Peak Car by 2035: The $1.2 Trillion Autonomous Vehicle Industry Takes Shape

Executive Summary

In this episode of The Road to Autonomy podcast, Grayson Brulte sat down with Martyn Briggs, Director, Thematic Investing Strategy, Bank of America to discuss the rapidly commercializing autonomous vehicle industry. During the episode, Grayson and Martyn dive into a recent Bank of America report that explores the huge economic potential, technological breakthroughs, and regulatory shifts shaping the future of autonomy.

Key The Road to Autonomy Episode Questions Answered

Key The Road to Autonomy Topics & Timestamps

[00:46] Rationale for the Bank of America Report

The Bank of America Won’t You Drive My Car report was created due to several converging factors: technology breakthroughs, huge economic potential, demographic factors like driver shortages, and more favorable regulation. The main catalyst, however, is the advancement of AI, which is making autonomous vehicles (AVs) more generalizable and cheaper to deploy.

[01:29] AI and Technology Breakthroughs

Key technological advancements discussed include simulation, which reduces the need for slow and expensive on-road testing; large language models for understanding driving decisions; falling sensor costs, such as lidar prices halving last year; and neural networks that learn to drive from data instead of relying on hard-coded software and maps. This collective progress is described as “gen AI on wheels”.

[03:00] Real-World Progress

Autonomous vehicles are no longer a “moonshot”. Four commercial AV operators have driven 190 million miles, which is more than a return trip from Earth to Venus, demonstrating rapid scaling. The technology is now global, with AVs operating in Asia, Europe, Australia, and soon on multiple continents.

[04:36] The $1.2 Trillion TAM

The projected $1.2 trillion total addressable market (TAM) for autonomous vehicles by 2040 is not limited to passenger cars. While vehicles make up $700 billion of this, the total is nearly doubled by including other sectors like ports, logistics, mining, agriculture, trucking, and delivery, which may be faster to automate with a clearer ROI. This TAM represents potential revenues for the autonomous vehicle technology itself (hardware and software).

[06:50] Serving New Demographics

Autonomous vehicles can extend mobility to new demographics, including older individuals who can no longer drive and younger people who may not get a driver’s license. This increases the addressable market beyond those with a license, but requires highly reliable Level 4 and higher autonomy systems that do not require human takeover.

[08:44] The “Great Unlock”

The “great true unlock” for autonomy is the ability to remove the steering wheel and pedals from a vehicle. This would allow for a change in the vehicle’s form factor, design, and capacity, leading to a huge opportunity for new uses and business models. However, it requires overcoming significant regulatory hurdles that differ by region, making scaling difficult.

[10:45] UK Regulatory and Public Sentiment

The UK government aims to allow fully autonomous vehicles on public roads by 2027, with some commercial activities starting that year. While there’s excitement in industry circles, public sentiment has some hesitation regarding safety and potential job disruption for groups like taxi drivers. However, there is a growing awareness campaign and a sense of “FOMO” from cities that have seen AVs operate successfully elsewhere.

[13:35] Driver Shortages

Driver shortages are a critical need that AVs can address. For example, the average age of a taxi driver in Japan is 60, and in London, the number of taxi drivers has been declining so severely that a study from Transport for London (TfL) suggests there could be none in 20 years. This decline is due to factors like the rise of ride-hailing platforms, the challenging “Knowledge” test, and the demanding nature of the job.

[17:15] Manufacturing and Scaling Autonomous Vehicles

Unlocking the vehicle constraint requires large-scale manufacturing, and an ecosystem of partners is emerging to build vehicles or handle fleet management. There is a significant role for contract manufacturers like Magna, as well as for OEMs who may want to build bespoke vehicles for specific cities, such as a “Big Apple Mobile” or “Big Ben Mobile”.

[20:10] Licensing and Business Models

Nuro and Wayve are exploring partnerships with manufacturers, such as the Nuro-Lucid deal for 10,000 to 20,000 vehicles, which may accelerate the market. The long-term vision for many tech companies is to license their autonomy software in a model similar to Android, where they receive a fee without having to own and operate the assets themselves. This could lead to new asset classes and investment opportunities.

[27:50] The Maturing Market

The AV market is transitioning from a “moonshot” to a commercial reality, with companies now showing what their technology can do in real-world situations. This is leading to more investment and a growing list of publicly traded companies, such as the recent IPO of Kodiak Robotics, a sign of things to come.

