Good morning, everyone. I am Stacy Feit, Vice President of Investor Relations, and on behalf of the entire Aurora team, welcome to our 2026 Analyst and Investor Day. The last time we got this group together was two and a half years ago. I want to thank everybody in Dallas here who joined us then, and who came now to come out to Dallas.
We know how much time it takes out of your schedule, and we really appreciate it, and we want to thank everyone online as well. Back two and a half years ago, we were in the process of closing our Safety Case for commercial launch. Today, we have our second-generation trucks operating driverlessly on the road every day. It has been quite the journey. We have made such tremendous progress and are in such a different place today, and we really want to thank everybody for being on the journey with us this time. We have a really great program set up for you today that will show you how Aurora is driving the inflection in autonomous trucking. As far as our agenda goes, our Co-founder and CEO, Chris Urmson, will set the stage for the day. Our President, Ossa Fisher, will then detail why our commercial flywheel is accelerating.
She will then host a driverless customer panel that brings together executive leadership from some of our customers, including Detmar Logistics, McLane, and Werner, and they will give you their first-hand experience hauling commercial freight with the Aurora Driver. From there, Sandor Barna, Senior Vice President of Hardware Products, will outline our multi-platform path to scale. He will then host our partnership ecosystem panel, which will feature partners from our OEMs, our tier 1 and upfitting partners, including Volvo, PACCAR, AUMOVIO, and Roush. They are going to show you how this all comes together. Next, CFO David Maday will host a fireside chat with one of our key insurance partners, Apollo. We know liability and insurance is top of mind for many of those in the investment community, so I think that will be a really exciting conversation. Dave will then walk you through our path to scaled economics.
We will bring the presentation portion to a close with some closing remarks from Chris before opening a Q&A. Also note, we have built in some brief windows for Q&A during all of the panels in the fireside chat, so there should be ample time for questions. Then finally, for the most exciting part of the day, I am obviously a bit biased, but most of our attendees here will have the opportunity to participate in a driverless ride-along on public roads, seeing the Aurora Driver in action with nobody behind the wheel. Now, before we dive in, we have the usual housekeeping item to cover. This is the most boring part of the day. You know the drill. I need to quickly read this. I need to note that we will be making forward-looking statements.
These statements are based on assumptions and beliefs as of today and are subject to risks and uncertainties that could cause actual results to differ materially. Please refer to the risk factors and other disclosures in our most recent 10-K and our other filings with the SEC. Our discussion today may also include non-GAAP financial measures. These non-GAAP measures should be considered in addition to and not substitute for or in isolation from our GAAP results. Now, let's get into the fun stuff. Enjoy.
[Presentation]
[Presentation]
Hell yeah. Welcome and thank you for joining us. I know it is a lot to ask you to come down here to Dallas and spend a whole day with us, so we really appreciate that. We hope this is going to be worth your time today. What I really want to get out of today is to learn a little bit about Aurora, a little bit about where we are going, and hopefully come away with the same level of excitement that I and we have for the road ahead. It is going to be a hell of a few years for us.
Today, Aurora and the freight industry are at a point of inflection. We no longer need to talk about when autonomous trucking is going to happen. We can talk about how it is happening on the road today. The Aurora Driver is here now and we are ready to scale.
I like spending time understanding how new technologies came into existence as you look back over history. In doing that, I have seen this common pattern. Most of the world's most transformative products have something in common. They spend a couple of decades wandering through the wilderness, really trying to figure out how to actually make it work, and then all of a sudden they look inevitable. For decades after Karl Benz invented the automobile, it really was this novelty for rich people that no one thought had a hope of replacing the horse-drawn carriage. It took 20 years of refinement. It took Henry Ford innovating the assembly line. Then we were able to get to the point where we could start to put a car in every driveway. Boy, did that change the world.
It didn't just change how we moved, it changed how we lived, it changed the world, the shape of the world we live in. In the shipping industry, Malcolm McLean began experimenting with containerization. At first it went nowhere. It was opposed. Freight had moved the same way on ships for literally centuries. But 20 years of refinement and this magic moment of product market fit in the Vietnam War suddenly opened the world's eyes to this, and all of a sudden, it was inevitable. It opened the way for new economic hubs and it built the global supply chain that we all rely on today. Finally, Vint Cerf and his buddies back in 1969 connected a few nodes to create the ARPANET. Initially, it was very much a curiosity. It was for academics to talk to one another over a network.
It took almost 30 years, the revolution in telecommunications, to create a global telecommunication network and the web browser, but ultimately, they created a platform that fundamentally changed not just the way we communicate, but the way we connect and how we do commerce. Cars, cargo, information. The same pattern. Investment for a period of 20 years, and then suddenly inevitable. Not coincidentally, these are all the bits of technology that we need to enable what we're doing now. Each of them transformed our world and created immense value, and our technology is on a similar shape of trajectory. 20 years ago, DARPA kickstarted automated vehicles with their grand challenges, these robot races across the desert. I took part in those early competitions, ultimately leading the team that won the 2007 DARPA Urban Challenge. We looked at what we'd built at the time, and we had big dreams.
This was going to be big. It was going to transform the world. We literally had no idea how hard it was going to be. I can tell you if I probably did, I wouldn't be here talking to you about it today, because I would've gone and found basket weaving or something. I then spent seven and a half years leading what's now Waymo. We made tremendous progress, real strides, but we still weren't ready. It's now been about 20 years since those challenges, and I can tell you we are ready today. Our first product, as you all know, is Aurora Driver for Freight. We chose trucking for a bunch of reasons. First and foremost, there's actually a huge need for this technology. The addressable market that we can reach is gigantic, and we think there's an opportunity for incredible unit economics.
The customer decision-makers are dispassionate. If we are safe and we can help them bring value and grow their business, they're going to adopt this technology. That is clear. Finally, there's a tremendous amount of existing infrastructure that we can leverage. As someone who's excited about scaling this business and doing that in a capital-light way, that's cool. I imagine that's pretty cool for all you that we don't have to continue to invest heavily to actually scale and grow the business. To have the privilege of serving this market, we have invested heavily. We've been developing our software and AI systems, our verifiable AI system that's on the road today. We've invested in our second and third-generation hardware that meet our affordability targets, that meet our robustness targets, and meet our ability to scale.
Our OEM strategy means that we can put the Aurora Driver on the platforms and vehicles that our customers demand. Our customers are excited for what we've built, and we're seeing that demand accelerating. Our philosophy from day one has been do what we do best in the world, and then work with amazing, great companies. I believe, and we believe, that we will scale faster and deliver more value if we can focus on what we do best, and they can lean into their strengths. We are proud of the ecosystem we've built. It contains world-class OEMs, amazing logistics companies, hardware partners. It's awesome, and you'll hear from many of them today. Building a connected ecosystem like this is hard. It takes time to build the relationships and build trust. There's really no shortcut to that.
What we are seeing is that with each proof point, each example we put on the board, our ecosystem is accelerating. People look at what we are doing, want to be part of this. They see the benefits to their businesses. This is why I believe we're at an inflection point. The tech, the partners, and the customers are all ready. It has been a hard journey, but the Aurora Driver is now on the same rapid adoption curve as we've seen with these other technologies in the past. Today, we are the only company operating driverless trucks on the road and scaling. Not in a pilot, not with a safety driver behind the wheel, but in actual day-in, day-out operation. This is just the beginning. By the end of the year, we expect to have 200 trucks on the road.
Aurora Driver 2 was a huge step forward for us. It allowed us to get our verifiable AI systems out on hardware that was cutting-edge, that had both the price, or cost, and scalability and durability that we needed. It allowed us to deploy the tools and services that will ultimately allow our customers to use this technology in their operations and allow us to scale with them to meet their demand. Trucking is a tough business and it's only getting tougher. Driving is difficult and demanding, and that means that there aren't enough people that want to do this important job. Our policy decisions are actually taking more supply out of the market. People need to rest. That means that these assets have to sit idle or companies have to do complicated logistical things with slip seat driving or team driving.
Driving a truck is dangerous. As a truck driver, you are 10 times as likely to die on the job as the average American. 5,000 people are killed in collisions with heavy trucks every year. This is something we can do something about. Fuel, labor, insurance, these are all costs that are increasing. When a customer integrates the Aurora Driver into their fleet, we will be able to help them drive all of these down simultaneously. The Aurora Driver makes trucking safer. 360-degree perception, validated safety case, millions of tests. This gives us conviction when we put it on the road, it will be the safest, best driver out there. The Aurora Driver never needs to take a day off, and our customers can deploy it where they need it, when they need it, allowing them to meet their customers' demands.
The Aurora Driver doesn't need to rest, allowing trucks to operate 24/7, really creating that opportunity to double utilization. Finally, all of this drives down the cost of labor, fuel, and insurance, again, simultaneously. Taken together, the Aurora Driver is nothing short of transformational for these businesses. These benefits are not hypothetical. We are delivering real-world value with customers today. When you look at the video here on the left, this is the Aurora Driver on I-20. Bombing along at 70 miles an hour, I think. At this moment, this construction worker decides it is time to go have lunch on the other side of the road. They run across, the Aurora Driver sees them, slows down, everybody goes home safe. In our long-haul applications with Werner and McLane, we are seeing trucks running 225,000 mi on an annualized basis.
Again, that is double the normal utilization for these assets. Then finally, in our short-haul application with Detmar, we are doubling the number of trips per day that they are able to operate. Safety is table stakes. Doubling utilization is a fundamental change in the value we can provide. As I mentioned before, we are fully allocated to exit 2026 with 200 driverless trucks on the road operating for customers on our existing routes. For customers to use the Aurora Driver, we need to meet them where they are. That means operating from and to their endpoints. Today, we are live offering driverlessly for Detmar between the Capital Sand mine and between their distribution point. In the coming weeks and months, you will see us begin to operate driverlessly for more customers.
Make no mistake, it is clear that the core value that we provide as part of the Aurora Driver is operating directly between customer endpoints, and that is the product we are building and deploying. Having had the opportunity to work with customers at their endpoints, it has really given us the ability to improve and optimize the Aurora Driver and ensure that it will fit seamlessly into their businesses. As we head into 2027, we are further expanding the network we are going to be supporting and growing into. It took us six years to develop, validate, and deploy for the first lane. It took us six months for the second lane. Earlier this year, we deployed a lane in six weeks. In the not-distant future, it is going to take us days.
That's because the Aurora Driver is now generalized, and every bit of further generalization we make eases the ability to open new lanes. In parallel, we're continuing to develop the tools that allow us to map, validate, and verify the Aurora Driver so we can have confidence it's safe on the road. The combination of these two is not just a linear improvement in how quickly we can open lanes, but an exponential one. Once again, we see this inflection point that means that we will be able to scale and build the business. Before I hand off to Ossa, I want to take a moment to zoom out. Over the long term, automating freight is going to bolster the U.S. economy. The Aurora Driver will make our logistics pipelines safer, more robust, and lower cost.
Lower cost logistics is great. It's going to both help the end consumer, because goods will cost less. We will help fight inflation. But beyond that, through Jevons paradox, lower cost freight will mean that there is more demand for freight, which will grow our customers' businesses. In parallel with that, automating the middle mile long-haul routes we're going to be focusing on will actually help our partners elevate the role of their logistics professionals, allowing them to focus on being the face of the company in short-haul applications, and importantly, allowing them to sleep in their own beds at night. We're going to help transform this industry. We're starting this flywheel in motion today. Of course, it is not going to happen overnight, but the long-term impact of what we're building is profound.
So thank you for being on the journey. Thank you again for taking the time with us today. Hope you find the day informative and useful. Really excited to get you in a truck. And with that, I'm going to hand it over to Ossa, our president. Take it away, Ossa.
Thanks, Chris. As you can see, it is an incredibly exciting time to be at Aurora. I have the pleasure of spending a good portion of my time with our customers, and their enthusiasm is simply palpable. I thought that before I get into the numbers, I might share a few stories from the road, just so you can get a feel for some of the things that I'm hearing. Last week, I had dinner with the CEO of one of our long-standing customers, and we were talking about S-curves and inflection points, sharing many of the same stories that Chris was just telling you about. And he said that the external factors or market forces that are often required for adoption to move from one part of the curve to the next had already arrived for autonomous trucking.
