Aurora Innovation, Inc. (AUR) Earnings Call Transcript & Summary

Hello, everyone. Thank you for joining us for this session with our Aurora Innovation as part of Deutsche Bank's 2023 Global Auto Conference. My name is Winnie Dong, and I work in the U.S. Auto and Technology Research team here at Deutsche Bank. I'm very pleased to be joined by Richard Tame, the CFO of Aurora. Aurora was founded in 2017 by CEO, Chris Urmson, the former CTO of Google's Self-Driving team, which is now Renault. Aurora have partnerships with companies like PACCAR, Toyota, FedEx and Volvo and recently introduced its Level 4 self-driving capabilities. It's looking to commercially launch autonomous trucking service by end of 2024. The format of this session will be a formal presentation from Richard, followed by a fireside chat around some of my prepared questions as well as topics from all of you on the call. [Operator Instructions] So with that, I will turn the floor over to you, Richard. Thank you for being with us.

Thank you very much for having us. Great to be here, and thank you to everybody who's tuned in to listen to learn a bit more about Aurora. So next slide, please. So I'll give you a brief overview of Aurora, like Winnie said, then we'll jump in to take some Q&A. But first, as is customary and because I'd like to keep the lawyers happy, this is our safe harbor statement. You might had a chance to read that. We'll move on to the next slide. So what do I want to do? So I want to talk briefly about the promise of self-driving technology. Why there's never been a greater need for autonomous vehicles. And why we expect that Aurora, that it's our technology and our path to commercialization, which will enable us to deliver this technology at scale. So our mission at Aurora is to deliver the benefits of self-driving technology safely, quickly and broadly. And together, we're building the next-generation of self-driving technology, which is focused on moving goods and people autonomously. And I think like just this vast opportunity to transform transportation motivates our work, whether it's to make the roads safer, whether it's to increase economic opportunity, increase people's access to transportation, there's just such a lot that we can do with this technology, and it's awesome to be able to work in this space. Next slide, please. So what are we building? We're building the Aurora Driver. This is our Level 4 self-driving product and this is a quick overview of what we're building. So the Aurora Driver's hardware, software, infrastructure and development tools are designed to work across all vehicle types. That's from Class 8 trucks through the light passenger vehicles. And this common core technology approach is unique in our space. And the commonality ensures that lessons learned, development efforts, hardware improvements and the cost reductions made the Aurora Driver benefit every vehicle that it powers. So something that the truck learns on the road, it's immediately available to the cars that are on the road and vice versa. But importantly, what we believe is that this common core technology, which again is unique in this space, unique to us, we believe that's going to enable us to deliver both on our first priority market, which is autonomous trucking, and then down the line, passenger mobility. And we think this sets us up really nicely to address not just one, but both of these very significant markets. Next slide. So you probably know, if you're familiar with the space, that these transportation markets are just massive. What Aurora plans to do is to first launch in trucking, which we see as a $700 billion market in the U.S. and that has attractive unit economics. One way to think of it is a truck driver gets paid significantly more than an Uber driver gets paid today. And then we believe that this technology has the potential to meaningfully reduce the cost of ride-hailing. And we think that as the cost of ride-hailing comes down, supported by autonomous technology in the future that this will help to unlock the $1 trillion personal mobility market via shared mobility. And then further down the line because this technology works and has been designed from day 1 to work on all vehicle types, we also expect it to expand into local goods delivery, which is another big market in its own right. Next slide, please. So Aurora is really motivated by the immense impact that our technology can have. The most important thing being safety. There's no surprise and no surprise that that's the first part of our mission statement. Safety is fundamental to everything that we do at Aurora. And specific to trucking, there's crazy numbers. There's like approximately 0.5 million truck crashes every year on the U.S. roads. 365 days in a year. So 0.5 million truck crashes is like a phenomenal number of accidents that are happening every day. So autonomy can help with that. It can make the roads safer. But also, what we're not doing at Aurora is trying to do a science experiment, we're really trying to build a business. And what autonomy can solve for is a number of the major pain points that you see in the trucking industry today. So we can help to alleviate driver shortages. The industry is short about 80,000 drivers to date and that shortage is projected to double by the end of the decade. The Aurora Driver provides a scalable, stable driver supply. You're probably familiar that truck drivers in the U.S. in particular, have hours of service limitations. So that's 11 hours now, so they can drive for 11 hours, but then they need to pull over and they need to stop and have a significant break before they can go and do another 11 hours. The Aurora Driver being a computer system doesn't have hours of service limitations. It can operate 24/7. And therefore, we can deliver great [ vesta ] with the Aurora Driver and we can cut transit times in half. It's no surprise and everyone is aware that there's high fuel costs that can really hurt the trucking industry. The Aurora Driver will allow more efficient vehicle