[30:00] Productivity and Health Benefits

Beyond just reducing labor costs, autonomy offers significant benefits in sectors like mining and agriculture. These benefits include a 30% increase in productivity, reduced tire wear, and enhanced safety by allowing humans to avoid dangerous areas. Autonomy can also free up workers to do other jobs and can extend operations beyond normal working hours.

[35:10] Global Scale of Autonomous Vehicles and Trucks

There are over 200 operational autonomous vehicles and trucks running globally, including roughly 60 commercial vehicles and 120 in robotaxis and shuttles. This is a larger number than many people think and provides a testbed for future commercial activities, which could lead to significant scaling if even 10% of them come to market in the next few years.

[39:00] Reducing Taxi Costs

By removing the driver, autonomous vehicles can reduce taxi costs by as much as 52% per mile. This cost reduction is not just about labor; it also includes optimizing operations and lowering the cost of autonomous vehicle technology itself. Passing on even a portion of these savings to consumers could significantly increase the number of miles traveled and make autonomy a viable alternative to car ownership.

[41:40] “Peak Car by 2035”

The “Peak Car by 2035” thesis suggests that while the number of cars on the road may decrease by 20% by 2050, those remaining cars will drive 50% more miles. This is based on the idea that autonomous mobility-as-a-service could reduce the need for private car ownership. The total market for miles traveled could actually increase due to economic growth, demographic shifts, and the lower cost per mile enabled by autonomous vehicles.

[43:55] Funding and Commercialization

From 2010 to 2024, over $200 billion was invested in more than 600 autonomous vehicle companies. While funding decreased between 2021 and 2023 due to consolidation, it has seen a resurgence in 2024, concentrating on companies that are ready to commercialize and scale operations. The money is now needed for building out infrastructure rather than just developing the tech.

[46:50] Investor Interest

Institutional investors are now actively engaging with the autonomous vehicle market, with inbound requests signaling a shift in interest. They are moving beyond simply looking at the top companies and are focusing on technology approaches (e.g., rules-based vs. end-to-end), global expansion potential, and geopolitics.

[54:40] The Geopolitics of Autonomy

China is rapidly advancing in the AV race, with a focus shifting from electric vehicles to autonomous vehicles. It is projected that 84% of Chinese cars will have some self-driving capability by 2030, with a notable portion being Level 2 and 3 systems. Chinese robotaxi companies are also expanding internationally, with announcements in several markets outside of China. The cost of autonomous vehicle tech is coming down, with one company announcing a next-gen robotaxi for $40,000.

[59:00] The “ChatGPT Moment”

The “ChatGPT moment” for autonomous vehicles is already happening in a sense, as AI and generative AI tools are speeding up development and making the technology more scalable. However, the point at which AVs become a meaningful part of daily life requires more patience. This will depend on the commercialization of the over 200 operational pilots around the world, focusing on profitable routes in logistics and big cities for robotaxis.

Full Episode Transcript

The Road to Autonomy Introduction: Welcome to the Road to Autonomy, the podcast shaping how Wall Street and industry executives think about the autonomy economy for today’s executives and investors. Simply following the headlines is not enough. That’s why our clients go beyond the podcast and the headlines. To get exclusive market intelligence and industry briefings, see what they see. Visit roadtoautonomy.com. That’s roadtoautonomy.com.

Grayson Brulte: Martyn, Bank of America’s everywhere in autonomy. I can’t turn left. I can’t turn right without running into somebody from Bank of America involved with autonomy and Bank of America recently released a perspective on all things autonomous vehicles. I was impressed. Why was the time to release the paper now?