He then went on to say that the decision to adopt AVs was no longer just a smart one, it was now an urgent one. That sentiment was actually echoed with another CEO of ours who we had the opportunity to introduce to Secretary of Transportation Sean Duffy. As he was talking to Secretary Duffy, he said this marked a high point in his career. He had never before seen such technological momentum coupled with regulatory acceleration. It is as if the two were working in lockstep with one another. I will share one story before I move on. A CEO of ours who might just happen to be in the room with us today, was driving on I-20 when up along beside him comes the Aurora Driver hauling his freight and no one behind the wheel, and it was just cool.
He said it marked a new level of excitement for him and for me. I share these stories so you can understand how fun it is to be at Aurora right now. We are in a materially different spot than when I joined four years ago. With that, let us get into a little bit more of the specifics. When customers come to us, what are they asking about? First and foremost, it is always safety. At this point, most of our customers, and frankly even our prospects, have had the opportunity to experience the Aurora Driver firsthand. They know that the Aurora Driver is one of the safest, most capable, most experienced drivers on the road. We did not want to stop there. We engaged third parties to validate how we approach safety. TÜV SÜD came and audited our safety management system.
They found that we were highly proficient, reliable, thorough. Basically, we passed with flying colors. We then went to Edge Case, and they did a first of its kind audit of our driverless safety case. They did a rigorous sampling of our facts and evidence, and they found that we had reached a whole new standard of transparency when it came to autonomous trucking. After safety, money talks. This is probably the most common conversation that I am having today. We took ATRI data and looked at total cost of ownership for a traditional driver versus the Aurora Driver. The savings are staggering, upwards of 20% or a full $0.40 per mi. There is a lot in this slide, but I really want to break it down into three key themes that we keep hearing and talking about. The first is, of course, the driver itself.
In this analysis, we compare it against solo driving, which is actually a conservative view. Team driving might be the more comparable approach, but even with solo driving, you see great savings. After the driver, there is fuel. We are regularly seeing 10% fuel efficiency on the lanes we are hauling today, and we see an opportunity of 15% depending on the lane and type of freight. That is quite notable to our customers, given diesel prices today and where we expect them to be. The third category is insurance. Dave, our CFO, is going to talk in detail about liability and insurance. At this point, I will just say that a safer driver is a cheaper driver to insure. The great thing about these economics for our customers is it affects both sides of a carrier's P&L. What do I mean by that? Let us take the top line.
Our truck is not subject to hour service limitations, and we don't have a driver that needs to get home. That means we can effectively double the utilization and in turn, the revenue, for any given truck. The bottom line, any carrier will tell you that this is a razor-thin margin business. Every penny counts. Based on the analysis I just showed you on the previous page, there is a ton of opportunity to improve the bottom line as well. But let me make this tangible. We did an analysis. We took our lane, Phoenix to Fort Worth. We took a single truck, and we mapped it over the course of a year. You see a doubling of revenue and a six-fold increase in margin. That is an incremental $340,000 in revenue and $160,000 in margin per truck, per year. Did I mention these conversations are fun?
Beyond that, beyond safety, beyond the economics, we also want to meet our customers where they are, and that's where endpoints come in. Opening up endpoints this year has been a great demand and growth driver for us. But beyond just opening the lanes, we want to make sure we're providing the service and support so that our customers can not only start operating with us, but they can scale operations with us. With that, I'd like to take the conversation just a little bit deeper, inviting three of my favorite people, industry experts, and also Aurora customers to the stage. Please help me welcome Matt Detmar from Detmar Logistics, Eric Hildenbrand from McLane, and Daragh Mahon from Werner.
Good. Good to see you.
Thank you, gentlemen. So appreciative of you making the trip. I think we're going to have a really fun conversation here.
I'm going to actually start the conversation with Daragh. For those of you who were here two and a half years ago, you'll remember that Daragh was on this panel with us as well. I'd like you to think back to that moment and then ahead to today. There's a lot that's happened. You've approved driverless operations with Aurora. Tell us a little bit about the last two and a half years and what gave you the conviction to move forward.
Yeah. Thanks for having me first.
Yeah.
I think back to two and a half years ago, it doesn't seem like that long ago. I feel like top of mind back then were three things. First was the tech. Was it there? Was it getting there? Second was safety, and third was reliability. I think I'll put safety first here because in my mind, that's what we eat, sleep every single day at Werner. It's safety, how safe, how do we drive safety into our fleet, how do we make sure that we maintain that, and we wanted to make sure that you guys could prove a safety case. I think back to even prior to that, maybe two years before that, the first conversation I had with Sterling and Chris was about safety.
I think you guys have done an excellent job up to that point, but up to this point, you've proved your safety case, and we feel like you guys are there from a safety perspective. Now, you got to get millions more miles on the road to keep proving it, but I think at this point, millions of miles in, the safety case has been proven, and we're very satisfied with that. Second thing is the tech. I feel really good about the tech. I don't think there's any question anymore. In fact, we talked about this internally all the time. Tech is there. We're not worried about the tech. We have some integration work to do, et cetera, but all of that's relatively easy in comparison to the heavy lift you guys have done to this point. The third part for us is reliability.
When we put these trucks on the road, we have to know they're safe, and we have to know they're reliable. Over the last eight, nine months, whenever it was we approved driverless operations, we've ran a lot of miles with you, but in comparison to what we run on a daily basis, it's not really a lot of miles. The reliability has been there. Again, you've got to run a lot more miles to prove it's there, but we feel really good about where we're at right now. We're still bullish, and I think we're just ready to keep moving forward with you guys.
We've really appreciated walking alongside of you on this journey. We've learned a lot from you guys as well. With that, I want to really turn it to Eric and to Matt. I think Eric, McLane's been working with us for two years. What prompted that decision, and where do you see it going from here?
Yeah, I think for us, first, it starts with safety as well. I think that is probably where everyone in the industry is, and we saw an opportunity to have a potentially very safe lever that we could use to grow. That, for us, was critical. How are we going to manage our growing middle mile logistics and redistribution operations? Drivers are scarce, as I think we all know. We need them desperately for our final mile into our customers. Finding a solution like yours that could help us in that middle mile and deliver the safety, the reliability, and those types of opportunities to put our drivers where we really needed them was why we turned to you.
Great.
Yeah. First of all, thanks for having me. Glad to be here. We have been working with Aurora for about a year now, and there is multiple things that prompted us to look at this and go down this journey. Safety, of course, being a big one. The Permian Basin is an incredibly dangerous place on the roads. A lot of traffic, a lot of truck traffic at night. A lot of people trying to get to these drilling and frack locations to get deliveries on time. Secondly is operational capacity. Not only are we in the trucking business, but we are also in the oil and gas business. So we are in two cyclical industries. The push for innovation and competition is very high.
As we look at the operational capabilities to be able to double our utilization and continue to look at how we can provide our customers a value-driven approach, when you look at both those things, it made the most sense for us to go down this path.
That is great. I remember the first time I visited in Midland, and there is very little infrastructure there, and I think you guys have a slightly different use case than the long haul we typically talk about. Can you talk, it is sort of a 60-mile loop that we run perpetually. Tell the audience a little bit about how that works and why autonomy is helpful in that regard.
Yeah, I will go ahead and kind of explain.
Yeah.
How the oil and gas logistics space works as well.
Yeah.
That could give more clarity.
Great. Yeah.
As we look at our trucks, especially in the sand side, and you look at pipe and a lot of other commodities that move in the Permian Basin and in different oil and gas basins. Sorry, I just lost my train of thought. As you look at that, most of the drivers, especially in the Permian Basin, do not live in the Permian Basin. They are coming from a lot of the southeastern U.S., different parts of Texas, Oklahoma, New Mexico. Almost all the trucks you see are running sleeper cabs. Day cabs have never really broke out into that market because drivers don't like to come live in man camps. They don't like to switch trucks because they don't know which driver was driving that truck before. Was the driver smoking in it? Did the driver not report something on his pre-trip or post-trip?
Anybody who's really come to try to work the day cab slip seat model to run 24/7, it's never worked out very well. We run sleeper cabs. Our drivers work on average, three weeks on, one week off. When you look at that from a utilization perspective, on the monthly, you're getting 75% utilization. Now go into their hours of service. They're driving about 12 hours a day. Okay, now you cut that in half. You also have their 34-hour reset that comes on top of that as well. When you look at the utilization of the tractor, you're looking at sub 40%.
With what we've seen with Aurora, our drivers are averaging two, 2.5 truckloads per day, getting into the five to six range on the 60-mi loop, is fantastic to see from a utilization perspective. You're taking that 40% to greater than 90%, effectively you're almost more than doubling the utilization. On top of that is the trailer utilization. As drivers go into their three-week shift, when they go back home, their trailer goes back in the pool. Because the drivers will stage next to the nearest sand mine that they might be picking up from the next day, they're always attached to that trailer for their shift. Not only are we doubling the utilization of the tractor, we're also doubling the utilization of the trailer.
That's where we see a lot of the use case and the optimization for the uptime and utilization of the asset.
That's fantastic. As we've talked before, it's almost insatiable appetite here because diesel prices and the need for oil is just growing. It's been really fun to watch and partner with you.
What's going on with diesel prices?
We monitor them occasionally. With that, Daragh, you've been in this industry a long time. You've seen a lot of changes, but in particular, I know you've looked at other potential AV trucking partners. What caused you to land on Aurora as differentiated?
There's a couple of things. First of all, again, I go back to the very first conversation I had with Chris and Sterling in a tent, I believe, in a drop yard somewhere.
Aurora Illuminated.
I think that the thing that came across most strongly was safety was the first conversation we had. Then you guys instinctively got the driver problem, right? This is not about replacing drivers. Maybe can I go on about drivers for a bit?
Yeah, please. Yeah.
First of all, the very first job I had after construction when I came to the U.S. 30-something years ago was I got a CDL, and I went driving for one of our competitors, who I don't think are in this room, but I can't mention. I'll get fired. The reality is I am a driver, and I did it for a short period of time, but I understand drivers. This makes me sensitive to this conversation, no question. I still actually go out and drive a little bit today at times. They get me to test-drive trucks every now and again. I have this connection to drivers, and drivers are the lifeblood of our industry. Our founder, one guy, C.L. Werner, with a truck, instilled that message and that just across the entire company, that drivers are the central part of what we do every day.
We have this connection to drivers. They are not just a guy in a truck or a lady in a truck. They are the lifeblood of our company and of our industry. The thing that Aurora intuitively got when we talked to them was this is not about replacing drivers. This is not what this is for. This is about taking the jobs that drivers don't want for us and giving them to a different mode of transportation, making drivers' lives better. Drivers today want to be home as often as possible. They don't want to be on the roads for two or three or four weeks at a time. They want to be home nightly, or at least every other night. We see this as there are places that we find it hard to hire drivers to do certain middle-mile routes that are long, lonely roads.
We think that that's where this really comes in. The way we look at it internally is it's almost like another mode of transport. We do van, we do dedicated, we do intermodal. We do everything, and then driverless becomes another thing that we offer to our customers. It also helps us fill some of the driver shortage that we're experiencing, and have been since there was a driver shortage when I took the job 32 years ago. There is still a driver shortage today, and it's not getting better, it's getting worse. It helps us fill that, but it does not and never will replace drivers. We say this all the time. If you wanted to become a driver today at Werner, we think you can retire as a driver 30, 35 years from now.
Can I echo that? Our choice for Aurora was very similar. It was about the shared values. McLane's been around for 130 years, and we care a lot about our culture, which is very driver-focused as well. I think between that and safety, working with you for two years, we feel that our cultures are similar in a way that's very differentiated from some of the other competitors, and it's why we're very happy to be the partner. For us, as I said, our drivers have to pull into a 7-Eleven parking lot with six other cars there and then take 40,000 pounds off of a truck with a hand cart. They're really critical to us, and they're basically our face to the customer, and we need them to understand this isn't a replacement.
