 2: we didn't see how it's going to be possible to
• 17:20
  Speaker 2: do trucking because based on the experience I had and
• 17:23
  Speaker 2: others had, we looked at how far you had to
• 17:25
  Speaker 2: see down the road to do that safely, and we
• 17:27
  Speaker 2: came to the conclusion you just couldn't do that with
• 17:30
  Speaker 2: the technology that exists all the time, and you have.
• 17:32
  Speaker 1: To see farther just because it takes longer for a
• 17:34
  Speaker 1: big truck to stop.
• 17:35
  Speaker 2: This is physics. There's just way more kinetic energy. Right,
• 17:40
  Speaker 2: They're heavier, and they're moving at seventy miles an hour
• 17:42
  Speaker 2: instead of in a city you're moving at fifteen miles
• 17:44
  Speaker 2: an hour. So the combination of the speed and the
• 17:46
  Speaker 2: way it means that it just takes longer to stop
• 17:49
  Speaker 2: farther distance down the road. And then you add to that,
• 17:54
  Speaker 2: it's much easier drive around a light vehicle than a
• 17:56
  Speaker 2: big truck, so you don't need a special license. They're
• 18:00
  Speaker 2: much smaller, you don't need special thing maintenance programs and whatnot.
• 18:04
  Speaker 2: And so we're like, okay, let's focus on light vehicles.
• 18:07
  Speaker 2: We see there's a real opportunity there. If we can
• 18:10
  Speaker 2: crack the can you see far enough problem, then trucking
• 18:16
  Speaker 2: to be a great application to go to. Yeah. Yeah uh.
• 18:18
  Speaker 2: And so part of what I spent my time doing
• 18:20
  Speaker 2: during the first couple of years of the company was looking,
• 18:22
  Speaker 2: you know, basically turning over every rock we could with
• 18:25
  Speaker 2: with Bart, one of our early team members, to find
• 18:28
  Speaker 2: a technology that would allow us to see for far
• 18:30
  Speaker 2: enough that we could actually do trucking. And ultimately we
• 18:35
  Speaker 2: found that in this little company in Montana called Blackmore
• 18:39
  Speaker 2: that had built this really special kind of laser range
• 18:42
  Speaker 2: finder that because of the way it did measurements and
• 18:46
  Speaker 2: what's called frequency modulated continuous wave it it could see
• 18:53
  Speaker 2: basically twice as far as the rest of the light
• 18:55
  Speaker 2: our stuff could see. Uh, And so we're like, huh,
• 18:59
  Speaker 2: we now we we have some really interesting driving capability.
• 19:02
  Speaker 2: We've now got this pseudomagical light our sensor that can
• 19:06
  Speaker 2: see far enough trucking for like a really great application
• 19:10
  Speaker 2: to go take on.
• 19:12
  Speaker 1: Let's talk about a few of the things you had
• 19:13
  Speaker 1: to figure out. So you decide you're going to do trucking. Yeah,
• 19:15
  Speaker 1: and presumably by trucking, Like I mean, I know where
• 19:19
  Speaker 1: you are now, Like, so presumably by trucking. You don't
• 19:22
  Speaker 1: mean like the bread truck that goes to the little
• 19:26
  Speaker 1: market by my house. You mean the great, big truck
• 19:28
  Speaker 1: that goes on the freeway.
• 19:30
  Speaker 2: Yeah, that's where we thought there was the biggest opportunity
• 19:33
  Speaker 2: was working what we call class eight tractor trailers, big
• 19:36
  Speaker 2: you know, semi trucks. You know, they drive a lot
• 19:40
  Speaker 2: of miles. We don't have enough people who want to
• 19:43
  Speaker 2: do that job, and yet it's absolutely essential to American
• 19:47
  Speaker 2: and worldwide way of life, and there's a real need.
• 19:53
  Speaker 2: Right as we've dug into it, we realized that there's
• 19:55
  Speaker 2: five thousand people killed every year in heavy truck accidents,
• 19:58
  Speaker 2: five hundred thousand people injured. And then we can move
• 20:01
  Speaker 2: goods more efficiently, which you know is important, particularly in
• 20:05
  Speaker 2: the you know, this age of e commerce where people
• 20:07
  Speaker 2: expect things next day or you know, be able to
• 20:11
  Speaker 2: move those goods efficiently the customers is really valuable.
• 20:17
  Speaker 1: Okay, this April, if I've got my dates right, you
• 20:21
  Speaker 1: actually have a truck drive for real stuff on the
• 20:26
  Speaker 1: freeway without someone behind the wheel, right.
• 20:28
  Speaker 2: That's exactly right. It was awesome, right. We've as a
• 20:31
  Speaker 2: company been building towards that for almost eight and a
• 20:34
  Speaker 2: half years, eight and fourty years, and you know, we
• 20:38
  Speaker 2: take doing these things safely, very seriously, and really a
• 20:42
  Speaker 2: lot of the last couple of years has been making
• 20:44
  Speaker 2: sure we could have confidence that when we let the
• 20:48
  Speaker 2: thing go, that it was actually going to be safe.