Martyn Briggs: Well, thank you very much, Grayson and good to be here. And uh, well it was actually based on a bigger report that I sent you in July. If you remember entitled AV. You can Drive My car now we can debate whose title’s better. Um, but to answer your question, why now? Um, there’s a few things that I’m sure we’ll unpack in the chat today, but it’s a combination of the technology breakthroughs. There’s huge economic potential that’s starting to get more visible as these services come to market. Uh, demographic factors as well. Of course, you know, AVs can solve driver shortages basically. And lastly, the increasingly favorable regulation towards autonomous vehicles. That’s starting to play out now partly due to escalating geopolitics and competition between US and China tech companies. But I’d say the main thing, Grayson, is on the tech side, right? Breakthroughs in AI are the main. Story here, as I know you’ve covered on the podcast before. Um, and AI is enabling autonomous vehicles to generalize, be deployed in new areas cheaper and faster. And just quickly, a couple of the tech, the technologies we looked at in the report that I’m sure you’re aware of is, uh, the simulation, right? Reducing the reliance or full reliance on, on-road testing requirements that’s traditionally been quite slow and expensive. Uh, secondly, on the large language models to better understand driving decisions that the AVs take. Uh, thirdly on the falling sensor costs, you know, we’re seeing lidar costs of halve last year. Um, and then finally neural networks that con, uh, that continuously learn how to drive powered by data rather than being dependent on hard-coded software and maps that most AV companies have previously. So in summary, a kind of gen AI on wheels on the tech front.

Grayson Brulte: So you name it, you can drive my AV car. Your mind’s with the Beatles baby. Won’t you drive my car? So I’m gonna see, I’m seeing you next week in London. Are we taking a trip to Abbey Road together? Is that where we’re going?

Martyn Briggs: Well, possibly. Yeah. And that’s a good idea, Grayson, and maybe we can do that. But yeah, we’re certainly big fans of having some, uh, things that are big catchy titles, whether they’re song titles or references to things. And you know, I think at the moment, one thing that we’ll maybe discuss a bit is, you know, whilst we joked AV you can drive my car, I think the big opportunity here in autonomy is far broader than that as we’ll discuss. Right. And obviously cars are, are very important in this, uh, in the addressable market. But actually, you know, there’s a number of other areas and, and the other thing is that AVs are no longer a moonshot, right? You and I have discussed over the years, when are we gonna get this? When’s it gonna be become important? And, but it’s still very much been seen, like a five, 10 year view or, or we haven’t had the commercial services available to actually go and ride and benchmark. I know you’ve ridden in, in loads of them now and, um, and helped to set some up for me of course, as well over the years, which is always fun. Um, but that’s the point now is that, you know, you’ve got. Services that you can actually start to, to look at gauge, why would this be important, um, and where we’re going. And a fun fact for you, actually from the report, just from the four commercial operators that are operating avs now have driven apparently 190 million miles, which is apparently more than a return trip from Murph to Venus. So increasing rapidly as you know.

Grayson Brulte: It’s going, it’s scaling, and most importantly it’s global. It’s not just Silicon Valley, it’s just not the United States. It’s global. We have AVs in Asia, we have a AVs in Europe. We have AVS in Australia, and soon we’re gonna have them on multiple continents. And eventually, at some point, if we don’t have ‘EM ready, we’re gonna have ’em on Antarctica. Yes, they will be used for research purposes, but we’re gonna have some form of autonomous vehicles on Antarctica at some point in the report. Bank of America said the TAM for autonomous vehicles is 1.2 trillion by 2040 with 700 billion of that making up vehicles. That’s big. Break the TAM for me. How did you get to the 1.2 trillion by 2024? ’cause to me this says that’s a massive, massive opportunity.

Martyn Briggs: Yeah, sure. Well, I mean, I was hoping Grayson, that there’d be some kind of third party study out there that I could easily just capture the whole market opportunity and quote that, that’s what I was hoping for. And I think I phoned you up first and a few other friends and consultants in the industry trying to find it, but we didn’t find it. Long story short. So, um, so what we did, uh, we didn’t create a model ourselves or some, you know, assumptions we’re gonna update every quarter. That’s not really what my team did, uh, or do, but we looked at about eight or nine different external, uh, forecasts for each of the potential AV markets. Um, so that’s the important thing, right? Um, is that, yes, the car’s very important, I guess, to highlight two things. One is that. Rightly, we see a lot of focus on autonomy on the car as we’ll discussed. That gets the headlines right. We can all relate to it. There’s an emotive aspect to it and we can all think of the use cases quite quickly. But actually there’s some markets that might be faster to automate with a clearer ROI. So think ports, logistics, mining, agriculture, of course, trucks and delivery. Um, and actually when you add all of those respective forecasts together for these industries, you more or less double the tam. So $700 billion is a big opportunity in 2040 for AVS in cars, but it jumps to that 1.2 trillion when you include all those other areas. Um, so what it actually refers to is the potential market or revenues for AV technologies, both the hardware and software investments in those respective markets. And a way to think about it is that, obviously it sounds like a huge market, but we’ve got a $75 billion market now for sensors and ADAS systems in cars. So that could, you know, potentially jump tenfold to 700 billion due to autonomy. I think both adas as well as the robotaxi markets combined, but it’s just for the AV tech spend. And you’ve probably seen other numbers much higher than 1.2 trillion, that talk about 5 trillion, 10 trillion markets, which usually, from what I understand is the, the total mobility market that they’re starting to talk about. EG you like what’s the, the end customer likely to pay for that service, which of course could be much bigger than 1.2 trillion. Particularly if AVS can lower the cost of travel, which I think in the medium to long term, we will see. So bottom line, huge economic potential for the companies that are developing the AV tech in the short to midterm, but could enable a much larger TAM than that over time.

