Name: Brainstorm Tech 2023: Autonomously On The Road
Uploaded: 2023-07-12
Duration: 17 min 16 s
Description: Shawn Kerrigan, Co-founder and COO, Plus Raquel Urtasun, Founder and CEO, Waabi; Professor, University of Toronto In conversation with: Phil Wahba, FORTUNE

Shawn Kerrigan, Co-founder and COO, Plus Raquel Urtasun, Founder and CEO, Waabi; Professor, University of Toronto In conversation with: Phil Wahba, FORTUNE

Brainstorm Tech 2023: Autonomously On The Road

Conferences

Autonomous trucking. It looks like self driving trucks are almost ready for their close up. We will find out it's promising technology that can help the industry solve problems like labor shortages since many truckers are retiring and even fewer are taking it up as a career. These days, there are also limits in terms of how many hours truck drivers can drive. So the largest retailers are starting to gingerly dip their toes in to see what is possible in the hopes that self driving trucks for at least part of the distribution process, at least can be faster, safer, more reliable than expected. And so far investors are are there, they're pouring billions of dollars into the 18 wheeler revolution. Estimates are suggesting that up to 90% of long haul trucking could soon be self driving. So those are some, some big numbers. We'll see what happens. Let's introduce our guests to hear more. Joining us today is Dr Sean Kerrigan, the coo and co founder of Plus, it's a company dedicated to transportation safer and greener. And we're also delighted to welcome Professor Raquel Orison, co founder of the Vector Institute for AI at the university. Of Toronto and CEO of Wabi, a pioneering force in self driving research. Please join me in welcoming to the stage along with Fortune's Phil Waba. Well, good morning. It's really fun to talk about technology that has real applications in life. But before we get started, let's familiarize ourselves with the technology. So why don't we keep up the video from Raquel and Wabi and as it runs, you can describe what we're seeing and then we'll get to you. So the video ready for or is this a film noir tribute? Ok. Well, then, you know, maybe just tell us and then the the the video will come on eventually, hopefully, right? So what you're seeing and your imagination, right? Is this amazing technology that we are building at? Um So the way to think about that is we are bringing next generation technology A I to solve uh at scale uh tracking uh level four tracking, which means without the driver. Um And in particular, we are, we're building technology that comes from a genetic A I which you know, these days, every one of you knows what it is. You know, we started a company 2.5 years ago. It was a big unknown. But the idea is that thanks to this technology, you can solve the problems that you see in the industry today that is too capital intensive, right? As well as that it's been making progress, but it's far from commercialization at this scale, which is what we all of us want to do. And so thanks to this technology, I guess we are able to generalize all the situations that happen on the road as well as we have a simulator. That is the same as the real world so that you don't need to drive, you know, millions of miles to really see all these rail cases that you need to handle during, during your driving. So very excited to be building this company and really bringing this next generation, you know, 82.0 to the market today. And you know, hopefully that's a good description of what the video is. OK. And Sean, yeah, thank you. So, so plus is an autonomous technology company focused on making driving safer, easier and more sustainable. So we started in Silicon Valley back in 2016 and since have expanded to a global footprint most recently opening up an office in Munich, Germany last year. So we focus on developing a range of level four to level two plus plus technology applications. So we focus within the level four space on fully driverless trucks. We think trucks is the place where driverless L four will first be commercialized. We think that that commercialization will take some time though and it requires a progressive approach to get there. And so that's why we've taken that same level four technology and actually applied it into level two plus plus type applications right, looking at what can we do for drivers today and with that, we can make driving safer, easier and more fuel efficient. And so we are now working with a range of partners, folks like Aveo and Bosch and Luman and Nicola to really bring that technology to market at scale. Ok. So one of the things you just mentioned was one of my first questions for you, which is why will trucks be self driving before cars would I mean, is it a question of you? Tell us both of both of you? So I think it's because it's a, it's that technology is a better fit for the problems that trucks face in terms of driving. So today in trucking drivers, drivers can drive 11 hours a day and during that time, it tends to be very stressful driving oftentimes if you're in stop and go traffic or other types of conditions. And so with this technology, you can really address many of the types of challenges that they face in trucking today. And we think that the ability to get to driverless first there, it's just that it's it's a better fit for the types of challenges. And so we think that will happen first. And then in addition to the driverless implementation, we also think that that's where applying that technology for drivers today has also had the most impact. And so that's why back in 2021 we actually took our technology and applied it first in this level two plus plus format. And so now we actually have customer trucks on the road today, delivering freight with our technology and it makes the driving safer, more efficient and it makes the driver shop more more convenient and more sustainable. And so that addresses some of the driver shortage challenge that I think was part of the introduction. Ok. Right. Go. Any