Now you're doing that versus having to hopefully drive very long miles or very repetitive routes. We see it the same, and I think that's what makes our partnership really valuable to us.
I would also add that I think you are probably the only one we spoke to ever who did not think initially they were going to become truckers. Move freight. That was a big deal for us, and I think all of the others have finally realized that being a trucker is tough, it is complicated. It takes years of experience to get there, and I do not think you guys ever felt like that was your role in the industry, which mattered to us as well.
That is actually a great commentary on why we picked Driver-as-a-Service. I think as Chris alluded to, there are certain things we think we can do better than some. We know there is a lot we cannot. Maybe as we talk about Driver-as-a-Service, you all know that is when our carriers own and operate the trucks. We provide the driving service. Tell me how you are thinking about that, why you are excited about that. Maybe Matt, I will start with you, sort of why a transition to DaaS would be an inevitable choice for you.
We have been in business for about 15 years now. We have purchased tractors, we have leased tractors, we have done full maintenance lease programs as well. What makes the most sense for us because of the use of the application, it is on the road, but it is also very vocational as well. Sometimes our miles are 10%, 15% off-road. The maintenance aspect of it is incredibly important to us for the uptime of the asset. As we buy the asset, we like to bring our own maintenance in-house because that really helps us accelerate the uptime as well. Of course, buying trucks also has its own tax benefits as well. There is that case as well.
Being able to handle the maintenance, bring it in-house, and as we look at scaling it, because it is going to be a whole different path to revolutionize, and we are actually really, really excited about that as well. That is one of our biggest drivers going into the DaaS model really.
That is great. Eric, how are you guys thinking about either the DaaS evolution or just the next several years at McLane with autonomy?
Yeah, I think similarly. I think we have a lot of infrastructure built to take care of trucks, whether it is our own or we have partners that do it. I think that having us be able to use our scale to do that while you do the technology and the thing that makes you special probably makes the most sense. There are also tax and other benefits, as you know. I think as we go into the future, we just see so much opportunity, right? I am always on the phone with you telling you about my next great idea. She is very polite, by the way.
They are fun conversations.
I am very excited to kind of expand from the lane we are running now, which has been incredibly successful, to some of these places that we mentioned about. Our drivers really do not want to do and are really critical and have this consistent freight running on this lane, and it is always like who wants to do that overnight trip to wherever, right? We would love to get that expanded as fast as possible under the model.
That is great. I know we want to spend some time getting questions from the audience, but before we do that, could each of you, maybe starting with Daragh, talk me through what does the strategy look like for you over the next three to five years? If you come back in two and a half years from now, what are you going to be saying to the audience then?
Yeah, I think for us, it really is about taking, I do not want to say tentative steps, but we have gotten our toe wet now, let us take the next move. It is about really proving out the safety case, making sure that we feel we are 100% comfortable, but let us make sure that we understand all the nuances of it. It is about the reliability that I talked about. I think it is about really exploring what lanes this is best suited to. We have a massive network countrywide across all 48 states. There are lanes that we already know that this will because we have trouble getting drivers or keeping drivers in those lanes. I think there is lanes that we can do it.
There is certainly the concept of completely utilizing this massive investment in an asset that we have that we do not get enough of today, or we could get more of. I think we want to explore all of those areas: safety, reliability, the tech. Let us see the tech keep moving forward because it is fantastic. How do we get to the point where I think we start to strip the cab of these trucks of their creature comforts and work with you guys on that so that becomes less expensive even. Then I think it is really finally about how do we fit this into our network and how do we make the best use of it.
I think a couple of years from now, hopefully we are much deeper into this and we have got more lanes operating, and we have given our drivers a better life and a better job across the board. I think that is what we would like to be.
Well said. Eric, how do you see it?
What Daragh said. I really agree. I think it is about use cases for us.
Yeah.
It is which different use cases can we use this in the north when it is snowing?
Yeah.
That kind of technology, that keep pushing to find the different use cases so we have more flexibility, because that is what we really need.
Great.
Yeah, for us, of course, we are running in the Permian Basin, and we are looking at areas beyond, but today we are brokering out about 75% of our freight in the Permian Basin. By the end of the year, we will be running about 1,200 or 1,300 loads a day. We are seeing the opportunities continue to grow out there for us. As we are continuing to prove out the safety case, which everything we have seen has been absolutely phenomenal on the Aurora side for the safety aspect.
We do expect to get into the hundreds of trucks over the next couple of years. Again, we are still proving out some of the deliveries closer to the well site, which I know we are looking at doing very soon. As we continue to get there, I do not see anything that is really getting in the way of that.
I do expect us to scale pretty significantly with Aurora.
That is great. We are looking forward to all the expansion opportunities and continue to work closely together. I do want to open it up for questions from the audience. Oh, lots. In the black shirt in the front, please. Oh, there is a, oh, sorry.
Yeah.
Sorry about that.
Thanks, it is Scott Group from Wolfe Research. Thanks for doing this panel. I am guessing you saw the slide earlier showing the total cost of ownership savings. Just curious, each of your perspectives, what you are seeing, and then maybe Daragh, if you think this is best as a middle mile application, does that change in any way the savings opportunity? Then maybe just my last question would be when do you think you transition from brokering trucks to Aurora to using your Aurora as a Driver-as-a-Service model?
On the economics first, we are negotiating, so I am not going to comment. I think if I had to say there are two areas we probably are working on. We are working on right now extensively with Aurora. One is legal, just contract work, which we will get through. It is all fine. The other is economics. I will be quite honest that we have a gap. We have got to figure this out. I think the economics become viable at scale. I mean, really viable at scale where nobody is eating some of the cost. I do not think that is too far ahead of us. I think that sometime in the next few months, we will get to a point where we believe that economically we can make this work. There is work to do there.
The second part of your question, yeah, I think absolutely where we run changes the economics for us, changes the viability. For us, honestly, we believe the long-haul route is the best. The middle mile where we are running 500, 600, 700, 2,000 mi, whatever that happens to be. The longer we can run, the more utilization we can get out of the asset, the better fuel economy we get once we are running those long hauls. When we can run it 22 hours a day or 20 hours a day out of 24, all of that matters. Shorter length of haul does not allow us to do quite that much. So I think the economics will get there, but we believe we have a gap that we have got to work through. I think we will get through it.
I think it comes at scale, then I think, yes, definitely the lanes we operate in change the financial model.
Yeah. Ravi? You are here. Sorry. Yeah. Hi.
Hi. George Gianarikas from Canaccord Genuity. I would like to ask about reliability, which you mentioned a few times. Is that something that just proves itself out over time? You just need more miles under your belt. And then second, what particular parts are you focused on from a reliability perspective? Is it the autonomous kit? Is it the redundant systems in the truck? Thank you.
Maybe Eric can take that as the lawyer on the panel.
I can't believe you outed me in front of all these people. I think that the reliability seems really good right now. We hope to just keep proving that out. We're seeing basically 100%, I don't want to say, I can't do it, 99 whatever percent uptime on stuff that's controllable. There's the stuff that isn't controllable. Some of it's on our side too. Getting the freight where it needs to be. I think that proves itself out over time. We just need more miles, more trucks running, and you're going to see. But I have an expectation that we'll just see that continue to be very positive. I forgot the last part of your question, sorry.
Are there any particular parts you're focused on?
Oh, I think we're looking at the whole truck and the whole system. I do think it's kind of early to say which parts of the Aurora Driver system we have to watch. We know what parts of the truck to watch, and I have a feeling if we can get everyone aligned, that because the truck needs maintenance in such a very specific way, particularly if you can get the utilization that we all want, that fixing or updating or all those things shouldn't actually lead to more downtime. Because we should be able to do that while we're doing maintenances that you have to do on the physical asset.
Thank you. Ravi Shanker, Morgan Stanley. A question for Daragh Mahon and Eric Hildenbrand. You were both very passionate about how you feel about drivers and how you're both going to be a very driver-first organization for a long time to come. Have you had this conversation with your drivers? Is this message resonating? If not, kind of how long or what do you think you need to build that trust? Thank you.
For us, yes. We have been very intentional over the last five or six years, where we talk about driverless operations at every opportunity. Just last week at National Truck Driver Appreciation Week, it is a topic that comes up. We get a lot of questions on it. We have been very open with drivers, not hiding anything.
The message is the same. We want drivers to understand that if you want to take a job at Werner today, if you are at Werner today, you are probably going to retire at Werner today, if that is what you choose to do. Because we keep reiterating that message that today, depending on what numbers you take, whether it is from the American Trucking Associations or the Department of Transportation, there is a shortage of anywhere between 100,000 and 200,000 drivers right now. That only grows, even in Aurora and all of the driverless operations trucks.
Most bullish case, we still need a million drivers in the next decade to come into the market. We have been very open about that. We have told drivers what we plan to do. We have told them that we are doing this slowly, methodically, and with a lot of intention and thought. But they know, and we have certainly talked to them.
You cannot hide things from drivers. It is the first thing you learn working for drivers. They have the best communication network you have ever seen in your life. We have been completely open about it. I think we have a little advantage over Werner there, which is, again, these guys have to deliver into small stores using hand trucks or lift gates. So they sort of do not, I do not think, have a fear because they know that right now that job is very safe and will be, we think, for the foreseeable future. But yes, we have been very open about it, and we were ready for pushback.
The drivers always ask great questions. But I think they actually get a little excited about it too, and they get the safety point as well. Good drivers want more good drivers on the road, and they sort of view it that way.
Yeah, and I very much resound what he is saying on that. We conduct a weekly driver call with our drivers. I get on and do a fully unscripted Q&A. It does come up often, but as these guys mentioned, drivers are the lifeblood of what we do. None of us would be even sitting on the stage if we did not have good, capable drivers to help us drive our business forward. So, what Eric Hildenbrand said is you do see good drivers, they get really excited about it because sometimes your good drivers are running next to drivers that might not be the best drivers. So when they know that you are building a team that is adding more reliability to the roads, more safety to the roads, you do see the more professional drivers be very excited about this technology.
Hi, David Vernon from Bernstein. Thanks for participating in the panel. Eric and Daragh, I guess it sounds like you're talking about a future where the driverless technology is doing the work your drivers don't want to do and is somehow complementary with your existing drivers. But if you're not taking the driver out of the cab, how does that affect the economics of the implementation, the first and last mile cost? When I talk to trucking companies about adoption of this stuff, there seems to be less certainty around what the actual total cost could be because of some of those business model issues around the first mile, the last mile, things like that. Could you talk a little bit about how the economics or your view of the economics of these technologies are affected by the fact that you might still also have a driver in the cab?
I think it's different types of freight. I think you said it maybe in a good way, which is like you use intermodal when that makes sense, you use drivers when that makes sense, you use Aurora when that makes sense. I think if you looked at the whole company and how it'd affect McLane's economics, I don't know that I want to get into that. What I do see is a tremendous amount of value that can be driven TCO-wise on millions upon millions of miles McLane runs today. Because of redistribution and other things, I have to move full truckloads of freight from A to B before that gets broken up or even gets unhooked and then goes into the city and makes all the deliveries. If I can get savings on those millions of miles.
Sure, I'm not getting savings on these other millions of miles. I still need a driver, but that's very beneficial to my economics as a whole.
Yeah. Same answer. If we have a driverless operation, we wouldn't have a driver in a truck between Houston and Atlanta, for example. If we're doing that 100 times a week or whatever the number is, significant savings. But I think we have to just think about the market's growing. Retail is growing. Retail is our biggest customer. So we're going to be moving more freight, not less freight over time, need more drivers are hard to get. So there is a hole that it fills for us just by nature of the fact that we're growing. Then I think it's also important to realize there's other costs besides just the cost of the driver and the truck. To get a driver and a truck, we have to recruit the driver, we have to train the driver. All of those costs are significant, as you guys know.