Grayson Brulte: The Tams are gonna grow. And I’ve been doing some in independent research, which hasn’t widely been discussed, but I think it’s important to get your perspective on, I’ve been speaking to a lot of older individuals, and when I say older, 75, 77, all the way up to 82, 83, and there’s that common denominator for certain individuals that. Know, I’m gonna get to the point where I can’t drive anymore. Then the interesting trend, this has probably come up, I would say about 30 to 35% of the time. What if I got Tesla, FSD? Could I keep my keys longer? Would I be able to go more places? I said, oh my God. And so you’re starting to get these, these older individuals that are starting to get interested in, in this ADAS technology one. What do you think of that trend that I’m starting to uncover? And two, do you think that could be a catalyst of those indi, those older individuals converting and eventually going to using robotaxis as they age outta the driving process?

Martyn Briggs: I think that that is a relevant use case, particularly so is the younger drivers, right? People that haven’t even got a driving license in the first place and you’re already starting to see a reliance on some of the robotaxi networks that are there already for, you know, kids and even if they’re, you know, not supposed to be signing up themselves. So I think it works both ways. One is that you could extend mobility for people that either unable to drive longer because of older age or disabilities and reasons that preclude them from driving or having a driving license, and equally on the younger side as well. So again, it increases the type of use cases and addressable market beyond those that have a driving license. But, you know, obviously the important caveat here is that, you know, once you are being totally dependent on that system, um, it has to be, depend, it has to work at all times, or at least in the operator, the ODD, that’s, that’s gonna be defined. Um, so at the moment, you know, some of those that you’ve mentioned in the automotive space are gradually getting higher levels of autonomy for it to be a fallback and dependent on people that. Can’t then take over control of vehicle. It needs to be more towards the so-called level four and higher levels of autonomy that will come eventually and, and are starting to come in certain cities, but it’s still quite expensive.

Grayson Brulte: I think the, the great unlock here in the United States, and we had, uh, Mr. Jonathan Morrison was recently sworn in, actually today, officially a swearing in ceremony with Secretary Duffy to be NHTSA administrator, you have to see the F-M-C-S-A rules I would call modified or enhanced for the future, where when you start removing the pedals in the steering wheel and you sell it, then you have a huge opportunity. So, so that younger individual that, that can’t legally drive, can go in that vehicle and then an older individual who’s incapable of, of driving because of the aging process. That becomes interesting. Is that the great true unlock when we remove the the steering wheel and the pedals? Do you think?

Martyn Briggs: Well, I think yes, to a certain extent for two reasons. One is you could change the form factor of the vehicle, right? You know, the current, uh, many of the current robotaxis and vehicles that we’re seeing are adapted, um, vehicles that we see on the road already. I think a bigger unlock to your point, both from a capacity standpoint and design and how we’re gonna utilize the vehicles will be once you’re removing the, the steering wheel and pedals altogether. But as you say, there are important regulatory hurdles to overcome. And they’re different in the US and, and other markets, right? That’s another thing that, uh, we wrote a bit about it in the report of the, on the regulatory side, at least being both an opportunity and a challenge now, right? The opportunity is that if you and I wanna start an autonomous business tomorrow and start in a certain city there regard, you know, rules and regulations, how you do that, but if we then wanna scale it to a hundred or a thousand cities, that’s when it starts to get a bit more tricky. Um, so, you know, to answer your question, I think as we see some of those regulations change on the need for steering wheels and pedals, it’s an unlock for the type of vehicle you can use. The capacity, therefore the money that you can charge and the, the extra, uh, revenue can make from the vehicles. But we also need to be realistic. What’s the kind of potential scale of that in the short term? We are seeing a few companies go down that route, like Zoox, of course, um, and, and others that are, who’ve got some, some vehicles they’re building already. But the volumes so far have been, I guess, largely retrofitting existing vehicles to get to market quicker.