thoughts on that? Yeah. So we'll echo, you know what he's saying that basically, you know, we see, you know, pure utilization, right, when you are a human driver because of not just the congestion, but it's also that the freight network is really not optimized. Oftentimes drivers are really waiting many, many hours for their cargo to be ready. So there's a lot of digitalization that can also help with this, with this problem. But I think something that many, many people don't know about is actually the conditions, the less conditions that some of these drivers have where if I think it's 7% there's only 7% of the parking spots for truck drivers that are actually available. What this means is that you know, what they have to do is that they need to go and for example, park on Walmart where they don't even have access to bathrooms, right? So you can imagine the conditions that these these truck drivers have. So this is one the reasons you know why we want to empower with our technology so that they can do focus on the local P and L right, be home every night, being able to have a family improve those conditions and safety, which is, you know, for some of us that have many people around us that have had severe traffic accidents, it's something that is very dear to our hearts and we believe that this technology is super human, right? And once it's ready and you know, it will really improve our safety of our roads and in terms of uh convincing regulators and the public that this is safe. I mean, do you think it's easier to explain to people why trucking uh can go autonomous before everyday uh drivers? You know, it's, it's uh there are obviously fewer truckers, truck drivers on the road than Joe and JQ public is. And will the success of this, do you think enable uh driverless cars for regular people? So, so I think we see trucking is the first place that this technology will be really commercially deployed. And I think then it will progressively expand scope in terms of what can be done. So I think yes, it will eventually contribute to also autonomous cars from a regulatory standpoint. I think what we've seen is that regulators are generally very supportive of this technology because everyone sees that the huge benefits this can bring in terms of safety. I think that the key hurdle that we all face in terms of getting to the point of being able to commercially deploy them as driverless trucks, though, is being able to have enough experience to really demonstrate both to ourselves and to the regulators and to the general public, just how safe the technology is and that it's mature enough to be deployed at scale. And so that's another reason why we think this progressive approach is really the way to go. But yeah, we generally see regulators as very supportive and engaged in this process. And so we participate in a lot of the different industry consortiums that are working with regulators on how do we, how do we adjust some of the regulations today to support the future deployment of this technology and what's going to fuel the who who will be buying this technology? Will it be the logistics companies we know today? Could we see the large retailers decide that they want, they truly want their own trucking fleet? What's going, what's this going to do to how goods get to stores? Yeah, I think that from the, we think today that the buyers are this technology because we think the first actually applications will be in the driver in format where you take it and apply it as an advanced level two plus plus system to support the driver. And there, I think the buyer is the traditional buyer, it's the fleets and they do that because they're really looking at how do they attract and retain the drivers, they need to support their fleet. There's a huge driver shortage in the US and that's projected to continue to grow. And so with this technology, it makes it easier to bring new drivers into the profession and enable existing drivers to stay in the profession longer. And then as we look ahead to business models and approaches for the fully driverless applications, I think it remains to be seen exactly how that's structured. But our view is that we are a software technology provider. And so we're really supplying that technology to tier one S and OS for them to deploy this technology. And so we have no aspiration to really own or operate our own fleet, right? So we'll see kind of how the value chain shakes out. But our view is that, you know, our goal is to supply the technology and focus on what we do best. Ok. Yeah, in the case of so we are already focused on level four, as I said, so our customers would be for a sample fleet shipper, right? As well as potentially OEM some of them which have ambitions, ambitions to become a transportation company and maybe giving a bit more color on the on the business models. There is like two business models out there. One is driver as a service. The other one is transportation as a service and transportation as a service is very capital intensive, you own the assets, you become a transportation company. For us, we are more on the technology front. We don't want to own those as is we want to empower people that know how, know how to run a business, a transportation business, which is something non trivial. The margins are actually pretty, pretty slim in the regular world, right? And what we want to give them is the ability to have this super human driver, which is our technology. Well, I'm delighted to have both of you here because as we were designing this session a few months ago, you know, we had a short list of companies and a couple of them are not here anymore. Uh So what does it take at this point for you guys to, you know, to keep growing and you know, to get to that point where commercialization at a scale that you need is happening. So I guess what's the secret to surviving at the moment? Yeah, so I I will keep my take on this, which may be a bit different than yours since our strategy