I think when you add up all the economics and when we get to a point where we're happy with our negotiation, and then when this gets to scale, and again, I don't know if the Aurora people will disagree with me here, but I think that's the real inflection point for us is when they get to scale, and all of a sudden those prices start to drop, not just for the truck and the hardware in the truck, but the tech and everything that Aurora does becomes more economically viable for us too. I think the driver out of certain routes helps us. It doesn't eliminate. We still want drivers talking to our customers, so it doesn't eliminate that. We have a similar type to you. We have retail stores that we unload and walk pallets into as well.
We still have those drivers at the final mile points, but middle mile for sure.
I am sure we have to transition. I do want to note, on the stuff we run with Aurora, we don't want a driver in the truck. There is no driver in the truck, just to make sure that was clear.
Great. I think we have time for one more for this panel. Good.
Thank you. Good morning, everyone. Andres Sheppard from Cantor Fitzgerald. First and foremost, thank you for the Aurora team for putting this day together, and congrats on all the great success. I think the team has done a great job articulating the value proposition, cost savings, higher efficiency, increased safety, of course. I am curious if we can maybe better understand your aha moment. How long did it take for you to visualize it, understand it, implement it? How long did it take you to convert into a customer? I am curious if you can maybe talk about what kind of industry reaction do you expect? Are you the consensus? Are you the outlier? I am just curious to understand what really drove it in home for you. Obviously, we talked about the benefits, but was there something specific, and how long did that process take?
Maybe we will start with Matt, who is the most recent adopter. Yeah.
Yeah. The aha moment is doubling utilization for us really. Again, mentioning the safety case. We are driving down Interstate 20. I do not know if you have ever driven down Interstate 20 between Midland and Odessa, but it is a disaster, and it is disaster all the time. It is always under construction. There are always accidents. There is always traffic. Since we have deployed this technology with Aurora, which we started at the beginning of this year, it has almost been an absolute perfect safety case. So that is an aha moment for us, the ability to double utilization. Customer feedback has been great. There has been a lot of excitement about it, especially right now. You have seen capacity shortages in the broader freight market over the last 8-12 months. Historically in oil and gas, we run about 8-12 months behind.
We are really starting to see a driver crunch right now, a big capacity crunch, which is developing even more excitement from our customers in the industry. So those two moments really being the aha moment for us. The industry, again, is very excited to look at this. When you look at oil and gas, they have been under significant pressure to get their operation under control. No one wants boom or bust anymore. Everybody wants to see significant returns. If they can base load their operations with guaranteed capacity, helping to fix their costs overall long-term brings excitement. All in all, those have been really the biggest positive things for us in what we have seen.
So multiple aha moments.
For me, I do not know if it was aha, but when our 20-plus-year safety person walked into my office and said, "Okay, you can go meet Aurora," because they had validated that the safety worked. I said, "Can I go now?" "No, you can not go yet." That was probably a great. I do not know if it was aha, but it was a real fun moment in my office because I thought, okay, we are really going to do this. That was pretty exciting.
Yeah, I think there is multiple for me. As a tech guy, as a geek, it was the first time I sat in the Aurora truck. You go, "Holy crap. This really works." That has been almost. It has been five years ago. For me, that was just exciting. It was one of the reasons I came to Werner was just this concept of autonomous trucking and how it would get there. Loved what I saw. That was a big aha moment for me. This actually really works. Then I think the second one was, there was probably multiple throughout the. The safety case always felt like the hardest one because everybody was struggling with it, and Aurora took a totally different approach to it.
When they finally. We sat down two years ago and went through it in great detail, there was multiple aha moments during that period where, okay, this really is safe. This can work. The tech is there. Safety is there. Now we have just got to work everything else. But I think in general, it has been a series of aha moments. It has been fantastic to be involved. Lawyers do not have aha moments.
I was going to say, Eric, I thought it was the first time we met. Well, that is a great note to bring us home on. Thank you all for your partnership and your conviction, and we are really excited to be on this ride with you all. Thank you.
Thanks, Ossa.
Thank you.
I would like to turn it over to our Senior Vice President of Hardware Products, Sandor Barna, who will talk to us about our multi-platform path to scale. Please welcome Sandor.
Thanks, Ossa. When it comes to physical AI, it is hard to be more physical than autonomous trucking. As with all physical AI, the actual hardware is critical. Frankly, hardware is cool. Our hardware roadmap and multi-vehicle approach is both deliberate and highly differentiated. We are the only autonomous trucking company with such a deep partnership ecosystem and such a carefully thought-out roadmap, and that is what will enable true industrial scale. Our second-generation hardware kit is currently being produced at our contract manufacturer, Fabrinet. Fabrinet is a top-tier CM based in Thailand, with over $4.5 billion of annual revenue and 3 million sq ft of manufacturing space. This is our 8,500 sq ft clean room in Chonburi, Thailand. The Aurora kit was designed with three key objectives. First, meaningful unit cost reduction on the order of 50%. Secondly, increased reliability to 1 million mi.
1 million mi drives down the cost per mile. In addition, this has been tested to demanding OEM specifications, with a goal of enabling line side production. It has the headroom to scale up to 1,500 trucks. It is already on the road, powering driverless operations today, and we are ramping to build over 50 kits per week this year. Once the kit is finished, it heads to our outfit partner, Roush, in Livonia, Michigan. Roush has five decades of excellence in vehicle outfit. They have extensive experience with autonomous vehicles. We have a 20,000 sq ft Aurora-dedicated facility with a multi-station assembly line. The first trucks are already off the line and here in Texas today. We are establishing capacity to 20 per week beginning next month. 20 per week is 1,000 trucks in a year.
Now let's take a moment to look at what's happening in Michigan right now. That's pretty fun. I've got to be honest, I think it was showing off a little bit. Our second-generation hardware is also being installed line side onto the Volvo VNL Autonomous truck at Volvo's New River Valley, Virginia manufacturing facility. Volvo Autonomous Solutions recently announced that we will have driverless operations in the first quarter of next year, and these trucks will be powered by the Aurora Driver. Volvo expects to exit 2027 with 300 driverless trucks, paving the way for industrial scaling in 2028. The second-generation hardware supports our initial scaling to over 1,000 trucks. But now let's talk about our long-term strategy. We're partnering with AUMOVIO to develop our third-generation kit. This will be industrialized automotive-grade hardware enabling tens of thousands of autonomous trucks over time.
It is an industry first Hardware-as-a-Service structure, so our hardware cost is paid per mile. That means no new upfront CapEx for our customers, and it enables an asset-light model for Aurora to support our SaaS-like gross margin strategy objective. For AUMOVIO, this partnership unlocks a brand-new recurring revenue stream. Incentives are mutually aligned across the entire ecosystem, and everyone benefits the more miles our trucks drive. The start of production is expected in the second half of next year with material economic benefit in 2028 and beyond. Our third-generation hardware kit will power all of our truck platforms, whether it's outfit with the International LT Series at Roush or line side install at Volvo and PACCAR. This multi-platform approach allows for customers' different preferences and perspectives, and provides scalable supply. This will position us to meet this market-defining opportunity.
Now I'd like to welcome a panel of partners to join me here on stage to talk about the ecosystem we're building together. Please help me welcome Jeremy McClain from AUMOVIO, Noelle Onstad from PACCAR, Brad Keselowski from Roush, and Sasko Cuklev from Volvo Autonomous Solutions. Thanks, everyone, for joining me. Let's start with you, Brad, from Roush. Roush has dedicated a facility to upfitting Aurora's second-generation hardware onto trucks. Why is this program so important for Roush?
Well, we're a product development company, we're a contract manufacturer, and we have products of our own. We very much understand the urgency and importance for our customers to get their products to market as fast as possible and of quality. So it's near and dear to our hearts. Also we have, as a company, 50 years of reputation for delivering for our customers. We plan to uphold that. So it's a very important program to us.
Great. Sasko, so together with Volvo Trucks, VAS has made remarkable progress towards line site integration of the Aurora Driver. What makes the depth and velocity of the Aurora relationship unique to you as you execute on your commercial autonomous roadmap?
It has been a long journey, right?
Yes.
I think the first time we met you guys was 2017 or something in Mountain View. I think at that point in time you were 20, 30 persons or so. You are a completely different company today. One year later, we integrated for the first time the Aurora Driver with a Volvo truck, a European FH. We tested that, worked out in a fantastic way. To be honest, we felt that we, in a way, clicked on many, many aspects with Aurora already from the start. Safety, we have talked about safety already. We both share the same values when it comes to safety, super important, the most important topic for autonomous. We aligned on responsibility splits. One thing that stood out as well was that you really wanted to, or Aurora really wanted to do this together with us and not as the other players.
We met all at that point in time, where they said more, "Give us the truck and then we will take care of the rest." It is a long-lasting partnership that we value a lot. Just to emphasize on the partnership, this is not about Aurora providing a driver to us, and then we integrate it, or we provide a truck with some CAD files to Aurora to integrate. It is really a joint program. We work side by side in developing and integrating and commercializing the solution. I think we have built something very, very strong here.
Thanks. Jeremy, tier 1 industrialization is super important to get to scale. We are incredibly proud of the partnership we have built together. Why is this program central to AUMOVIO's growth strategy, and how do you view this opportunity?
Yeah. First of all, we are a very new but a very old company. Having spun off from Continental just last year, a year ago, we have got a long legacy and it is a very long story, to Sasko's point, building safety into products at scale. What do we do well? We build safe, reliable products at large scale. We have been working in ADAS and autonomy for many, many years. This is roughly three decades. We put our first automotive radar sensor into the market in 1999. I have been working personally on autonomy for many years, and it was about finding the right point in time. What is the inflection point when it makes sense to bring that expertise, building safe, reliable products at large commercial industrial scale to autonomy?
Exactly that moment is here and the partnership has been running for a couple of years. I think the interesting thing about that partnership was exactly seeing the need to be able to prove that the technology was ready and was safe and could be deployed, and in parallel, making sure that we are ready for that industrial scale. That is what we have been doing, and the third-generation hardware is back there. It is exactly that inflection point that makes it exciting for us, because that is the next growth opportunity. We typically grow with volumes. There you get a kind of saturated market if you look at ADAS. It is about how do we unlock that next opportunity? You unlock that next opportunity by bringing autonomy to industrial scale, so we are quite excited about it.
Nice. Turning to PACCAR, Noelle, the freight industry, as we have heard earlier, continues to face a lot of challenges. What are you hearing from your customers about the prospect and value of autonomy?
Yeah, thank you. Thank you, Sandor, for having me on the panel. The customer panel that was up here earlier did a great job answering this, but I think the big change has been from a focus on the technology to discussions now about the business efficiency opportunities. It has really pivoted with our customers, there is a huge conversation about just the challenges the industry is facing with the diesel prices, with labor shortages, they want solutions that are safe, reliable, making sure that they are efficient in their business, because really the businesses there and are successful are making money. So they are shifting to how can they incorporate this technology into their businesses to be more efficient.
At the end of the day, it is not just about our customers that want this technology, but it is what will it do for them to deliver to their customers more reliably, freight on time, meeting their commitments at a scalable option. So they are very excited about where this can go for them and how they can incorporate it into their business.
Okay. All right. Let us go back to you, Brad. A lot of people may not be that familiar with the Upfit model. Could you walk through how the Upfit process works and how we can get to 20 trucks per week?
Absolutely. We saw a quick two minute version of it in the drone footage, which was great. There are eight overall steps that happen. There are many stations in the manufacturing process that you could see. Step 1 is we receive the vehicle, we inspect it, make sure it meets the specifications, and it is good to proceed to the next step. Once we get it into our facility, we wash all the trucks.
Our certified technicians then start actually de-contenting the trucks, taking the parts off, getting it ready for the hardware system. The third step is the fabrication. So that is when we are drilling holes, we are cutting holes into the body panels in the truck to actually prep it for the hardware install. Simultaneously, we are doing offline sub-assembly. A lot of the components that go into the truck still require some form of assembly that we do in our facility offline.
It is a lot more efficient and repeatable to do it that way. Then once those sub-assemblies are complete, they move to the main line, which you saw, and they get installed into the vehicle. Really the fifth step is the installing of the actual hardware, the autonomous equipment. That is done through our MES system, our manufacturing system. We have digital work instructions.