Grayson Brulte: Volume’s really gonna come down to, we’ll get this a little bit as contract manufacturing and the manufacturing capabilities. Then I asked this, this. Is, what role is Magna gonna play? ’cause they play a very large contract manufacturing role. They have, they have world-class engineers. But the stand the regulatory front for, for a moment here, it’s, it is 2025. Today, the UK government has very publicly stated by 2027, they’re gonna allow fully autonomous vehicles on the roads of the United Kingdom. You, you, you live in London. You’re based in London. How is the UK public thinking about this technology where it’s two years away from a regulatory perspective?

Martyn Briggs: Well, sadly I can’t speak for the whole UK public yet. Not, not yet. Maybe one day. Um, but I, I think, you know, was, I’m kind of in the, involved in a lot of the discussions with some of the companies and certainly investors looking at this. So I dunno if my views are a little bit tainted towards the more optimistic side, but it does feel that you’ve had a lot more announcements on, you know, what the, the capabilities are from the companies. Obviously you’ve got Wayve here in London that get a lot of focus. Also oxa on the industrial logistics side. And you’ve got a, a number of companies that are now, um, have come through some of the initial proof of concept to deploying stuff with, you know, with bigger partners. Um, so there are, there’s an activity on the company side as well as some of the international companies that are hiring and expanding, uh, both from China and the US into, into London as a market. So to answer your question though. It goes back to what we were saying around what the, these approvals will actually entail, right? Initially you’ll start to see some trials with safety drivers, um, when you can remove the safety driver from the equation and operate more of a commercial service. Looks like that will be in the, some tests in 2026, some commercial activities from 2027. And that’s not very far away now, to your point, right? So I think there’s already a bit of an awareness campaign coming from some of the, the government department for transport and the various different sectors involved in that. Um, you know, but. Ultimately, there’s a lot of excitement there, certainly in the circles that the, that I’ve been speaking with. Um, there, there’s always gonna be a bit of, I guess, hesitation on the, um, the use cases, the safety, the, you know, maybe pushback from some of the, the, the people that could maybe get disrupted from this in the mid to long term in taxi drivers, et cetera. But by and large, I think the, the, the sentiment for AVs in, in the UK is positive as we’ve seen in many cities, right? Once you, now you’re seeing expansion in areas like where you are, and certainly in San Francisco and in China, there is becoming a bit of FOMO from other cities as well. We’ve been to San Francisco and tried this, we want it in London or we want it in Paris or other areas. And I think that’s also helping to change a bit the, uh, the opinion as to why we need it. And from the regulator standpoint, starting to, um, make some of the, the rules, regulations, and standards for which these vehicles can operate in.

Grayson Brulte: I said this years ago when I was the co-chair at Autonomous Vehicle Task Force at City of Beverly Hills. I said, you know it when a autonomous vehicles made it. When the queen, when the queen mom goes for a ride and the vehicle, obviously the queen Mom ha has passed on, but now we, you have King Charles. I’ll say this, when King Charles goes for a ride, and I’ll caveat this, either at Windsor in a secure location or in some sort of formal royal parade, you know that the UK is ready for autonomous vehicles.

Martyn Briggs: Yeah, I think, well, you aren’t getting higher profile visits and, and use of the vehicles, right. You know, whether it’s the, the royal family. I understand. Um, bill Gates did a, did a ride in one and, and we’ve got quite a lot of focus, Richard Branson as well. So all these names are starting to, um, to. Talk positively about the technology and what it can do, but I think that’s, uh, you know, that that’s one aspect of, you know, persuading people that this is, this is gonna be useful. I think on the other side, and we wrote a bit about this in the more in the report is the critical need, right? We’ve spoke a bit just already about the economic potential, the huge tam, which has always been the underlying principle of why, you know, autonomy should become a thing. Um, but you’ve also got the demographic aspect as well, right? Driver shortages in, certainly in the more the trucks and buses and commercial vehicles, but impacts the taxi market in the same. The same way, I think the average age of a taxi driver in Japan, something like 60, um, in London, you know, you’ve got, uh, already a, a short or potential shortage of taxi drivers that’s been reducing in the last 10 years. I think there was a study from TFL showing if the current rate of decline of drivers were to continue within 20 years, there would be no London taxi drivers, right? That’s how severe the decline has been. So, long story short, there’s, yes, there might be some campaign on the awareness and, um, why people should use AVs in the positive side and the safety case, but also on the demographic side as well, underpinning the, the need for, uh, for, for these vehicles in the first place.