is very different to whether what we see today. So there has been a lot consolidation, right? And you know what I will call 81.0 where, you know, people thought that technology is ready, you just need to harden it and then you will solve this as a scale. And you know what we see today is that this approach is far from actually being solved, it's far from being a solution is extremely capital intensive. You look at some of the players go all the way to 3 $4 billion a year. Right. And, you know, in a market, in an economy like today, it's really hard to see that for many, many years to come. Right. So there's two things you can do. One is, you know, the approach that they are taking off, they try to commercialize with level two and then you have the customer base and then you know, try to go to level four for us. What we did is something very different, which is transport fully, you know, build new technology that scale level super capital efficient from day one. And then you are in a different ball game compared to everybody else. So we spend a fraction of the cost of anybody else with technology that you can develop so much faster. And just to give you an understanding what this gene technology is for us is think of the transformation that the GP the GPT staff has made for natural language processing, right large in which models. So we are bringing this to self drive. So the abilities that this technology has is far superior than everything that you have seen before. And since you can learn from this automatically, then suddenly you don't need humans to develop that technology as much. You don't need a fleet of vehicles. So it's very large to drive millions of miles, right? So that suddenly you are in a totally different ball game. That's what the capital needs. So I'm very excited when we are today. And, uh you know, we have some really exciting, you know, commercial and other announcements coming soon. We have, you know, we driving fleet out there in the world, um you know, driving autonomously with a safe driver now, um and definitely making tremendous progress in the industry. Yeah. And so I agree with you that there's been a lot of consolidation in the industry. I think that what I say is that everybody that we talked to still believes that trucking will be the first place. This technology is really commercially deployed at scale because the value proposition is just so clear in terms of the safety and the efficiency improvements that come with it. But we think it will take some time. And so that's why I think that we've taken this progressive approach because that really enables us to demonstrate the capability of the technology and acquire enough data to prove out that we've really covered all the scenarios that are necessary to scalability, deploy it. And then there's also the notion that this actually creates tremendous value for drivers today when you apply it to vehicles with them. And, and so what we're seeing with this sort of shift in the market is people start to think well, ok, level four and trucking is still coming just a little bit longer. But what can we do for drivers today? And the fact that we started commercial deployment in 2021 and we're now working with partners like Veco and Bosch and Lumina Nico and others on factory scaled production of this technology in the future is really, I think what's driving most of our focus and where we see the, you'll get the initial scale deployment which will then really facilitate the eventual deployment of level four as well. So how far are we away from when I peer up at a truck? Um, going next to me on interstate 87 to seeing no driver. Uh, no, nobody in the driver's seat, but maybe somebody still in there and how far until we see them in city centers, maybe dropping off a shipment at one of the big box stores? I mean, is this, you know, a few years away, more than a few months away probably, but you know, what, what's your sense of when this is really going to be a reality that we'll notice in the public? So definitely more, more than a few months away. We think it's, it's some years away. But I think that the first thing you'll actually see is trucks that have a comprehensive sensor suite around them with the driver still there behind the wheel, but more supervising the system than really fighting to keep it in the lane all the time. And so I think that will be the first applications that you'll see on the roads and there are actually some already out there. Uh, and then I think it will be some years after that when you start to actually see the, the driverless trucks first on the sort of the highway, long haul segments. Uh, and, and I think it will be much longer before it goes into any urban settings. Ok. Yes, I think, you know, maybe one thing to, to mention that we haven't talked about this unit economics, something that is not as trivial as people have said for self driving. And one of the things to, you know, in the mother that you're talking about to think about is whether you need to reduce the sensors because the economics of having a human uh supervisor, you know, makes things much harder versus not having a human there in the first place. Now, in terms of when it will be ready. So for us, we believe that to deploy at the scale uh you, you need to integrate our technology at the factory level with our OEM partners. And the reason we believe this is that this is the safest uh you know, more scalable uh robust solution that you can have. And the, and the O EMS are all building these platforms that will come live depending on which OEM you know, in 23 years. Um and sort of that, that's a precursor to having this technology deploy at this scale as well. But we will see this, you know, relatively soon and you know, it's going to be very, very excited. Ok, great. Well, join me in, thanking Sean and for joining us and telling us about what's coming to our roads in a few years and I will try not to flip any truck to the bird. Thank you for having us. Thank you.