operation and has also an ability to reduce fuel use and emissions by 10% and even upwards of 20% to 25% depending on the speed at which our trucks will travel. So because we don't have hours of service limitations and we don't have to have those big break periods, it's likely that in operation our trucks could drive slower and still get to places much, much faster. And because of the way the trucks are designed, if you could drive slower, you feel efficiency has significantly increased. And also, there's very high insurance costs in the industry. And we believe that our safer operation will help to get insurers comfortable with reducing the price of insurance for trucks. And also because we have this full rich sensor suite on the vehicles, we think there will be significant data available for fault attribution. So it will be much less of the he said, she said kind of thing that happens in accidents that can cause costs and claims to go up because we'll have a record from our various sensors of exactly what happened in any given incident. And longer term, we believe that this technology will transform the way freight moves. It's going to bring previously impossible nearly 24/7 truck utilization and faster freight than we've ever seen before on the roads. So a trip from Houston to L.A. that would ordinarily take about 2 to 3 days, could be completed in about 24 hours with an autonomous truck. So that's really transformational. It also means things like you could reach the entire U.S. within a day with only 2 to 3 distribution centers across the country. Whereas now, you have this deep footprint of distribution centers for anyone who's trying to do next-day delivery. And then also with something that's very important to our potential trucking customers is the potential to double revenue per truck because it can drive almost 24/7 and that's a metric that they think is very important to their business. And then the Aurora Driver is going to help them by making that metric significantly better than they can achieve with a human driver. Next slide, please. So to meet the needs of our customers and to most effectively monetize our technology, the business model that Aurora is pursuing is something that we call driver-as-a-service, and this is structured to be highly capital efficient. So we expect to have very attractive SaaS-like margin structures in the future, and we expect to provide our technology to fleet owners and operators, and we expect to be paid a fee per mile. So every mile that the Aurora Driver is driving your truck, Aurora will be getting a fee per mile for that. And this model is great for us, the capital-efficient nature, the very high margins into the future is awesome for the company. But interestingly, the model also aligns very well with the customers' needs because they like the certainty of being able to pay per mile for their driving task. Next slide, please. So we really believe we're extremely well positioned for success in trucking. We're successfully executing against our road map. We put a road map out publicly. We'll get to that on the next slide. And we're moving very nicely across that road map. At the same time, we're seeing that the competitive landscape is clearer. Given recent news, such as a key competitor who was deemphasizing the autonomous truck inside of their business. We've seen another player winding down and selling the technology. And yet another player has really quite a bit of uncertainty with respect to their path forward. So this is something that we either were expected to happen. It seems to have happened quicker than we thought it would, but we believe that this leaves us as the clear leader in the AV trucking space, best positioned to commercialize this technology and thus to create significant value in the freight ecosystem and also for our shareholders. So we're about halfway through 2023 now and it's proven to be a really pivotal year for Aurora. We continue to execute against our road map to launch Aurora Horizon, which is our autonomous trucking service. Next slide, please. So drilling into our road map, again, we're not trying to do a science experiment here like Aurora is really trying to build the business and the road map is a kind of testament to that and something that you as investors can follow along with. So drilling in, it covers 3 components of the business that are needed to bring the autonomous trucking service Aurora Horizon to market. The first section, the blue section at the top, is the Aurora Driver. That's our self-driving technology that I just walked through. The middle section is -- covers our truck platform and we're working with world-class OEM partners that will enable us to operate at scale. And then there's also the bottom section, which covers commercialization. This covers everything around our customer relationships, our operations and then the infrastructure necessary to effectively integrate the Aurora Horizon into the existing freight ecosystem. So I want to call out that at the end of the first quarter of this year, we achieved our critical Feature Complete milestone and we did that on plan. What this means is that we've implemented all of the capabilities we believe are necessary for our Dallas to Houston launch line and we've removed all policy interventions on that line. We see this milestone as an inflection point on our path to commercial launch because it unlocks the final phase of refinement and validation to close the safety case for our launch. So we expect to plan -- we plan to spend the rest of 2023 working on this. And that will help us to achieve our next critical milestone, which is Aurora Driver Ready by the end of the year. And that's on our path to commercial launch, which we expect to be by the end of 2024. So with our first quarter results, we also shared that the Autonomous Readiness Measure, the ARM, was already at 44% as of March 31. The bar for our commercial launch, people ask us is, how do you know that you're ready and the bar for our commercial launch is a closed