Grayson Brulte: What is driving the decline of the London tax driver? Is it the, the education they have to go through, the tests they have to pass of knowing every in and out in London? Is that one of it, or is it just individuals or? Interested in in other career path.

Martyn Briggs: I think it’s a combination of things. I mean, on the taxi front, you know, you have seen a number of other technology platforms, you know, the rise of right. Hailing being one of the, of the obvious areas. But as you say, there are a number of hurdles to go through and it’s, this isn’t just London, you know, you know yourself in New York and other, other markets. There’s always been that kind of barrier of whether it’s cost or in London, it’s the case of getting the knowledge of being able to, uh, know every street and be able to, uh, to be tested on that without the use of a, of a map or technology. So there is a barrier to entry there, which, you know, starting to reduce a little bit going forward. I think ultimately it’s kind of persuading, you know, people into that, into that career path as a driver, whether it’s a taxi, uh, taxi driver, bus driver, truck driver. There’s a reason why there’s kind of shortages in all of these different areas. Um, and it’s not just about pay, it’s about the kind of the, you know, the conditions and whether people wanna do that for 8, 12, 16 hours a day. And, uh, so I think that, you know, maybe autonomy can help some of that if there’s, in, in the short term, you’ve got more assisted driving features and technology could, uh, you know, could maybe be a positive in that front. But across the world, you’re seeing driver shortages in, in the commercial vehicle segments and in taxis and autonomy isn’t gonna address all of that. Right. You know, and, and I think that’s the important thing to note is that in any of these markets, you’re not gonna see zero to a hundred percent penetration in, in any of the areas. And many of the companies have spoken about this, the limitations they have on supply in the short term are, are notable, right? Either getting access to the type of vehicles, the technology. Um, but yeah, nevertheless, the, the, all of these things combined mean that it, it could be another driver of sentiment and importance for autonomy.

Grayson Brulte: You hit the nail on the head. Right now feel the global market for autonomy is energy constrained and it’s vehicle constraint on the, on the vehicle front. What has to happen for that constraint to be unlocked? Is it large scale manufacturing? And if so, when do you really see that truly coming online?

Martyn Briggs: I think there’s a couple of things, right? Yeah. I think on, you mentioned about Magna and some of the companies that are, that are involved on the contract manufacturing side. I think what we’re starting to see now is that, you know, you’ve got this ecosystem emerging of partners that are starting to, um, to, uh, to be involved in either building the vehicles or the fleet management or the several different parts that are, uh, that are starting to take place here. Um, but again, that those investment decisions are, um, are not just gonna take place by one or two companies. There’s a number of people that need to make those investment decisions. Um, I think what we have seen started to change and, um, is on the, the different, um, uh, that the ecosystem around fleet management, charging, cleaning, maintenance, those things are becoming a bit more evident. On the manufacturing side, I think you alluded to it earlier on in, you know, not just in Magna, but others, there’s a big piece for contract manufacturers, um, that I, and maybe even OEMs wanna do that themself as well, if the volumes are big enough. But given the kind of, uh, opportunity that autonomy can bring for bespoke, um, or designing vehicles bespoke to even certain cities, um, that’s what many of the taxi regulators wanted years ago. Right. Is a, just a specific, just taxi for London. A specific taxi for New York. And I think actually autonomy could be something that, uh, that maybe brings back those conversations if somebody’s prepared to manufacture them at a, at a sensible price.

Grayson Brulte: And you know who loves that? The governments, because it increases tourism revenue, they love that. Anything to increase tourism revenue, governments around the world will applause and cheers. Do you so? Do you see to a point where perhaps let’s call it, you have the very famous London Black cabs in New York. You have the yellow cabs and various other cities. But fast forward to the future where there’s autonomous versions of that, where it’s called the Big Apple Mobile, and then in in London you have the Big Ben Mobile. Is that kind of where you see this maybe going?