• 20:51
  Speaker 2: And then we got to a point where we're like, yeah,
• 20:52
  Speaker 2: this is safe. We can go send this out in
• 20:54
  Speaker 2: the daylight and dry weather and have it drive back
• 20:57
  Speaker 2: and forth between Dallas and Houston. And I had the
• 21:00
  Speaker 2: privilege of right along in the back seat. And it
• 21:05
  Speaker 2: was awesome right in that it worked. Incredibly boring, right,
• 21:10
  Speaker 2: It turns out it's like it's.
• 21:11
  Speaker 1: Wanted to be so boring. You wanted to be very boring.
• 21:14
  Speaker 2: But but it just kind of worked. And I went
• 21:18
  Speaker 2: down and back that day in the truck and were.
• 21:21
  Speaker 1: You hauling stuff at that point? Was it was it?
• 21:23
  Speaker 2: We were? We were all the histories.
• 21:25
  Speaker 1: Yeah, oh pastries, Okay, pastries both ways, that would be
• 21:31
  Speaker 1: a bit. There's some pastries. It just that they want
• 21:34
  Speaker 1: to know.
• 21:34
  Speaker 2: I think there is a pastry imbalance. I think we
• 21:37
  Speaker 2: for the first trip, I think we may have actually
• 21:39
  Speaker 2: been empty. But on the way back, I'm pretty sure
• 21:41
  Speaker 2: from for whatever reasons, the pastries diffused from Houston to Dallas.
• 21:47
  Speaker 1: So then what happens in May and may you put
• 21:50
  Speaker 1: a person back behind the wheel, right? Tell me about that.
• 21:54
  Speaker 2: Yeah. We so as a company, we've really focused on
• 21:57
  Speaker 2: do what we do best, and what we do best
• 21:59
  Speaker 2: is building a safe, capable driving system. The oral driver
• 22:03
  Speaker 2: and we worked with other companies like Peterbilt at Pacar
• 22:06
  Speaker 2: and Volvo Trucks, and our partners at peter Bolt called
• 22:12
  Speaker 2: up and said, hey, we'd like you to have an
• 22:14
  Speaker 2: observer in the seat because there's some prototype parts in
• 22:18
  Speaker 2: our truck and it matters to us that you put
• 22:21
  Speaker 2: an observer in. And so we said, okay, uh, you know,
• 22:25
  Speaker 2: we had a conversation. We respect them, and we ultimately
• 22:28
  Speaker 2: put the observer in the seat, and we have a
• 22:30
  Speaker 2: person sat on board who is sat in the driver's seat,
• 22:33
  Speaker 2: but they're an observer and they're really just there because
• 22:36
  Speaker 2: our partner asked them to put them there.
• 22:38
  Speaker 1: The VIHA presumably they are a license to drive an
• 22:41
  Speaker 1: eighteen year It's not just some n week watching them.
• 22:43
  Speaker 1: What's going on. That's the truck driver.
• 22:45
  Speaker 2: You know. We we we don't want to put anyone
• 22:47
  Speaker 2: in place where you know, the high control has to
• 22:49
  Speaker 2: debate whether you know, it's just like, okay, let's just
• 22:51
  Speaker 2: put a truck driver there. They sit there. If you
• 22:54
  Speaker 2: go to YouTube dot com slash or row driver, you
• 22:58
  Speaker 2: can see a live video between eight and five o'clock
• 23:01
  Speaker 2: every day of our trucks drive down the road and
• 23:04
  Speaker 2: you can see that you know, this person sat there
• 23:07
  Speaker 2: and sometimes they're twiddling their sums, sometimes they're eating Fredo's,
• 23:11
  Speaker 2: you know, just observing.
• 23:14
  Speaker 1: I mean, at least from a narrative standpoint, it's a
• 23:19
  Speaker 1: bummer to have to put the person back behind the wheel.
• 23:21
  Speaker 2: Surely, absolutely, from a narrative standpoint, it's a bummer, right,
• 23:24
  Speaker 2: And it creates a nuance that is, you know, certainly
• 23:28
  Speaker 2: in today's.
• 23:29
  Speaker 1: You don't want a person buy the wheel. It's the
• 23:31
  Speaker 1: whole point. It's the whole point. It's you don't want
• 23:33
  Speaker 1: a person buy the wheel, right.
• 23:34
  Speaker 2: Totally agree. But in terms of the development and delivery
• 23:38
  Speaker 2: of what we're building, it just doesn't matter, right, it's
• 23:40
  Speaker 2: a complete nothing.