Everything talks to our system, torques, traceability, as those get installed. The sixth step is throughout the entire process, which is in-process quality. We do not just wait until the end to check to see if we are building what we are supposed to be building. That is built in across the entire process from when that truck arrives at our facility all the way to the end. Throughout that process, we are checking and making sure we are doing what we are supposed to be doing as we go.
Seventh part of the process is the bring-up part. That is when everything is really installed. That is really to bring the truck up, test the software, it is really a functionality check of everything that was just installed, and make sure that is operating properly. Then the final step is the final quality sign-off. Our quality team will go through and check to make sure that everything is completed that is supposed to be completed, there are no open items that need to be addressed, and that the vehicle is 100% ready to go to Aurora.
How are we doing on getting to 20 a week?
Right on plan. The facility is and has been up and running. We are fully staffed. All of our equipment, tools, fixtures, those are all validated and up and running. As of last week, we actually just brought on our second shift operation. We are building trucks 16 hours per day right now.
Nice. Let's go back to you, Noelle. PACCAR is known for being very disciplined and having a very responsible approach to new vehicle introduction. As we work together towards defining the third-generation integration, how do you see the autonomy-ready truck fitting into the advancement of Peterbilt and Kenworth in general?
Mm-hmm. Yeah, we do have a very disciplined approach to new product development at PACCAR. Whether it's new powertrains, it's advanced safety systems, it's connectivity, and now autonomy, we follow very similar processes. That's why it's been a joy, really, to partner with Aurora to come up with a way that we can have this new technology in line on our Peterbilt and Kenworth products. The reason we're so disciplined in how we approach new technology and integrating new technology into our trucks is because we need to make sure that we are delivering safe, proven, reliable products to our customers, as well as the infrastructure to support them once they leave our facilities. It's really the approach that we take in the partnership with Aurora that we're following.
Great. Sasko, as you go out and engage with major fleet operations about the Volvo Autonomous Solutions offering, how would you characterize that discussion?
I would say that the discussions we have had, or we have with the fleets are going extremely well. We have had an approach where we said early on, "Let's not go out and talk to everyone because it creates a lot of expectations. Let's instead join forces with a handful or 5- 10 customers that we work extremely tight with in a partnership approach." We have done that and built the ecosystem, developed the solutions together with them, and now when we have announced that we are launching in Q1 next year, Q1 2027, we are launching. By the end of next year, we should have 300 plus trucks in operation. Then we have opened up and talked to more customers, and the interest is huge.
And have in mind when we talk about those numbers, these are purpose-built trucks for autonomy, with the redundancies in place for the safety critical systems, with the Aurora Driver integrated in our manufacturing facility in New River Valley, which has been an important thing for us. Our manufacturing facility in New River Valley, that is really the flagship of the Volvo Group, where all our Volvo trucks are coming out and we have already built the first batches of trucks coming out from the factory. As I said, 300 by the end of the year. So huge interest. I can share some examples. Last week, we had a discussion with one of the big fleets that are super interested because today they have problems. They have to turn down business. They do not have drivers.
They really see autonomous as the alternative here so that they can capture that business. Another one of the big ones, they said, "We want half of the 300 directly." So, I would say that the confidence is high that we should be able to allocate those trucks.
Great. So back to you, Jeremy. Under this Hardware-as-a-Service structure, AUMOVIO's economics are actually tied to operation of the fleet, operation of the kits. How does this mutual incentive shape prioritization and execution at AUMOVIO?
Yeah. It is really about aligned incentives, and I think that is the important thing. We heard about it in the earlier panel as well. Our traditional business scales with volume, maybe scales with software-defined vehicles, and so on. But if you look beyond that and where we stand here with autonomy, it is all about as a service business. And we see it scaling in a very different way. It, of course, scales with miles. The value of autonomous trucks especially, comes with safe, reliable, and quality service. And of course, when we build a model around that, we find other unit economics that help us to recuperate those investments, which is, of course, important.
It also aligns the incentives to make sure that when we design together with Aurora, that third-generation kit, we design it in a way that it is going to operate over the full life cycle of the vehicle. Then we measure the performance against that target by the way it actually operates in the field. We get incentivized to design it properly, and as well to make sure that it operates properly in exactly that way. So it is driving a different kind of business model for us.
We did that on purpose with the vision that that is one of the next frontiers as far as business models in our tier 1 space as a service business where we build that into the hardware, and we maintain it over its life together with the fleet partners, together with the OEs in a way that generates new value streams and generates value for the end customer. The end customer in this case is not a consumer, but the fleets.
Great. Thanks, Jeremy. Actually, at this point, we would like to open it up to questions. We have got quite a few. We are going to grab a mic.
Thank you.
All right.
Thanks so much. This is Colin Rusch from Oppenheimer. This is really for the OEMs. I am curious about how you are thinking about single source risk and how you manage that. You talked about discussing with multiple partners. As this industry starts to take shape and scale, how are you thinking about that risk, managing that, and preparing for the potential for multiple partners from a technology perspective?
Our approach at PACCAR is really to develop the truck as a system. We call it an AVP, an autonomous vehicle platform, so that when we can in the future, as the need grows, we can bring on more partners. Aurora is obviously the one we are working with right now, and then it is really a platform approach for us so that we can grow in the future.
Yeah. Similar answer from my side. We are also developing our truck, our autonomy-enabled truck with a platform approach that we call COST. This is not something that is specific for autonomy. It is how we develop all our products. COST stands for common architecture and shared technology. We develop the autonomous truck according to that. We already have two partners that we are doing this with. It is the same product, the same platform for several partners. With this approach, we should be able to bring the autonomy also to different regions because we have other trucks in Europe, for example. It will make it faster and easier for us to also scale in other regions, on other truck specifications, on different brands. We also have both Mack Trucks and Renault Trucks within the Volvo Group.
Thanks. Ravi Shanker, Morgan Stanley. A question to PACCAR and Volvo. Again, we can debate the timing and the slope of the curve, but it is very clear that the trucking industry is changing. I would love to know your internal discussions on what the truck market looks like 10, 20 years from now. Who is going to be buying your trucks? How many OEMs, et cetera? Also, with both AUMOVIO and Aurora moving to a per mile fee to monetize the product, is there any thought on you guys doing that for the truck itself and how that might potentially change your business model over time? Thank you.
I can start. Let us start with the business models. We have been clear from the start that, and this is based on the dialogues we have had with the customers, actually, where in the early stages, they told us more or less, "Hey, guys, we work tightly together with you, and we would like to continue to do that. We want you to take the full responsibility." We have said that the starting phase, we will offer this as a Transportation-as-a-Service. Basically, like what everyone else, I would say, is doing at this stage. Have in mind that we have been operating in a commercial setup almost for two years. The first phase is Transportation-as-a-Service.
Then we have always said that we are extremely humble to different types of models, and we are already now looking into additional business models where it is more that the customer operates and so on. We will, of course, follow what will happen in the industry, and listen in. We would be stupid if we go the other direction when everyone wants something else. That is on the business model. Then the first question was-
What the industry looks like in 10 years.
The industry looks like in 10 years. I think we had a capital markets day, when was that? Mid this year where Nils, our president, talked about that in five years or so. Every 10th truck will be an autonomous truck. That is what we see.
Yeah, on the PACCAR side, I would definitely echo the Volvo view, is that we are really partnering with our customers, and we want to deliver what our customers want. Whether that is truck as a service or per mile kind of approach from a PACCAR lease sense, for example, or if they just want to purchase the vehicle and then work with Aurora separately. We are really open to making sure that our business model is what customers want and deserve. We are not going to push that on them. We have been very open in talking with our customers around that. On where we see the industry in 10 years, other than PACCAR obviously growing market share rapidly. Thank you for laughing. It is going to happen, but thank you for recognizing the joke.
Other than that, it is really we are going to make sure that we are ready to deliver what customers need and want and grow their business with us, because it is really all about partnering with our customers.
Mark Delaney from Goldman Sachs. Thank you very much for the time and doing the presentation and panel. A question for Noelle Onstad and Sasko Cuklev. You both mentioned you are developing your autonomous trucks as platforms and would consider other technology partners or maybe already are in some instances. Can you just talk a little bit more around your experiences specifically with Aurora and how you see Aurora relative to some of the other competitors? Why are you working with Aurora now, and just other things that may stand out from Aurora compared to other technology providers? Thanks.
Yeah, as we mentioned, PACCAR's approach is very disciplined in product development, so we partnered with Aurora because we see that they also are very disciplined, focused on safety, and that's our number one focus is we want to make sure that we have safety at the forefront, proven technology, reliability for our customers, and that's why we partnered with Aurora and are working with them to make sure that we have Kenworth and Peterbilt trucks that will support the Aurora Driver. I forgot your second part of your question, sorry.
Just how Aurora compares to other technology partners.
Well, we picked Aurora for a reason, and I think that's probably enough said for me on the PACCAR side.
Yeah, I touched upon it in one of the previous questions where we have felt that already from the start, that we share a lot of the values that are important for us, in terms of safety and other things. So, we click there. Then again, a partnership is it's like a marriage. So you need to find your ways, and we have done that and built something strong. We have not always agreed on everything, but we've sorted it out. So it's really a joint program, joint initiative. We work hand in hand. Now when we're talking about integrating Aurora into our manufacturing facility, we do that together with Aurora, so that we really take care of everything. It's not always easy discussions, but in a true partnership way, we solve it.
It is, again, a true partnership that we feel is extremely strong and we value a lot. Then we have other partners. I do not want to go in here and start to compare. But yeah, we value Aurora a lot in our partnership.
Okay, I think we have time for one more.
Thank you. Ryan Sigdahl, Craig-Hallum. Maybe this is for everyone, just from a capacity standpoint. Brad, it is great to hear you guys are on track, but we always want more and look to the future. As you think about capacity expansion, if demand is there, can you ramp faster? Can you go quicker? Then maybe for Sasko and Noelle, from an OEM assembly line standpoint, talk about retooling. If the demand is there, again, what the process is, how quickly you can actually ramp volume capacity should there be the demand quicker. Thank you.
Yeah, as mentioned, we are completely on plan right now to reach the planned volume that we are talking about. Our team feels, after building, obviously, we have many vehicles off the line. We feel very confident, very good in meeting or exceeding that with our current line. Then we always have the option to expand capacity if needed. We are very flexible on the Roush side. If that demand were to increase, we can pretty rapidly grow that operation.
Yeah, on the PACCAR side, the program that we're working on right now is to really tool and have in place the online installation of AVP. That's really for as the customer demand grows, we are ready to achieve that.
Yeah, on our side, the whole program has been driven with scaling and industrialization in mind from day one. This is not about producing 10 trucks and go to hit the pilots. It has been with scaling and industrialization in mind. We have, in the plan, as we said, 300+ trucks from our factory next year, and then we want to be able to go to 1,000. I don't see that that will be the problem from a manufacturing perspective to meet the demand.
I just would address it from the AUMOVIO side as well. There's a plant in New Braunfels, Texas, four and a half hours south of here, which is exactly being ramped up to be prepared to produce that third generation hardware that's in the back of the room, at scale, to be ready to hit those volume demands. I would encourage you all to go in the back and take a look at that third generation of hardware. That is now design complete. It's in design validation, and the product validation runs will happen from the plant in Texas here in the very near future. It's really exciting to see that we ramp towards scale. That's really an exciting milestone for the industry, for the business, for the partnerships, and something we're excited about.
Okay. Well, thank you everyone. Really appreciate your time here. Okay, now I'll hand it over to our CFO, David Maday. He'll have more to share with you, but starting with a conversation about insurance. Thank you.