safety case. And the ARM demonstrates a strong process -- progress we're making towards closing our safety case with nearly half completed already, and this keeps us on track to achieve the Aurora Driver rating milestone at the end of this year. Next slide, please. So I want to move and take a moment to highlight our philosophy and strategy around bringing our self-driving technology to market. So at Aurora, we've spent the last 6-plus years building the technology to capture the tremendous opportunity in autonomous driving. It's very important for us that we focus just on what we do best, which is building the self-driving technology itself. We then partner with industry leaders to support commercialization of the Aurora Driver. We're great at building technology, but we don't want to be a truck manufacturer because there's awesome truck manufacturers who've been doing it for a long time and can do it much better than we ever could. Similarly, we don't want to run a freight network because there's great freight network -- freight customers who are out there with very dense and sophisticated networks and we can help them to grow their business. We don't need to develop a freight network ourselves. So in trucking specifically, we're partnered with 2 of the top 3 North American trucking OEMs, PACCAR and Volvo. They together make up nearly 50% of the market. We're currently out on the roads in Texas, pulling loads for customers in autonomy every day under the supervision of vehicle operators for now and we're pulling loads for FedEx, Werner, Schneider, Hirschbach and Uber Freight. We also have a couple of collaboration partnerships with U.S. Express and Covenant. And we have an on-site fleet maintenance pilot program with Ryder, which is supporting our pilot operations, and we expect, ultimately, it will support our commercial operations at scale. And most recently, you may have seen that we announced an exciting first-of-its-kind partnership with Continental, which is of course one of the world's leading technology manufacturers and Tier 1 automotive suppliers. I'm going to dive into that a bit deeper in a moment. And lastly, once we've launched our trucking product, we expect to move into the ride-hailing product because you'll remember that I said our technology was designed from day 1 to work with any vehicle type. And when we're ready to target that market, we also have very strong partnerships with [ FedEx and Uber ]. [Technical Difficulty] So moving to the Continental relationship. This is a really, really, really exciting partnership for Aurora. But what they're going to do is they're going to co-develop and manufacture a future generation of the Aurora Driver hardware kit and they're going to manage the complete hardware life cycle; design, manufacturing, food service and then warranty. And Continental have already begun to invest in what might be ultimately up to $300 million in NRE to scale the Aurora Driver hardware, and that's awesome for Aurora because it allows us to avoid near-term capital commitments to develop and manufacture that hardware. When we look at this, we believe this is a testament to Continental's confidence in both an autonomous future, but also Aurora's leadership position in that feature because they've looked at the market, they've talked to people and they've chosen to partner with us on this important technology, which is new to their business. What's super unique about this partnership is that it's a Hardware-as-a-Service model. So under that structure, Aurora is going to pay for the hardware in the future on a per mile basis. So as revenue comes in because we're driving with the Aurora Driver, we're going to be paying for the hardware that's part of that system. And this aligns with and supports a highly capital-efficient DaaS business model that I talked about before. So financially speaking, we believe industrializing our hardware kit through this partnership is going to help us achieve the commercial scale and the cost structure necessary to support our long-term profitability objectives. Next slide, please. So we often get questions about regulation in the U.S. And looking towards commercial deployment, you can see why people ask that question. But this is what the map of the U.S. regulations look like today. Very different to Europe, the U.S. is a self-certification framework. What that means is that if we were ready to deploy, we could do so in the vast majority of states in the U.S. today, and that's what stark blue on the map. So in those states, if we as a company were comfortable, we could have trucks driving down the roads without human drivers today. That being said, we are very proactive on the government relations front today and we have been throughout the life of the company. And we work collaboratively with regulators and lawmakers at the federal, the state and the local levels because we really believe that's the right approach. We want to be able to educate these stakeholders as well as the public about AV technology and so build trust because we understand that this industry is in its infancy. Next slide, please. So wrapping up, we really believe we're very well positioned for success. So Aurora has a leadership team with unparalleled cumulative industry perspective supported by deep technical experience all across the organization. We have differentiated technology across the software and hardware stack. And we have strong strategic partnerships that really support our commercialization. Again, we're not interested in doing a science experiment, we're interested in building a business that's going to grow and be substantial and create a lot of value in the world. To that end, we have a defined path to commercial launch in trucking. And we believe that that can then rapidly scale operationally. And we have a driver-as-a-service business model that underpins really attractive unit economics and high gross margins into the future. So with that, Winnie, I look forward to taking your questions and any that the audience may have.