• 23:41
  Speaker 1: Well, I mean that's some margin. It matters at some point,
• 23:44
  Speaker 1: it matters, right. The whole premise is you can do
• 23:46
  Speaker 1: it without a person there, and so like at some point,
• 23:49
  Speaker 1: like relatively soon, I would imagine you need to do
• 23:53
  Speaker 1: it without a person there.
• 23:54
  Speaker 2: That's right, and uh, look forward to doing that soon.
• 24:00
  Speaker 3: We'll be back in just a minute.
• 24:14
  Speaker 1: Yeah, So where are you now, Like, give me where
• 24:16
  Speaker 1: we're basically to the present now after our twenty year journey,
• 24:21
  Speaker 1: Like what's like, are you are people paying you to
• 24:24
  Speaker 1: haul stuff.
• 24:25
  Speaker 2: Yeah, absolutely, So to like today is an incredibly exciting moment.
• 24:30
  Speaker 2: So we have a technology that can drive a truck.
• 24:32
  Speaker 2: It has the skills necessary to drive on the freeway
• 24:36
  Speaker 2: and drive to sites off the freeway in these industrial
• 24:40
  Speaker 2: park areas, and it does that safely. And then internally
• 24:44
  Speaker 2: we've got the capability say, okay, what are the new
• 24:47
  Speaker 2: features we need to make this more useful for our customers.
• 24:51
  Speaker 2: And so we launched in April. About three months later
• 24:55
  Speaker 2: we were able to show, you know, be able to
• 24:57
  Speaker 2: launch the next version that now doesn't just operate day
• 25:00
  Speaker 2: in the daytime, but operates at night as well. And
• 25:04
  Speaker 2: of the course the rest of this year, it's really
• 25:06
  Speaker 2: about taking and building the you know, in enhancing the
• 25:09
  Speaker 2: skills that it has to add the few new skills
• 25:11
  Speaker 2: it needs to be able to drive between Fort Worth
• 25:14
  Speaker 2: and Phoenix and El Paso. And so at that point, just.
• 25:17
  Speaker 1: To be clear, the route you have now is Houston Dallas,
• 25:20
  Speaker 1: Is that right?
• 25:20
  Speaker 2: That's right? I'm sorry, We drive tween thousand Houston and
• 25:23
  Speaker 2: by the end of the year we expect to be
• 25:24
  Speaker 2: driving between Fort Worth and Phoenix. And what's exciting about
• 25:29
  Speaker 2: and fort Worth and El Paso and El Paso and Phoenix.
• 25:33
  Speaker 2: And what's exciting about that is fort worth of Phoenix
• 25:37
  Speaker 2: is a thousand miles and a person is not able
• 25:40
  Speaker 2: to legally drive that in a day, And so we'll
• 25:45
  Speaker 2: be able to start doing things that are superhuman in
• 25:49
  Speaker 2: more and more dimensions.
• 25:50
  Speaker 1: Presumably the value preposition is really long drives, right, Like
• 25:55
  Speaker 1: it makes absolutely no sense for whatever one hundred miles
• 25:58
  Speaker 1: or something. And the longer it gets, the more the
• 26:00
  Speaker 1: economics makes sense.
• 26:01
  Speaker 2: Is that right? I mean, that's think it makes sense
• 26:04
  Speaker 2: across all different kinds of scales smiles. But I think
• 26:07
  Speaker 2: we're the biggest impact we'll ultimately is on these long
• 26:09
  Speaker 2: call trips.
• 26:11
  Speaker 1: And like, I get the freeway part in Broadway, How
• 26:17
  Speaker 1: does the not on the freeway part work? How constrained
• 26:20
  Speaker 1: are you as to where you can go when you're
• 26:22
  Speaker 1: not on the freeway.
• 26:23
  Speaker 2: Yeah, So the capability we're building is to be able
• 26:26
  Speaker 2: to drive everywhere. Sure, but like now, so today we
• 26:32
  Speaker 2: drive from a place that's off the freeway onto the freeway,
• 26:37
  Speaker 2: and then when we get in at the Dallas end
• 26:39
  Speaker 2: of the route, and then at the Houston end, we
• 26:41
  Speaker 2: drive about five miles from where we get off the
• 26:44
  Speaker 2: freeway to our terminal side, and along the way we
• 26:48
  Speaker 2: drive past customer locations. So we're making the choice today
• 26:51
  Speaker 2: to go to our terminals, but it's relatively straightforward to
• 26:57
  Speaker 2: stop and go to a different terminal instead or a
• 26:59
  Speaker 2: different See.
• 27:01
  Speaker 1: Like, how hard is that off the freeway part, the
• 27:05
  Speaker 1: sort of marginal you know, mile when not on the freeway?
• 27:11
  Speaker 2: Not, right, would be the short answer. Right, it's you know,
• 27:16
  Speaker 2: we've put a lot of for for a lot of time,
• 27:18
  Speaker 2: the primary focus what we're building was the freeway part, because,
• 27:22
  Speaker 2: as you pointed out, that's where the most kinetic energy is.