Where's the clicker? Oh, right there. Thanks. All right. That was super exciting. Thanks to the panelists for participating. I think when Sasko mentioned we've had our share of arguments, that could be with me, so that's interesting. We've talked a lot from the beginning about safety and the importance of safety. We're getting strong receptivity from customers, from our OEM partners, and as well from our insurance providers. I know it's top of mind to many of you how the insurance and liability pieces can work. We are going to be joined, if this all works out brilliantly, via a video link from London. We're going to be joined by Chris Moore. Chris is the Chief Underwriting Officer at Apollo ibott. This is a syndicate for Lloyd's of London. They are one of the recognized leaders in complex, specialized technology and mobility risk.
They were the first in really leading the way with companies like Uber, Airbnb, and of course, Aurora. Chris has a background in mathematics and machine learning engineering, so he's a whole lot smarter than me. He also understands how to look at potential liability through a distinct quantitative lens. Hopefully here we're going to be able to be joined by Chris. Hey, Chris. Can you hear us?
Can you hear me? Yes.
Excellent. I'm going to sit by myself since he's on video. Thanks again for joining from London through his busy schedule. I'm going to have a couple of prepared questions but we got about 15 minutes. I want to enable you guys to ask a lot of questions. This is for you. I'm going to start though, Chris, if you can just tell us a little bit about Apollo and what excites you about the AV industry and then Aurora in particular.
Yeah, I think autonomy is probably the biggest opportunity facing the insurance world today. It also comes with risk for us as an industry in some ways as well. If you think about the U.S. auto insurance market alone, that is about a half a trillion dollars. If I can see the answers in the data from people like Aurora and other partners in this space, you are talking about frequency reductions in accidents that could be as high as 99.5%. So the premium that can be taken out of that market is phenomenal. We are talking about shrinking from $500 billion to $50 billion.
Now, that is the risk side of it. But for us, leaning into that, because I do not work for Progressive, and I do not work for State Farm or GEICO, so I would be worried for those guys. But I am leaning into really deep embedded partnerships with companies like Aurora.
If I can capture a large share of that $50 billion results in market, I am not saying I need more than 50, but I will be happy with 20. So it is a huge opportunity for us. Not to mention all the other benefits that come with autonomy. Saving lives, getting goods and services to people in America and globally at a more affordable rate. There are loads of those benefits. But yeah, for us as an insurer, it is so nice to have a partnership where the insurance costs for those companies, specifically trucking. Trucking is a very hard insurance risk. When you get those partnerships where they reel it in, where insurance is really valuable, that is a really great place for me to work.
Excellent. Great. Now, I know it looks like from here, the reception is a little bit good as everybody. I know we are going to be continuing to fine-tune it. So if we have some disruptions, we might go to no video and just the audio. All right. Let us talk a little bit about from an underwriter's perspective and you being the chief underwriter, how do you evaluate and differentiate risk across different autonomous technology stacks in the marketplace today?
Yeah, sure. So we have been actually insuring autonomous vehicles for over a decade. So we have built up quite a large dataset. We have also built a risk assessment framework for autonomy. So we look at what is the use case of the technology? Is it trucking? Is it robotaxi? Is it autonomous drones? We will have that as a key consideration. Then we will look at the ODD. So where are we doing the operation themselves? A truck driving up and down Arizona is very different to a robotaxi operating on the streets of San Francisco. Then we will start looking at the experience, which again, autonomy has performed very well compared to the human drivers that we insure. So there is not a huge amount of developed claims data, but we are certainly building up that dataset.
The final piece that we look at when we are trying to assess autonomy is all about culture, and it is super important. If you asked me five, six, seven years ago, that may not have been part of our risk assessment framework, but it is so important about how we collect data, how we are going to look at those expansion, what are we doing from a safety perspective? How are we partnering? How do we get very purposeful in how we want to position the insurance? Because the U.S. legal system can be a very challenging one from an insurance perspective, and trucking has had a lot of nuclear verdicts. We want to make sure that if there is ever an occurrence or an accident involving an autonomous vehicle, that we are very well prepared for that.
We know exactly how we are going to defend the technology and potentially in front of a jury in a court of law in America.
All right. Great. I will ask one more question that is definitely top of mind. We talked a lot about cost structure for the trucking industry. Insurance is obviously a big cost item. When you look at the economics of underwriting autonomous trucks compared to traditional fleets, how do you see the pricing evolving over time as these platforms continue to scale?
Yeah. I think we split it up into two key considerations, which would be frequency and severity. If I deal with frequency, so the number of accidents, the number of collisions, we already have enough developed data that we can really lean into that. No, we are not at the 90%-95% that certain mathematical reports have been released, but every year we are seeing that frequency reduction, and we can lean in and price that in. The unknown part is the severity, because we have not seen a large number of claims involving autonomous vehicles hitting courtroom steps. So that is some of the uncertainty. But I would say if I am insuring a human driver, it is a very linear relationship to the risk. So Dave, if you do 10,000 mi in one year, my premium is X.
If you do 20,000 mi, I am pretty much going to charge you double because it is a linear relationship. It is not that for autonomy, because every mile driven is slightly better than the mile previously. And so you are starting to see a different curve. With every mile driven, you are seeing the price of insurance lower. Now, again, that is subject to the U.S. legal system, and we have seen some difficulties in human driving, like the Montgomery Supreme Court case. For me, leaning into that regulation and legal side is where I am spending a lot of my effort at the moment. But undoubtedly, in this space, rates for insurance are going down every year.
That is excellent. All right. Well, why don't I, instead of me asking a whole bunch of questions, I am sure you have a lot, and you usually ask me these questions, so might as well ask somebody that knows more about it than I do. Let's just take some questions from the crowd. Oh, no, I let Stacy pick. Sorry.
Thanks. Mike Latimore, Northland Capital. I guess, in terms of today's pricing, what is the price of a virtual driver versus a human driver with similar frequency severity?
Yeah, good question. I would say it differs by state, but we are probably slightly above where a human driver is purely on that unknown with the severity. That being said, I think that changes rapidly, and by quite a large amount, to be honest. So I think what we are kind of seeing at the moment when I was talking about those rate decreases, you are probably looking in the 15%-25% range every year, and that doesn't take very long for that to compound and be a really significant factor.
All right. Other questions?
Thanks. Ravi Shanker from Morgan Stanley. That was a very interesting point about you guys already preparing to defend the technology in court against any potential accident. Can you just unpack that a little bit more? You said you've been working with autonomy for over 10 years. When you go to prior technologies, what does that first precedent case look like? How easy does that get over time? Thank you.
Great question. I think my concern is slightly alleviated, and we can thank our friends in the robotaxi space for this. I think Uber now making these partnerships with having autonomy on the platform and having human driving on the platform has unlocked a defense for me in that scenario that I haven't been able to rely on before. Please indulge me. What I would say now is if an autonomous vehicle got into an accident, I imagine a plaintiff's lawyer that's looking for a big nuclear verdict and make big headlines would say something along the lines of, "You put a dangerous vehicle on the roads, and we need to punish you because you put our community at risk. They scare people," and that's how it works.
That we would try to defend that, and what I mean by being proactive is working with the engineering team to say, "How would we defend the claim? How would we use the sensors? How would we use the cameras and what technology to show what actually happened here and how we've tried to be the safest possible company in operation?" What's brilliant now is because Uber have a direct comparison, if that happens and they sue Uber, they'll say, "Well, my human drivers get into an accident once every 50,000 mi, and the average severity of the injury is X." Now they can have a direct comparison to say, "But the autonomous vehicles operating are 90% less likely to get into an accident, and because of the braking speeds and the reaction times, the severity is 50%. So I've actually connected you with the safest form of transportation.
You can't come for me." My concern then for the rest of the market is the plaintiff, if the human driver gets into an accident, they go, "You just told me that the safest form of transportation was the AV for the money anyway." I think the cost of human driving is just going to keep increasing, and the cost of autonomy will keep lowering, and then eventually you'll get to an inflection point that says it's just too expensive to allow a human to drive.
His best part was when it broke up a little bit, but I think you guys got the message there.
Well said.
All right. Oh, sorry, David.
Hey. David Vernon with Bernstein. As you think about the technology being in the cab and having all the information about causal liability being one source of it being cheaper for autonomy versus the actual driver being safer, how do you think of what the technology brings to the table, just in terms of being able to say, "Look, it wasn't the truck's fault, it was somebody else's fault"? How do you separate the benefit you get out of that from knowing how much safer the driver actually is?
Yeah. It's that dampening severity point again. I look at that as it's a great discovery tool. I just need to have a playbook that I can use it in a court of law where it's as impactful as it should be. There's lots of different jurisdictions. Some jurisdictions may say you're confusing the jury, where it's inadmissible. That's why we have to prepare that playbook. What I love about the partnership with Helen and her risk team at Aurora is they're having those active conversations before anything's happened. That's someone that has proven to me that they have a safety culture, because you don't really talk about things like this unless it really means something and you see the importance.
I do think what you touched on with the product itself is there's an argument to say that autonomy has no place to sit in auto liability insurance. There is an argument to say it's a product liability risk, and that comes with a completely different framework for litigation and defense and pricing. It's a really fascinating time that we're seeing companies like Aurora partner with insurers that say, "Hey, we're going to create an insurance product that's not just fit for us, it's fit for our clients and all our stakeholders that we work with.
All right. Got time for a few more. Ryan?
Thanks. Ryan Sigdahl, Craig-Hallum. How do you think about underwriting different AV technology providers and the due diligence as you think about building that defense case, the safety case, everything that goes in it, and is there a differentiation between Aurora and others? Maybe talk through that process, and thanks.
Yeah, it's a great question. I think when I was talking about that risk assessment framework, that culture pillar is really key, and that's kind of where I think your question's sitting. We won't insure anyone in autonomy unless we meet them in person, have numerous conversations. It's quite a long process. There's not 20, 50, 100 insurers queuing up to write autonomous vehicle risk, which I find staggering, but that's just me. We have those conversations, and that fleshes out whether we have appetite. It is a true partnership, the way that we look at things, and there has to be a lot of transparency and a willingness to share information both ways.
Yes, I'm receiving information on the exposures that Aurora are operating in, but I'm also sending back trends, and I'm sending back, "This is how our pricing model works," to the Aurora team so they can see and balance those financials. I do think the financials will be so compelling to empower the future of autonomy. For me, I'd find out straight away whether a company is going to be a good partner for us, and we are super selective in who we're going to insure.
Okay, I think we have time for one more question in this section.
Oh, hey. It is Chris Pierce from Needham. I just want to tie together, you mentioned prices coming down per year, but then you also mentioned you are surprised by the lack of competition in the space. As you get these miles, are you building in a more competitive space and more competitive bidding in that 15%-20%, or could we see it come down even more as there are more companies like yourselves that want to insure these drive miles?
Yeah, I think it is a great question. It could come down further than that, again, if the market does dip in and provide capacity. What I hope for Apollo is obviously we have long-term partnerships, and then there would be an open conversation about what those rates do. We tend to have a renewal retention on portfolio in the high 90s. I think it is a really difficult one. If you ask me why are not more insurers jumping into this risk, it is because it is kind of cannibalizing a lot of their business. They have 50 years of developed loss data. Why would they go into an industry and support an industry that is going to potentially completely disrupt the cash cow that they have had for a long time? I am not in the personal line space, so it does not really affect me.
This is why I can lean into this very exciting space and try to lead it. The capacity will come. You cannot ignore the experience. That is what I love about, again, the data and the transparency. I am not going to insure someone for 10 years unless I show them how they perform, and we have a very sensible discussion on what margin looks like for both parties.
I would just add for that, we treat insurance partners, we use the word partners a lot, they are just like every one of our other partners, whether they are on the customer panel, they are the OEMs or the tier ones. We are all partners in this together. We will not be successful if we try, especially early on, just to bid out to the cheapest person. We really believe in select folks that believe in the technology. Apollo is a leader in this technology. They believe it. They have a substantial. They carry a lot of insurance for us, and they will in the future. But of course, as new competition comes in, it keeps them honest, but more than anything else, they are looking to actually increase their level of exposure into us. So I think it is a healthy relationship that we have. All right.
Chris, I am going to let you go. Thank you very much for the session and shedding some light into insurance because you are a lot better at it than I am.
Pleasure. Appreciate it.