To start, I wanted to go back to the competitive landscape that you were talking about, the implementation, how you mentioned some competitors perhaps deemphasizing the space. I think they're all trying to solve similar problems they are facing in the industry. So I guess, as it currently stands today, why do you think that some of these competitors are deemphasizing? What are the broader challenges that maybe perhaps you all now face with the technology progress that the industry has made? And what are your key differentiations from these competitors?

No, it's a great question. So maybe I'll take the key differentiators first. So our team, the Aurora team, has been at this the longest, both their time at Aurora, but also in prior companies. So they fundamentally know what's going to advance this technology to market. And they also, on the flip side, really understand what's held it back to date. So when the founders started Aurora 6-plus years ago, they were confident that the recent revolutions that had happened in AI and machine learning was going to be a necessary breakthrough to really commercialize self-driving technology. So that insight, as they looked at it, really drove their clean sheet approach to building the Aurora Driver and they were able to leverage those advancements in the work that we're doing. So this AI research has led to cutting-edge machine learning architectures with diverse usage -- diverse uses. And we've also made foundational technology investments in our proprietary FirstLight Lidar and simulations. So when you think about the competitors, we don't think anyone in trucking has the ability to see as far as we can. We have FirstLight Lidar, it's a proprietary LiDAR FMCW technology that we think -- well, we know nobody else has it. And what this allows us to do is see far and then it also allows us to get like simultaneous as well seeing how far enough is we can see the velocity is traveling. So we know can if it's stopped, if it's moved away from us or moving towards us, and all of that information is really awesome. As you can imagine, on the highway when you're driving a truck 65 miles an hour to really allow you to make decisions, the system to make decisions as well. So there's all of this kind of differentiation that we have internally combined with that knowledge of how to make this work based on prior experience and we really see that that puts us in a really great position. And then, of course, as you mentioned in the question, the competitive market is certainly being clearing. Things that we thought would happen at Aurora have been happening. And some of the other companies have run into these kind of challenges that we feel like we've already addressed because we can't see far enough on the right because we did make those foundational investments in technology in our development program, and they are paying off as we continue to tick the boxes on our road map as you move towards expected commercial launch at the end of next year.

From an industry pain point perspective, we've addressed -- you're addressing essentially long-haul drivers from your achievement, you mentioned 80,000 of driver shortage right now that stands in the industry. But there's also like great liability in the trucking space, right? One single accident can cost millions of dollars for a fleet, which they're trying to avoid, and they come across as extremely risk-averse. So what is your process of approaching your potential fleet customers with this new technology? Imagine no one being in a truck driving down the highway, how do you bring them on board with your products?