• 27:26
  Speaker 2: That's where you know, that's where the bad things could happen. Yeah.
• 27:30
  Speaker 1: I mean it's also simpler in certain ways, right, Like
• 27:34
  Speaker 1: it seems like turning a big truck around or just
• 27:37
  Speaker 1: getting a truck around. Like I live in New York
• 27:38
  Speaker 1: City and I see eighteen wheelers in New York City,
• 27:41
  Speaker 1: and I don't know, maybe that's maybe that's not that
• 27:44
  Speaker 1: hard to engineer for autonomy. You tell me it seems hard.
• 27:47
  Speaker 1: It looks hard.
• 27:48
  Speaker 2: Yeah, I think New York City is hard, right, And
• 27:51
  Speaker 2: I think there'll be a time where we get to
• 27:53
  Speaker 2: that where I'm really focused and we're really focused on
• 27:56
  Speaker 2: in the near term is how do we get to
• 27:58
  Speaker 2: the place where most freight moves right, And a lot
• 28:02
  Speaker 2: of that will go to these industrial parks in regions
• 28:06
  Speaker 2: of a city. And the reason why it's hard to
• 28:10
  Speaker 2: drive a truck in New York City right, and it's slow.
• 28:14
  Speaker 2: And so if you're trying to build a distribution center
• 28:16
  Speaker 2: for these Class eight tractors and trailers, you put them
• 28:18
  Speaker 2: in places where it's convenient because you're trying to run
• 28:21
  Speaker 2: a business, not build a technology showcase.
• 28:25
  Speaker 1: So what can you not do now ya?
• 28:30
  Speaker 2: So today we can't drive in the rain. Well, in fact,
• 28:34
  Speaker 2: that's not actually true. So today we drive in the
• 28:36
  Speaker 2: rain really well, but we haven't validated that fully for
• 28:42
  Speaker 2: driverless operations. And this is really an important idea, right,
• 28:45
  Speaker 2: is if you came down to Texas and got a
• 28:47
  Speaker 2: ride in one of our trucks and it was raining out,
• 28:50
  Speaker 2: it would work, and you would go, huh, why if
• 28:54
  Speaker 2: you guys not launched this, and we would say, yeah,
• 28:56
  Speaker 2: it basically works. But that's not good enough. It's not
• 29:00
  Speaker 2: good enough that it feels like it works. What we
• 29:02
  Speaker 2: want to have done is.
• 29:03
  Speaker 1: All basically it makes me nervous with a ten roil truck.
• 29:07
  Speaker 2: Exactly and it should right, And so we put a
• 29:10
  Speaker 2: lot of effort into making sure that it doesn't just
• 29:13
  Speaker 2: seem like it works, but we have conviction that it
• 29:15
  Speaker 2: will work right. And you know, as an example, two
• 29:21
  Speaker 2: of the last things that we we got checked off
• 29:24
  Speaker 2: when we launch, where it's one really high speed motorbikes
• 29:31
  Speaker 2: running red lights in the city. So we had done
• 29:35
  Speaker 2: a bunch of testing, I think up to let me say,
• 29:38
  Speaker 2: I'm gonna say some numbers here. These are probably not
• 29:39
  Speaker 2: quite right, but I'll use some numbers, you know, at
• 29:43
  Speaker 2: ninety miles an hour, because we figure, you know, ninety
• 29:45
  Speaker 2: miles an hour in a surface road, nobody.
• 29:48
  Speaker 1: Need the motorcycles going ninety.
• 29:50
  Speaker 2: Motorcycles going ninety miles an hour and they're going to
• 29:53
  Speaker 2: run a red light in front of us, and we
• 29:55
  Speaker 2: want the truck to not hit that motorbike. And then
• 29:57
  Speaker 2: we're on the freeway and we see a motorbike doing
• 29:59
  Speaker 2: one hundred and fifty miles an hour, and we're like, okay,
• 30:03
  Speaker 2: that is let's go make sure that that works right,
• 30:07
  Speaker 2: because you know, at some point we just want to
• 30:10
  Speaker 2: have comps. And so we went through and we found
• 30:12
  Speaker 2: out that when motorbike's going that extremely fast. First, we
• 30:18
  Speaker 2: found that our simulation wasn't quite up to snuff to
• 30:20
  Speaker 2: deal with that, so we had to go fix that.
• 30:23
  Speaker 2: And then we found that our perception system wasn't able
• 30:26
  Speaker 2: to quite track it as well as we'd want, so
• 30:28
  Speaker 2: we had some things to fix there, and so ultimately
• 30:31
  Speaker 2: we got to the point where if you're a motorbike,
• 30:33
  Speaker 2: please don't go test this run in a red light
• 30:36
  Speaker 2: at one hundred and fifty miles an hour. We're going
• 30:38
  Speaker 2: to see you and assuming you know, we can actually
• 30:41
  Speaker 2: see you, because there's not a building in the way,
• 30:43
  Speaker 2: we're going to stop and you know, not have an
• 30:46
  Speaker 2: event with you.