Okay, we are almost to the end run. Almost time to drive trucks, which is going to be way more exciting. But I got to tell you, this has been a great event. I am so excited. The ability to not only tell our story and talk about this commercial inflection point, but also listen to so many of the people that are helping us get here, and their excitement and this aligned vision is what keeps us all excited. But I know we are also excited about the financials, and you all want to hear about them. We will talk a little bit about our path to scaled economics and long-term value creation. Of course, our press release went out this morning, so a lot of you saw the punchline at the end. But we will build you into it. Okay.
It is important to start off, again, with the market opportunity, and we talked about this multiple times. The U.S. is a trillion-dollar market based on 200 billion vehicle miles traveled, and this is a place where we can have an opportunity for unmatched value creation with autonomy. We really can help influence this on a lot of levels.
For us, the market has very attractive unit economics, and as Chris, Ossa, and a lot of the panel members talked about, it resonates. They can see the value in the technology. In the near term, our focus, as you can see, is a 60 billion vehicle mile traveled market by 2028. For context, we are about $4 billion today. So we are going to be operating in a much bigger place. Even at a very reasonable single-digit percentage market share, that represents more than $1 billion of revenue opportunity for Aurora.
It is a great opportunity. It is also important for us to remember what we have accomplished thus far. We always want to reflect on our roadmap. Are we delivering to the things that we said? You have heard multiple times that we are the only company operating driverless on public roads today in trucking. We have surpassed 500,000 mi, and this is not just demos. These are commercial miles with all the commercial pains and challenges that trucking experiences today.
We are fully allocated to exit 2026 with 200 driverless trucks, which would represent a $80 million revenue run rate on our Transportation-as-a-Service fleet. As a reminder, that is where we own and operate the trucks on behalf of our customers. As Sandor outlined, we are building the capacity with our upfit center at Roush to achieve more than 1,000 trucks. We are really on a good path. This is what we are delivering today.
We really believe we are hitting this commercial inflection point at the end of this year, and it really is going to be the key to our growth trajectory in 2027 and beyond. For us, in 2027, DaaS is our business model. That is our Driver-as-a-Service business model. That is the one that you heard a lot of earlier today. That is the one that the customers would like to own and operate the trucks. They like that because they can deliver real value. They can maximize the value of the Transportation-as-a-Service or the Driver-as-a-Service business. By the end of 2027, we will have also launched our AUMOVIO Hardware-as-a-Service partnership. This is our asset-light DaaS framework. This is the one that allows us to build tens of thousands of trucks, and this is where we believe we will surpass 30,000 trucks in 2030.
Now, to really understand the financial trajectory, let us just make sure we understand the two models. Our Transportation-as-a-Service model is our customer adoption model. That is where we are today. We have talked about the ownership. We expect to get roughly 200,000 mi per truck on average per year. Pricing is roughly $2 per mi, including fuel surcharge, on average. It varies by lane. It varies by customer. This is a great customer adoption model, but it is also capital intensive, and we are going to our Driver-as-a-Service asset-light business model. We are going to limit this fleet to roughly 500 trucks. We then shift to our Driver-as-a-Service business model. That is where all the growth will occur.
We also expect because the actual customers are owning and operating the trucks, they are able to maximize through their network optimization, the miles per truck, and we expect the actual miles per truck to increase closer to 250,000 mi per year. The revenue drops down from the $2 to $0.85 plus per mi. That is natural, right? As Ossa pointed out in the cost comparison, now we are just focused on replacing the driver cost and not the overall cost. This still allows us to have high margin potential for the business, and we think it is a great strategy. We know that is what the customers want to have, and that is what we have been building for the last several years. Now, if I look out to the next four years, we project rapid top-line revenue growth and margin expansion.
Starting in 2027, we expect to end the year at over 1,000 trucks. That will represent roughly $200 million in revenue for the year. We also expect to achieve positive gross profit. Our target to achieve positive gross profit on a run rate basis is in the first half of 2027. This target slightly moved from where we were talking about the end of this year, and that is really just reflecting a slightly slower fleet ramp. All the revenue enablers and all the cost reduction enablers, including the labor and support scale efficiencies, delivering to customer endpoints, and even really getting the full benefit of our hardware set are in place. We are excited. We are on a path to achieve positive gross margin. If you look at 2028, we again believe that we will achieve positive free cash flow on a run rate basis in 2028.
We expect to achieve this with roughly 7,500 trucks operating on the road. If you look into the future, and again, our targets going into the future, we really do expect rapid growth. We will have the full benefit of everything that we have put in, our DaaS business model, our AUMOVIO partnership, being able to operate everywhere. We expect $5 billion of revenue, and we expect to exceed 60% gross margin, really on our pathway to get to 70% gross margin. With this rapid revenue expansion, we do expect that we would deliver compelling return on operating expenses and capital efficiency. That is really important. We have spent a significant amount of money getting to this leadership position, putting ourselves in a position to be at this commercial inflection point.
We are going to continue to spend money to accelerate our lead and grow the top-line revenue. From an R&D perspective, we expect to continue R&D roughly where it is at today. I would say adjusting for inflation, of course, but we are going to continue to invest in our leadership position. If you look in the future, that means that an R&D dollar is generating $7 revenue in the future. It is only going to get better. We really think this is going to get to, I think, 12%-14% is really industry leading at that point. We are well on our path to get there. If you look at capital expenditures, for 2027, we expect to spend about $185 million. This number is largely comprised of two factors.
Number one, building out the rest of our Transportation-as-a-Service fleet, so buying more trucks and getting them on the roads. The second thing is buying the remainder of the second-generation hardware kits that we are going to put on the road before we go to the AUMOVIO business model. If you look at 2028, we expect our capital expenditures to drop to below $50 million, and on a go-forward basis, CapEx as a percentage of revenue is going to be less than 1%. We will be fully in our Driver-as-a-Service business model. The last thing is SG&A. We are going to increase spending to achieve the commercial growth that we are targeting. We are establishing a target of roughly 7% SG&A as a percent of revenue out in 2030, and we will continue to monitor that each and every year.
This is really set up so that we can scale the business in the long term. 30,000 trucks is a lot of trucks. You need a lot of support to be able to do that. All right. Let's close it out where Chris began. We are extremely excited about the industrialization of the Aurora Driver. If you look at 2030, we expect to be operating in a $150 billion VMT market. Remember, the total market is about 200. We expect to be operating almost everywhere. We are going to deliver more than $5 billion in revenue, more than 30,000 trucks, exceeding 60% gross margin, and having high return on our R&D spend. This is just the U.S. trucking market. We are going to be able to take opportunities to go to global markets and additional adjacent applications. This is just the beginning for us.
This is a compelling business just for the U.S. Honestly, Aurora is driving the commercial inflection point in autonomous freight today. Okay? We are exceptionally well positioned to maintain our leadership position. I can say that. I know you asked others what they thought. We really truly feel we are in a leadership position. We are taking advantage of all of the investments we have made, and we are really at that commercial inflection point. We also believe that we are well positioned to generate tremendous long-term shareholder value for our investors. With that, I will ask Chris to come back on stage and bring us home.
Cool. Thank you, Dave. If you talk to Dave, the next few years are going to be tremendous. If you talk to Ossa Fisher, the next few years are going to be fun. If you talk to me, they are going to be exciting. One thing is clear, the pieces are in place. We are on the road operating day in, day out, commercially, driverlessly for our customers. The thing to remember is trucking in the U.S. is only the beginning. You can count on the fact that we are going to take this show on the road and go and deploy this to global markets and expand in that direction. It is also important to remember that the foundational investments we have made in building safety critical verifiable AI sets us up to go and take on other exciting commercial applications.
You can think about wherever a vehicle needs to be guided to operate safely through the world, that is a place where you are going to find Aurora. Transforming trucking is really just the first step in delivering the benefit to self-driving technology safely, quickly, and broadly. With that, I want to say thank you again for spending your valuable time with us today. I am going to welcome the executive team up on the stage here so we can answer the questions that you have not got to ask yet today. Maybe you guys want to join us up here. I do not know why I am clapping. Felt the moment.
Because you are so damn excited.
I am so damn excited. It is going to be fun and tremendous.
Just go to the right and you can get up to the left.
Great, thanks. Ravi Shanker, Morgan Stanley. Thank you for the presentation today. Maybe one for David, one for Chris. David, can you give us some level of detail on how you got to the 30,000 target number, kind of top-down, bottom-up, et cetera? Chris, I liked your "and one more thing" video at the very end with exploring the other end markets. When do you start thinking about those? When do we start thinking about those? Thank you.
Okay. It's a little bit of both, right? We have a top-down vision of where we expect to go to, and that's largely based on macroeconomic factors. 30,000 trucks does sound like a lot to some people, but when you look at the market, there's over two million trucks operating on the roads today. They build over 250,000 trucks every year. I know it seems like a lot, and it will be. There's a lot of research out there that says the autonomous trucking market is going to be 10% of the market. There's a little bit of that. But we always have to balance that with our plan. We look at both the supply side.
What do we have relative to our hardware set, working with the AUMOVIO team, working with our OEM partners, where are we going to build, how much can we build? What do we think is the desired and reasonable capacity each year? We balance that with Ossa and the business development side to understand where we think the customers are going. Again, we look at it from a VMT perspective as well. If you think about 150 billion VMT that you can operate in, 30,000 trucks, and you do the math, it's not that much of a market share gain. I think we're appropriately conservative in some ways, but we want to also put a vision out there that we really are at the inflection point. Some of the biggest challenges we've faced in the past is the technology going to be there?
Will customers be ready? Can we insure it? Can we build it? All those things have been answered, and now we just are going to go and execute it and deliver the value. That doesn't mean that there's not a ton of hard work that has to be done, but the doubt about whether it could happen, I think has been erased. Now it's about taking advantage and getting to that true commercial inflection.
Yeah. It's going to be a lot of fun.
It's exciting, right?
Tremendous.
And on the other domains. First of all, as Dave has made the case, if all we do is deliver the market leader in trucking, I'm going to be pretty damn satisfied. I think for everyone here, we'll make a hell of a lot of money for you and your clients, and we'd be proud to do that. At the same time, we're ambitious. We have a lead, and we have a capability that no one else can replicate right now. We are starting in a very lightweight way to assess where's the right opportunity to go and apply this technology yet. This is not taking our eye off the ball on the core business because we have to really get that established and build it, and shame on me and shame on us if we let that slip.
There's no effort to take the foot off the gas in any way in that space. But we are starting to think, okay, the gen 3 hardware is in validation at this point, right? Where do we want to go apply next? Where are the markets? What are the opportunities? We're starting to think about it. You'll start to see more over the next couple of years is my expectation.
Thank you. Ryan Sigdahl, Craig-Hallum. Dave, maybe first, gross margin 60% in 2030 or previously at 70% in 2028. Maybe talk through the puts takes on the change there. Chris, just on the last point, are those all AI creations or is any of that actually real products that were shown in the depictions at the end?
That today is AI creations. Yes.
Thank you.
Yeah. I think when we did it two years ago, I think our target is still to get to about 70%. I think we are going to be above 60%. We put a little bit of a hedge in there in terms of that, just understanding the business a little bit stronger than what we did before. I would say, though, that it is probably reasonably conservative as well. I thought it was appropriate to set it at that mark. There is no reason we cannot deliver outstanding gross margins and reach 70%. There are some uncertainties out there of things that we have to do. Frankly, there has been a lot of supply chain headwinds as well that we are accounting for and addressing. So, it is just appropriately conservative, I think, at this point.
By the way, somebody asked that question earlier, and at 60% at $5 billion, our free cash flow looks awesome, or tremendous, I guess.
Mark Delaney with Goldman. Well, first, in terms of the exciting and fun, I am very interested and excited to do the actually on-road demos this afternoon. Remembering a couple of years ago, your last Investor Day was on your closed course in Pittsburgh. Really exciting to see the progress and actually get to participate in that. Thanks to all of you for allowing us to do that later. Quick question for Dave on the financials. Maybe talk a little bit more on the 2027 topics you are starting to lay out. You talked about exiting this year at 200 trucks, over 1,000. Should we think about a back-half-weighted 2027 when AUMOVIO comes on or more linear?