So again, a great question. Those nuclear verdicts that you hear about in the press and even like in Texas you see all of these billboards about people who want to sort of pursue trucking claims, they're absolutely top of mind for our customers. But at the same time, we see tremendous enthusiasm from our customers to get the Aurora Driver technologies launched. So we're running autonomous pilots, again, under the supervision of a vehicle operator for now, for many potential customers like the FedEx, Werner, Schneider, Hirschbach and Uber Freight. So that's a testament to the fact that they see that this technology is going to be important to them in the future. And irrespective of what might be happening in sort of the trucking space with that liability and those high claims, I think they fundamentally understand that a computer is going to be safer than a human. A computer doesn't get tired. A computer doesn't fail drug tests. A computer is not worried about the argument that they had with their family before they left for that long-haul trucking journey. So we -- the customers understand this and they think that they see the value in the technology in the form of safety that will pay off in ideally fewer accidents, maybe less severe accidents, and thus, some of those large claims you would expect to go down in frequency and the time. That being said, we're very aware of the environment that we operate in and we've made a really conscious effort to engage with insurance carriers. So we want to educate them on this new technology and we want to educate them on our planned commercial launch because we really believe them understanding what we're doing and bringing them along for the journey will benefit us and also our customers. By having them see as soon as possible that this is great technology from a safety perspective, obviously, some insurers -- insurers often, because of their actuarial backgrounds, they typically like to see a lot of data before they're comfortable about assessing risk. So we think that what makes sense for us is to really engaging them at different level and try to say like, look, this is why we're safer, this is why we are going to be safer and you don't need hopefully like 20 years of data before you can get comfortable that this is going to be safer technology. And as an example, just last week, I was in Texas and we were hosting a group of insurers and they got to talk to us, they got to listen to us and then they also go to experience the technology first-hand. So we really believe the Aurora Driver is going to be safer. And we also think that when there are accidents, there's going to be a ton more data available for fault attribution. And those 2 things together, we fundamentally believe means that the insurance industry is going to recognize that autonomy is much safer than a human driver and then we expect the cost of insurance will come down significantly.

There's a slide in the presentation that sort of captures the regulatory landscape across the U.S. on many different states. In general, do they represent [ tailwind ] or is there a need to convince the regulatory body state-by-state of your technology? And then how do you see that regulatory environment evolving in the next 5 to 10 years as the technology continues to move forward? You also mentioned, obviously, some competitors also deemphasizing this technology. How is that? Is that good Or is that bad for you from a regulatory perspective?

So again, like it might be sort of unintuitive for people who aren't familiar with the space. But as we said in the presentation, if we were ready to deploy today, and by that I mean have vehicles drive themselves without a human driver, we would be able to do so in the vast majority of U.S. states today, about 45 of the 50. We'll note the one meaningful outlier is California. You see progress now with the DMV moving to enable heavy trucks to operate autonomously. But then you're also starting to see some pushback from labor by kind of labor supported driver bills. So we have a very, very strong government relations team. It's been something that's been invested in throughout the life of the company. And it's very engaged on the issue in California. We've seen similar proposals in probably like 7 other states, probably through Republic and Democratic controlled legislatures. And all of those either stalled or would defeat it during the legislative process. So California is something that we're monitoring, but it doesn't seem like representative of the rest of the country. On the federal front, when our Founder and CEO, Chris Urmson was at Google, he understood that this would be important to kind of bring the legislation, bring regulators along for the right -- and he was one of the first people in the autonomous industry to go spend time with the Federal Department of Transportation. So we continued to kind of build and nurture that relationship and thus build that trust over the years. And we've had senior members of the Department of Transportation down in Texas, experiencing our trucks on the road. We continue to engage with them. There's a complexity of understanding. And I think the regulators, they understand that not everybody in the industry is the same. So when we talk with regulators, like we really get the sense that they seem to appreciate our approach and specifically our transparency in contrast to maybe some of the other people who either don't talk to them or either don't talk to them to the level in depths that we do. Federal legislation, we've seen goes of that in the past. And what we really expect is, once the promise of the technology becomes more visible and then the impact is more tangible on the world, like there might be a push for federal legislation. But we all know that that's a really, really long process. So to the timeline that you referenced in the question, that's kind of likely when you see something on the federal legislative front. But again, before we go to the next question, super important point is that we don't need federal legislation to be able to launch and operate, but we do think that it would likely be beneficial for the industry over the very long-term.

There are some indicators that you also alluded to in your presentation, sort of measuring how ready your technology is for testing and for commercialization. So I'm just going to allude to them quickly. So can you explain what is the Autonomy Readiness Measure and Autonomy Performance Indicator? In Q1, these metrics were 44%, 96% respectively. Can you just dive more into, was it a measure? And what do they inform about your technology coming closer to launch?