• 30:47
  Speaker 1: So is the rain analog of that, like more rain
• 30:50
  Speaker 1: than I can imagine and flooding and it's dark or something.
• 30:54
  Speaker 1: I mean, yeah, maybe frozen frozen also.
• 30:58
  Speaker 2: And so this is this is the answers. We we
• 31:01
  Speaker 2: have some good ideas of what these challenges are, and
• 31:04
  Speaker 2: so we're in the process of putting place the last
• 31:06
  Speaker 2: of the tests that we need to say yeah, across
• 31:08
  Speaker 2: the things that might be a problem, we actually handle
• 31:12
  Speaker 2: those in a way that's safe.
• 31:14
  Speaker 1: Once you got rain, are you good? I mean, I
• 31:16
  Speaker 1: guess you have the whole southern US without snow and ice? Right, Well,
• 31:20
  Speaker 1: think it not freezes in Texas? Right? Does it freeze in?
• 31:23
  Speaker 2: Does we get we get snow in Texas? And but
• 31:27
  Speaker 2: the good news is, like the reason why a lot
• 31:29
  Speaker 2: of freight moves wrong the sun belt is because trucks
• 31:33
  Speaker 2: don't like driving in snow. Right, It's just, you know,
• 31:36
  Speaker 2: much like people in general don't like driving in snow.
• 31:39
  Speaker 2: But there is snow. But if you've ever been in
• 31:40
  Speaker 2: Texas when there's a snowstorm, it basically shuts down.
• 31:43
  Speaker 1: Yeah, so you don't you don't have to solve snow.
• 31:46
  Speaker 1: You don't have to solve snow for now.
• 31:48
  Speaker 2: For now. But we will solve snow, right because all
• 31:50
  Speaker 2: of the stuff that works for rain mostly works for snow.
• 31:52
  Speaker 1: To me, all of these problems we can stipulate will
• 31:55
  Speaker 1: be solved by somebody, right, Yeah, plausibly by you, if
• 31:58
  Speaker 1: not you, by somebody else. But like I feel like
• 32:01
  Speaker 1: there's some sense of urgency. Like I'll say so interestingly,
• 32:04
  Speaker 1: just purely coincidentally, I talked to uh Worris softman?
• 32:10
  Speaker 2: Is it softman? Yeah?
• 32:10
  Speaker 1: Have his name is Med and so he said, and
• 32:13
  Speaker 1: we were not talking about you. This was not a
• 32:16
  Speaker 1: dig on you. It was more talking about why he's
• 32:18
  Speaker 1: doing what he's doing. He said, I don't think Open
• 32:21
  Speaker 1: Road Autonomy is a startup game. Uh. And he basically
• 32:25
  Speaker 1: you probably know he thinks, I mean, he basically thinks
• 32:28
  Speaker 1: it's too expensive.
• 32:29
  Speaker 2: Yeah.
• 32:29
  Speaker 1: In short, too expensive and too hard in that way
• 32:33
  Speaker 1: for a company that is capital constrained, for a company
• 32:35
  Speaker 1: that you know doesn't have a monopoly on search.
• 32:37
  Speaker 2: Let's see.
• 32:39
  Speaker 1: And so I do feel like in talking to you
• 32:40
  Speaker 1: like I, you plausibly may solve them. But there must
• 32:44
  Speaker 1: be some clock for you, right, You must have to
• 32:46
  Speaker 1: get to a point where you're making money relatively soon.
• 32:50
  Speaker 2: Is that right? Yeah, Well we've been at the game
• 32:52
  Speaker 2: for a while. You know, we went into an eyes
• 32:55
  Speaker 2: open right and looked at the you know, having spent
• 32:58
  Speaker 2: the time with Waymo, having the history I have, I
• 33:00
  Speaker 2: was like, this is not one hundred million dollar problem
• 33:02
  Speaker 2: or a ten million dollar problem. This is a multi
• 33:04
  Speaker 2: billion dollar problem. And so as we've built Aurora, it's
• 33:08
  Speaker 2: been Okay, we need to set ourselves up with the staff,
• 33:11
  Speaker 2: we need to set ourselves up with the capital partners,
• 33:13
  Speaker 2: we need to set ourselves up with the technology partners
• 33:18
  Speaker 2: that allow us to take that run. And I think
• 33:20
  Speaker 2: we've been very careful about how we've done that, and
• 33:23
  Speaker 2: we've put ourselves in a position to succeed. And I
• 33:25
  Speaker 2: agree with with Boris this is you know, I hear
• 33:28
  Speaker 2: people throwing out, oh, we're gonna, you know, spend one
• 33:30
  Speaker 2: hundred million dollars to do this. I think, you know,
• 33:32
  Speaker 2: you haven't even met the ante, right, And so for us,
• 33:39
  Speaker 2: we look at say, driving in the rain, we expect
• 33:41
  Speaker 2: to be solving that by the end of this year, right,
• 33:44
  Speaker 2: And at that point, as you point out, we've basically
• 33:47
  Speaker 2: unlocked the Southern freight belt. The amount of truck travel
• 33:51
  Speaker 2: that's there is gigantic, and so you know, from now
• 33:56
  Speaker 2: where we have a small number of trucks in the road,
• 33:59
  Speaker 2: by the end of next year, end of twenty six,
• 34:01
  Speaker 2: we expect to have a few hundred trucks on the road,
• 34:05
  Speaker 2: and then the year after that we expect to have
• 34:06
  Speaker 2: one thousand plus trucks on the road, and then we'll
• 34:09
  Speaker 2: built from there. And so it's you know, the conviction
• 34:14
  Speaker 2: we're starting to build internally, like yeah, this is just
• 34:16
  Speaker 2: just turn over the kind of the way we want it,
• 34:17
  Speaker 2: and we expect it is happening now and so it's
• 34:20
  Speaker 2: not a hypothetical. It some day we'll solve it. It's like, no,
• 34:22
  Speaker 2: we're you know, four or five months kind of time frame. So, well,
• 34:27
  Speaker 2: three four months timeframe.