In terms of some of the CapEx numbers you shared, are those numbers that you think maybe you can offset if you find financing and partnership options, or is that a pretty firm expectation at this point around cash out from Aurora next year? Thanks.
Two really good questions. I think on the first, it will be a little bit more linear next year. We really have the installed capacity in place to build 1,000 trucks, 20 per week. I think this year it was really about getting to that capacity, and that is why it was back-end loaded. We are just going to build upon that over time. I think the numbers are going to consistently get higher. It is going to be more of a linear trend than back-weighted. With regard to the CapEx, you are exactly right. We did identify it as CapEx. We have looked at the cost of equity, the cost of capital. We have looked at various financing options. Right now, I think I am leaning more towards just purchasing them and calling it CapEx. That doesn't mean we are not evaluating other options as well.
It is just, I think that is probably the more likely option at this point in time.
George Gianarikas here from Canaccord Genuity. Just a couple of clarifying questions for Dave, and I am sorry if I missed some of this stuff, but I just found out my water heater broke, so I am a little bit thrown off. Maybe on 2027 guidance, you mentioned you would be exiting with 1,000 trucks. I am assuming that is both DaaS and TaaS. You have capped TaaS, I think, at 500, and Volvo just told us they will be at 300, so I am assuming that is an inclusive number that may include other OEM trucks as well. Is that an accurate conclusion?
So, we are going to be greater than 1,000. This is our targets. We will give you your guidance like we would normally do after an earnings, after the Q4 earnings. Yes, we will be over 1,000 trucks. It is a combination of our international upfit, which can be both TaaS and DaaS, as well as it will be Volvo. In Volvo's case, we would consider them more DaaS because they are owning and operating it. We are not doing the owning and operating. Anything that was in Volvo, we would probably consider under the DaaS. We will definitely have more DaaS trucks operating than TaaS trucks by the end of next year, or at least that is our expectation.
Just a couple more clarifying. I am sorry.
That is why we are here.
The 4%, I think, SG&A guidance long term.
7%.
Okay, good. The 7x the R&D efficiency, what does that mean? Can we help us translate?
Yeah. For every R&D dollar, we are going to be generating at least $7 in revenue. We have done some benchmarking, and some of the very best are at about 12. A lot of people talk about our R&D investments and what we are making. We firmly believe this is what separates us from everybody else. Everybody talks about doing a tech stack and the things, but there is a lot to this, building that amazing architecture and back building our Aurora services platform to make sure we operate. We want to continue to invest in that.
The way we look at it is everybody says, "Yeah, it is really high today versus the revenue." Sure, in four short years, we are going to be generating $7 per every R&D dollar, and beyond that, we are going to be achieving $12, $14 per every R&D dollar, and that is actually some of the benchmarking that some of the industry leaders.
Okay, and then last clarifying point. On the 2030 guidance, it implies at least a run rate, I think, of close to 7 billion mi, and your revenue guidance is for 5 billion. I know these are sort of approximations, but that is about $0.75. Is there maybe room for upside to that because you talked about $0.80?
There is a little bit of upside, for sure.
Thank you.
Right here. Good afternoon, everyone. Andres Sheppard from Cantor Fitzgerald again. First of all, once again, congratulations on all the great accomplishments and successes. I guess first clarifying question is, Dave, can we say we are now in the walking phase? We have gone from the crawling phase to the. Or are we in a light jog? I am just curious to explain kind of where we are.
I think we're in the tremendous phase.
We've certainly stretched, and we're starting to jog, I think.
Yeah.
I guess my other two quick questions are, the first is I wonder if you can maybe expand on what are some other use cases and verticals, perhaps industries that we haven't talked about today that you could see this being a great solution for, whether it's military defense, whether it's international markets. Curious to get your take there. Then maybe the second part of the question, I was just wondering if you could maybe give us an update on your relationship with or partnership with Uber.
Yep.
There's been some conversations there, so just curious if you could maybe give us their perspective. Thank you very much.
Yeah. First, let's talk international and other markets. When we look at it globally, the U.S. is clearly, outside of China, the largest possible market, so we want to make sure we win here. China, we think of as basically off limits for as far out as I can imagine. Next, places like Japan and Korea are particularly interesting because the cost of labor is high. With the population dynamics, the need and the amount of logistics, or freight logistics to have them moved by truck is large, so they seem like very interesting markets. They are also long lead time markets, right? There is a lot of engagement. You have to have partners. The good news is, in both of those countries, we have long-standing relationships, whether it is with Toyota or Hyundai, that may ultimately be helpful in building entry there.
The Gulf region is also interesting, right? They want to be technology leading. They have a large non-immigrant workforce. The challenge there is one of labor cost. As a capital allocator, do I want to put a truck on the road in the U.S. where I can generate this much revenue, or do I put it on the road in Saudi or somewhere where I can generate a fraction of that? We have conversations globally about what are the incentives and how would you align interests so that that becomes something that meets their objectives and meets our economic objectives for deploying the technology. So those are kind of globally. Adjacent markets, there is clearly opportunities in nearby logistics, so moving from Class 8 to Class 6, 7 trucks. Glad that worked. Going to be a legend.
Yeah, we see that as just a very clear, obvious next step. Robotaxi is interesting, but it is a much different and, I think, harder business. There is a lot more capital investment that you have to make to scale, and we really like the idea of working with partners who have those capital and have already made that investment, and we can leverage it. We will continue to pay attention to robotaxi and figure out when the right entry point is there. Off-highway mining, ultra trucks, that is another place where there is already significant capital that has been invested, and we can help those companies accelerate and improve their economics. As I said, what we have built is a competence to deliver safety-critical physical AI systems. Anywhere something drives, anywhere something moves, we can go help with that.
We just need to be disciplined in when we start to invest in that. On the Uber relationship, Uber is great. Daragh was a board member, continue to have tight relationship with him. They have a business objective that they have been clear about, which is creating competition in the AV robotaxi space. They have been clear for some time about the need to take capital and recirculate it from us into funding that kind of defensive measure for their business. We have seen this coming. I actually think it is a great thing. This was a large, concentrated position from a holder that had no intent to hold it long-term. It has been an overhang in the stock.
the fact that those shares have recirculated to folks who we expect will want to hold it for the long term, I think is a really good thing for us as shareholders, so I'm actually pretty excited about it.
Great. It's Itay Michaeli from TD Cowen over here. Thanks for hosting the event. Just two questions. Chris, you mentioned opening lanes going from six weeks to six days. Curious when you think you can hit that, what the significance is maybe for costs when you get there.
Yeah.
Then how transferable is that rate of change to international markets? Then one for Dave, just on the 60% gross margins, is that all assumed to be on the third-generation hardware, and could a fourth generation down the road present some upside to that? Thank you.
You want to go first?
Yeah. I will talk first about the lane expansion. This has been our expectation. This has been our thesis, as we continue to improve the generalization, there is fewer new things to learn, so the Aurora Driver obviously more rapidly deploys. We do not believe in this one-shot YOLO and hope kind of strategy. We actually want to know when we put it there, it would not just get lucky today, that every time it is on that route, it is going to be safe. There is an investment we make in verifying and validating the systems for these new lanes. That is something that did not matter initially, but we have been accelerating that work, and I think it is less about cost and more about flexibility and the ability to go and be adaptive and responsive to customer need.
When a customer comes to us and says, "I really want to operate between two places that currently are not in our network," it goes from, "Okay, we need to plan ahead," to allowing our sales team to say, "Yeah, we will be there next week." That flexibility and ultimately the compounding and network effects that will come from this interconnected network of routes will really, again, compound into the value that we provide to customers, is my expectation.
Relative to the 60% to 70%, the short answer is yeah, I still think there are a lot of ways that we will get to 70%. I still think that is going to be our target. If we work with our AUMOVIO partners and Jeremy back there, we are already talking about, all right, what changes would we make on the hardware that could be really productive as well as more cost-effective? I think there is a lot of things also, like four years from now, the cost of the human driver is going to be substantially more than it is today. I think the cost of insuring AV trucks is going to be substantially lower for our customers, and I think that is going to create more value. I think there is a ton of opportunities there.
We certainly believe that 70% is the right target, and it might just take a year longer to get there.
Awesome. I think we have time for two quick questions. We have them right here.
Is that a comment there on how long we are taking to answer?
Hi. Mike Latimore, Northland Capital. On the economics you showed, the economics, is it materially different between long-haul and, say, more regional use cases? When you think about your mix, what do you think about the mix of customers doing long-haul versus maybe more regional use cases?
Yeah.
Yeah. Long-haul has definitely been the most popular thesis that people come to us on, but as you heard from Matt, the short-haul use case is tremendous and sees value near instantaneous. We have customers in both camps that are growing very quickly. I would say the growth in each segment, now that they are better understood, is equal, but we just have a much larger pipeline on the long-haul.
In terms of improving the economics, our cost basis for serving them, I don't think there's a material difference to us, whether it's short-haul or long-haul. We really think about it as amortized on a per mile basis, so not particularly.
Ken Hoexter from Bank of America. Thank you for the presentation and the rides a little later on. Maybe just talk a little bit about your lane expansion. You've talked about the speed you can do it moving north, going into. I mean, we haven't really hit on-
Yeah.
Weather and different kind of lanes and your expansion speed and where you want to target. Maybe throw in also regulation in terms of-
Yeah.
As you do that, are you now free to go everywhere? Thanks.
Yeah. So, one question turns into like five. That was well played, sir. Well played.
You gotta answer faster.
Yeah. I just gotta. Okay. It's all good. On the lane expansion side, yes. From a build the map and verify, validate, it doesn't really matter whether we're in Texas or Arkansas or Minnesota, right? The competence for building those maps, that's not it, and verifying, validating them doesn't really matter across them. On the dimension of weather, today we're operating in a variety of weather conditions, including good rain. We'll continue to push that. It's really incremental work. By the end of this year, we expect to be operating in light snow. When we start to move to snow, this is one of the things that I, by the way, about 15 years ago, I created a bugaboo about snow, and I wish I could go back and reset that conversation.
But if you think about what does snow mean, there's kind of three ways it affects the Aurora Driver. One is there's stuff in the air, which makes it harder to see. In certain conditions, the ground can have a lower friction, which means it's harder to control a vehicle. At least you have to be less dynamic. And the third is that the world looks different. The first of those, we already have them, right? We drive through dust. We drive through rain. We need to do the work to make sure there's nothing surprising. But this is not really a thing that we worry about. The low friction part of this, again, this is an incremental step of validation where we'll go through and check that the control system doesn't do anything goofy when you get to particularly low degree of traction on the road.
And then make sure that we've encoded the behavior so that we can respond when we identify those things. When the world looks a little bit different, again, this is something that's just incremental work. It's not a big deal. The last part of the question was around regulation. As Ossa's anecdote went into, we're seeing an incredible amount of progress on that front. More action at the federal level than I've seen in, I don't know, 20 years of working this. That's exciting to see. We're also seeing a lot of enthusiasm from the states we operate in, too. It's going to be a while where we continue to operate with 50 different regulatory regimes. Even with the enthusiasm we see for BUILD 250.
Yeah.
Which is the legislation, basically, the Surface Transportation Reauthorization Act, which is putting in place a framework for regulation for our vehicles. That's going to take a while to come into action. But we have a team that has put a lot of effort into building relationships and trust with the federal and state regulators, and so we feel very comfortable at our ability to unlock key economic elements of expansion across the U.S. It'll take effort, but we feel good about it. Okay. I took too long with that last question. I blame Ken. With that, thank you all. I know this was a big commitment of time. Hopefully, you found it useful. Hopefully, you come away with the enthusiasm and energy and excitement that we have from this.
For those of you here in Dallas, I am psyched to go get you in a driverless truck operating on the freeway here. It is something that just does not get boring. So, glad to have you with us, glad to have you come do that. With that, let's roll. Thank you.