We don't have revenue right now, as you know. So it's kind of hard for people to track along the company's progress. And so both of these measures were things that we've put out into the world to try to help people kind of follow along. First of all, the Autonomous Readiness Measure or ARM as we refer to it, what that does is, it demonstrates the percent complete of our safety case. So we know we'll be ready for launch when we have a closed safety case. And you can think of the ARM as the amount of work needed to move from a Feature Complete to our next critical milestone, Aurora Driver Ready and then to commercial launch. So with the first quarter, we showed that the ARM was already at 44% as of March the 31st. And what that does is, it kind of demonstrates the strong progress that we're making towards closing the safety case, because you can think of it as like we're nearly half completed on the safety case. And that's obviously keeping us on track to achieve Aurora Drive Ready later this year. The metric, we consider that to be a supplementary metric. The ARM is the thing that tracks our ability to launch, because that's what's tracking the progress on closing on our safety case. But we also put out the API, the Autonomy Performance Indicator. And what that does is it's tracking our progress, tracking our performance to successfully operate the Aurora Horizon service and it's commercially representing the same. So it's allowing us to understand not just kind of the state of our technology, but also the maturity of our processes and our procedures and operate in a natural business in the future. So what the indicator does is it penalizes the use of on-site support, and that's going to be the most expensive support that's provided to enable the Aurora Horizon service. The API was 96% for quarter 1 as we reported. We don't anticipate that the aggregate API will be 100% even at launch because certain situations, like flat tires for example, always going to require on-site support. And thus, we don't look at the API as a launch path.

Can you dive a bit more into how your technology can -- or how transferable is it from autonomous driving in the trucking space to the ride-hailing space? I would imagine the type of growing environment they each go through is somewhat a little bit different. So I guess, how transferable will your technology and your driving policy be in that trucking type of growth to more urban dense ride-hauling type?

So as we said before, we're unique in this space. The Aurora Driver has been built from day 1 to operate on any vehicle type. And this common core approach, we believe we're the only people who have it. So today, we have trucks driving autonomously under the supervision of humans in Texas and we also have passenger vehicles driven autonomously under the supervision of humans. So the technology is the same, right? It's the same hardware, it's the same software that powers both types of vehicles. And then what a car learns in more dense urban environments translates directly to a truck and what a truck learns on the highway translates directly to cars. So it's extremely transferable. It's basically the same common core technology. So what is really interesting about this is it really sets us up for success in the future because what we're going to intend to do is we're going to launch in the trucking market, which we know to have the best unit economics and has the similarity of the infrastructure, like an interstate is kind of the same in Texas as is in Arizona. So you have that similarity that kind of helps you to operationally scale. Then when we're ready to launch a ride-hailing product, because we can see further than other people with our FMCW proprietary FirstLight LiDAR and because we're partnered with Uber, what we expect to be able to do is say, look, like we can do rides that other companies might not be able to, so we can start having a ride-hailing product, to say, doing airport runs, going from an airport to a hotel district or an airport to a convention center. And that in many places will look very much like a trucking thing. It's going to go from the airport, it's going to go short distance, it's going to go into the freeway and then it's going to come off again and it's going to go to the hotel district. We can do that. If you're a customer of Uber and Aurora Driver can do that, even that small subset of routes to start off with, you're going to be happy. And if there's something that we can't do then, you're also going to be happy because Uber has many human drivers who can come and pick you up and take where you need to go. So this kind of drafting between the truck and the car is super exciting for us because then as the car learns more and gets more and more dense into the cities, that's when the truck can also go more and more dense into the cities and start going kind of store-to-store. So we can go to the Walmart ultimately. So that kind of [indiscernible] and Flywheel is only impossible because we have the same technology on all the vehicles. And again, that's a conscious decision and a conscious development decision to make sure that whatever hardware and software is on the trucks, powers cars and the vice versa.

I actually wanted to discuss a bit more, but I realized that we're actually up on time. So thank you for such an interesting discussion and conversation. Thank you for your presence. Really appreciate it.

Yes, thanks for having us. It's been really awesome, and look forward to talking to you more.