• 34:28
  Speaker 1: What's the business model?
• 34:31
  Speaker 2: Yeah, so our business model is to work with the
• 34:34
  Speaker 2: ecosystem that's there today. So today, if you're a truck
• 34:37
  Speaker 2: and company, you buy a truck and then you pay
• 34:39
  Speaker 2: somebody to drive that truck for you. Uh, and so
• 34:42
  Speaker 2: that's what our company's going to do. You'll if you're
• 34:45
  Speaker 2: a company and wants some Automas truck, you're gonna go
• 34:48
  Speaker 2: to pac r and peter Bilt, or you're going to
• 34:50
  Speaker 2: go in Evolve and say I want to buy a truck,
• 34:52
  Speaker 2: and they buy the truck stuff and then it'll roll
• 34:57
  Speaker 2: off the line from our partner and it'll have the
• 34:59
  Speaker 2: row driver installed on it and you'll pay a subscription
• 35:02
  Speaker 2: to Aurora to drive that truck for you.
• 35:04
  Speaker 1: I see. Okay, So so you guys get paid by
• 35:08
  Speaker 1: the by the mile or by the by the hall
• 35:12
  Speaker 1: or something like that.
• 35:13
  Speaker 2: Yeah, that's exactly right. You know, it's at the risk
• 35:16
  Speaker 2: of at a service as a services you know, it's
• 35:19
  Speaker 2: driver as a service. That's the way we were thinking
• 35:21
  Speaker 2: about the business.
• 35:23
  Speaker 1: And the trucking company is buying the truck, So that's
• 35:25
  Speaker 1: good for you. You don't have to When you say
• 35:26
  Speaker 1: you'll have a thousand trucks out there, you don't mean
• 35:28
  Speaker 1: you're going to own a thousand trucks. You mean you'll
• 35:30
  Speaker 1: have essentially a thousand autonomous drivers in trucks that other
• 35:33
  Speaker 1: people own.
• 35:34
  Speaker 2: That's right, And this is because that's what our customers
• 35:36
  Speaker 2: want right that there Also it.
• 35:39
  Speaker 1: Seems easier as a business, and I still not have
• 35:42
  Speaker 1: to get into the trucking business.
• 35:44
  Speaker 2: I agreed again back to the philosophy of we think
• 35:47
  Speaker 2: we're really good at building the driver. And I look
• 35:50
  Speaker 2: at our partners, whether it's Werner or Hirshbock or Federal Express,
• 35:55
  Speaker 2: they know what they're doing. Why don't we help them
• 35:57
  Speaker 2: make their business a bit better and work together?
• 36:03
  Speaker 1: Can we talk about autonomy a little bit more broadly,
• 36:06
  Speaker 1: just since you're at it so long and I'm curious
• 36:08
  Speaker 1: what you think, So, like, what do you think are
• 36:12
  Speaker 1: the important constraints on autonomous vehicles right now? Sort of
• 36:17
  Speaker 1: broadly technologically, you know, in terms of policy.
• 36:21
  Speaker 2: I think it's primarily technological right now. Okay, right, And
• 36:25
  Speaker 2: I've said that. You know, people for for the better
• 36:28
  Speaker 2: part of a decade have been you know, when when
• 36:30
  Speaker 2: it's convenient hiding behind you know, regulation or policy. And
• 36:35
  Speaker 2: you know, the US policy of regulatory environment is permissive.
• 36:38
  Speaker 2: It's one of the things that's allowed innovation to flourish
• 36:40
  Speaker 2: in this country, certainly in the automotive space.
• 36:44
  Speaker 1: So what are the what are the important technological constraints
• 36:47
  Speaker 1: right now? Like, I mean, obviously WEIMO is thriving where
• 36:52
  Speaker 1: it's living, but like, also there are no autonomous cars
• 36:55
  Speaker 1: where I live and people aren't buying them, and so
• 36:58
  Speaker 1: like what's the what's what are the important bottlenecks right now?
• 37:01
  Speaker 2: Uh? Execution madwidth? Right? Like, So so at least where
• 37:06
  Speaker 2: I sit at a roar today, I don't out and see,
• 37:10
  Speaker 2: Oh my gosh, I don't know how we're going to
• 37:12
  Speaker 2: solve that. It's hey, we have this many people, we
• 37:16
  Speaker 2: have this much leadership band with we just need to
• 37:18
  Speaker 2: go execute right.
• 37:19
  Speaker 1: Like thinking industry wide and not to be like petty
• 37:24
  Speaker 1: or something, but like why aren't autonomous cars everywhere?
• 37:26
  Speaker 2: Right? Yeah? Because it's really hard, right, And I think
• 37:29
  Speaker 2: the right metaphor for this is commercial aviation, like the
• 37:34
  Speaker 2: physics are really in grasp and like you can both
• 37:36
  Speaker 2: things together. And yet you know, we basically have two
• 37:42
  Speaker 2: commercial aviation manufacturers globally, right, we have Boeing and Airbus.
• 37:47
  Speaker 1: Because it's hard to build big jet aircraft.
• 37:50
  Speaker 2: It's hard to build them, it's hard to have conviction
• 37:53
  Speaker 2: they're going to be safe, it's hard to produce them right.
• 37:56
  Speaker 2: And when I look at automated driving, I think it's
• 37:58
  Speaker 2: in a very similar space in that a lot of
• 38:01
  Speaker 2: things have to work well. You have to have the
• 38:03
  Speaker 2: discipline to be able to work across all of them,
• 38:06
  Speaker 2: make sure they all fit together, and then you have
• 38:08
  Speaker 2: to have the processes. And process in Silicon Valley is
• 38:12
  Speaker 2: often kind of a four letter word, like you know,
• 38:14
  Speaker 2: you've got to move past and break things, but when
• 38:16
  Speaker 2: you're building something safety critical, you actually have to get
• 38:19
  Speaker 2: that stuff right, and the art is getting it right,
• 38:22
  Speaker 2: we'll also be able to do it quickly.
• 38:25
  Speaker 1: So so when you think about the spread of autonomy
• 38:30
  Speaker 1: from here, again not just within your company, but sort
• 38:33
  Speaker 1: of more broadly, like how do you think it'll play out?
• 38:37
  Speaker 2: I think it's going to usher incredible new age all right,
• 38:40
  Speaker 2: that we should be able to dramatically reduce traffic fatalities.
• 38:48
  Speaker 2: The fact that you know, it had been a decade
• 38:51
  Speaker 2: without a fatality on civil aviation, and yet we had
• 38:55
  Speaker 2: forty thousand people killed on the road every year in
• 38:58
  Speaker 2: the US. Like, we can bring a technology to bear
• 39:01
  Speaker 2: to help solve this, right, and I think that's profound.
• 39:03
  Speaker 2: I think we can have a huge sustainability impact all
• 39:06
  Speaker 2: while actually kind of power up the US economy. And
• 39:12
  Speaker 2: so for us, you know, we've got this mission right
• 39:14
  Speaker 2: now that's focused on trucking, but what we're really building
• 39:17
  Speaker 2: is this capability to release safety critical software and systems
• 39:22
  Speaker 2: and this ability to understand the world. And so you
• 39:25
  Speaker 2: combine those two things, the places we can go with
• 39:27
  Speaker 2: that are profound, right, whether it is you know, last
• 39:30
  Speaker 2: mile delivery or ride hailing applications, or mining or farming
• 39:35
  Speaker 2: or aviation. Like there's it's just going to be a
• 39:37
  Speaker 2: lot of fun over the next decade. So yeah, I'm
• 39:40
  Speaker 2: psyched for it.
• 39:41
  Speaker 1: Next decade sounds fast given the pace so far. I
• 39:45
  Speaker 1: was surprised when you said next decade at the end there.
• 39:48
  Speaker 2: Yeah, I think that like anything, there's a point where
• 39:52
  Speaker 2: you have built a foundational base and then suddenly you're like, oh,
• 39:58
  Speaker 2: now I have this power. Now I have the superpower
• 40:00
  Speaker 2: to go do these things.
• 40:01
  Speaker 1: It's basically the package of hardware plus software at a
• 40:04
  Speaker 1: reductive level, and you can sort of put it on
• 40:07
  Speaker 1: other things and it'll more or less work.
• 40:09
  Speaker 2: More importantly the process and the data and again does
• 40:12
  Speaker 2: not sound sexy.
• 40:14
  Speaker 1: But the data sounds sexy, Like the more you talk
• 40:17
  Speaker 1: to machine learning people, the more sexy data sounds. Right.
• 40:20
  Speaker 2: Yeah, And you know, and we joke internally, you know,
• 40:23
  Speaker 2: if you if we have three kind of big artifacts,
• 40:26
  Speaker 2: being our software, our data, and our process, if you
• 40:30
  Speaker 2: held my gun in my head and said delete one
• 40:31
  Speaker 2: of them, I'd say delete the software, right, Please do
• 40:37
  Speaker 2: not delete our software.
• 40:40
  Speaker 1: Yeah, that's really interesting. Yeah, we'll be back in a
• 40:46
  Speaker 1: minute with the life around. We're gonna finish with the
• 40:59
  Speaker 1: lightning round. I appreciate your time, and we were almost done.
• 41:02
  Speaker 2: Cool. Thanks for the conversation.
• 41:05
  Speaker 1: So right that your father was a prison warden he was,
• 41:08
  Speaker 1: in fact, Yeah, did you learn any management and were
• 41:11
  Speaker 1: parenting tips from him?
• 41:14
  Speaker 2: Lock them up? Uh? No, absolutely not. You know. One
• 41:19
  Speaker 2: of the things that that stuck with me is, you know,
• 41:22
  Speaker 2: it's it's a difficult job being a prison warden because
• 41:26
  Speaker 2: you have the staff and then you have the inmates
• 41:29
  Speaker 2: and right, and and they're often at odds. And one
• 41:34
  Speaker 2: of the things he said was, you know, make sure
• 41:36
  Speaker 2: you treat everyone with respect. Right that he would, you know,
• 41:39
  Speaker 2: these are people who had made a mistake in life,
• 41:42
  Speaker 2: you know, and you treat them with respect. Uh. And
• 41:46
  Speaker 2: you know, I think that's really powerful and important.
• 41:49
  Speaker 1: What's one thing you learned working at Google?
• 41:53
  Speaker 2: I think big, right, I think I think that was,
• 41:58
  Speaker 2: you know, Larry and Sergey never lacked for vision, and
• 42:01
  Speaker 2: I think you can often find yourself constrained by kind
• 42:04
  Speaker 2: of your own kind of sense of what's a limit,
• 42:07
  Speaker 2: and they just thinking like, no, what, why can't you
• 42:10
  Speaker 2: do that? Why can't you do that? Yeah?
• 42:13
  Speaker 1: I remember, just as a consumer, just as a as
• 42:16
  Speaker 1: an ordinary person in the world, when they started what
• 42:20
  Speaker 1: became Google street View, I was like, yeah, surely you
• 42:23
  Speaker 1: can't take a picture of every bit of every road
• 42:27
  Speaker 1: in America, and actually you can and they did.
• 42:31
  Speaker 2: Yeah. Right. Like I think that that sense of like
• 42:34
  Speaker 2: you don't need to be intimidated by this, like you
• 42:36
  Speaker 2: need to think rationally, you need to figure it out.
• 42:39
  Speaker 2: But you know, I think it's a power one of
• 42:41
  Speaker 2: my professors at Carnegie Mellon said, dream like an amateur,
• 42:46
  Speaker 2: execute like professional, right, And I think that's just a
• 42:49
  Speaker 2: really kind of profound sentiment.
• 42:52
  Speaker 1: What's one thing your time at Google taught you not
• 42:54
  Speaker 1: to do?
• 42:55
  Speaker 2: Oh? What it teged me not to do?
• 43:02
  Speaker 1: Uh?
• 43:03
  Speaker 2: Maybe kind of adjacent that. Like, so when I went
• 43:06
  Speaker 2: to Google, I thought I was a pretty good programmer,
• 43:08
  Speaker 2: you know, I was, Yeah, Carnie Melon, I'm just good
• 43:10
  Speaker 2: at what I did, and I'm not a bad programmer.
• 43:15
  Speaker 2: But like, there are truly exceptional people out in the world, right,
• 43:18
  Speaker 2: And so I think, you know, maybe don't overestimate yourself
• 43:23
  Speaker 2: and kind of again kind of think, kind of think big,
• 43:27
  Speaker 2: don't underestimate what's out there.
• 43:34
  Speaker 1: Chris Sumson is the co founder and CEO of Aurora.
• 43:38
  Speaker 1: Please email us at problem at pushkin dot fm. We
• 43:41
  Speaker 1: are always looking for new guests for the show. Today's
• 43:45
  Speaker 1: show was produced by Trinamnino and Gabriel Hunter Chang. It
• 43:49
  Speaker 1: was edited by Alexander Garretson and engineered by Sarah Burgher.
• 43:53
  Speaker 1: I'm Jacob Goldstein and we'll be back next week with
• 43:56
  Speaker 1: another episode of What's Your Problem.