Martyn Briggs: I think in the, in the long term for sure. Right. Because if you, you know, going back to your point earlier around the, the regulations around steering wheels and pedals, I think that’s, that’s obviously dictating a lot of the current, um, how much you can do with a vehicle in the first place. And it’s obviously then retrofitting existing ones. But I think that the vision for many of the cities could be if you, you, you’ve got the opportunity to, I guess take a reference design, um, you know, whether it’s a skateboard platform, ev something that then you could add a different top hat to it. Um, and if that becomes much cheaper to do, then it would be now in smaller volumes then, you know, the business case could start to stack up. Um, I guess for the more mass market appeal though, um. We might start to see some different regional collaborations and partnerships take place between car companies and others to, uh, to have a bespoke autonomy platform. You’ve seen a few of those examples coming already. Um, and, but ultimately you’re gonna need scale to make it profitable and make more people want to want to go down that route. But, um, I think, yeah, either, either route is, is viable in the long term. You might some more bespoke design in the short term. I think you’ll start to see more of the autonomy kits being, um, designed in from the factory, right. Or built in to make it cheaper. I mean, even you, you look at some of the, the, the Chinese robotaxi companies that are announcing, um, the kind of price points now, um, of, you know, 40, $50,000 in some cases. And so, you know, whilst that might not include all of the work that’s gone to date in the r and d, the, the cost is, is definitely coming down of, of what the additional AV technology requirements would be. Um, and that’s the first step to getting more of these bespoke vehicles that, that you’ve just mentioned.

Grayson Brulte: Looking to the long term and, and we have some data to analyze. Foxconn has had ambitions over the years. They bought the old Lordstown motor plant, which they converted for Monarch Tracker in Ohio. But then a few weeks ago, the Nikki out of Japan reported that they, and I was able to verify this from a press release from Nissan that they bought an old Nissan factory. Long-term, does Foxconn enter this potentially.

Martyn Briggs: I know you’ve spoke about it a bit on the, on the podcast before Grayson, and I mean, we, we hosted the company, or MIH, the, um, the, the, the consortium that are developing, um, uh, EVs or a number of different parts of a couple of years ago at our future car conference that you spoke at. Um, so I think they’ve, you know, made quite clear that the ambitions is to, uh, to start to piece together partners that could ultimately design an ev themselves from the ground up and take whatever, you know, go as far as they would like, whether it’s just taking the batteries or just taking the, the skateboard chassis and then building on top of that. But ultimately, I think it’s more, so far we don’t see any evidence of companies like that wanting to vertically integrate autonomy and do that themselves. I think it’s more working in a contract manufacturing facility on behalf of other partners, and whether it’s them or some of the other companies you mentioned earlier on, it feels like that could become more of a, uh, of a viable business for the, for these companies, especially given initially the volumes are not likely to be. Millions. Right? Um, it could be something that, uh, that, that’s outsourced to any of those, those providers. So it feels like the building blocks are starting to be there. Now. You’ve got EVs that are easier to maybe, uh, make a reference design and build on top of with a certain chassis and then customize that dependent on the design, the, the different, um, aspects that you would need for autonomy and make it so with a different top out on top. So it feels like that’s where we could see some traction going forward. Maybe not as many clear cut evidence of it today, but you do see a number of different robo shuttles and um, uh, and, and different form factors. Certainly those that are in the Middle East actually, that are being announced now as a lot of the activity there, um, around the robo shuttle front.

Grayson Brulte: The UAE is becoming a, a very interesting market, what I’m looking for and watching in the market for any signs at the Foxconn. Potentially could have legs is the Nuro lucid deal. I think it’s very interesting. So the SEC filing states a minimum 10,000 vehicles contracts were guaranteed. Nuro received the first robotaxi this week built on the lucid line. That gets interesting. I’m looking to see, okay, does another OEM, perhaps it’s Nissan, which has a, a deal with Wayve, is that Waymo finally gives us some, some data to analyze on the Toyota deal and do we start to see a pattern emerge there around licensing? And if we do see that pattern, do you think that accelerates what you did? A well, brilliant job covering in the report the growth of the personally owned autonomous vehicle market.

Martyn Briggs: I think so far what we’ve seen is there’s been a maybe a bit of a, um. A lack of willingness from those companies to give a, a number. I think the lucid Nuro Uber deal was the first that actually put a, a, a potential number of what, as you said, at 10,000. Now I think it’s up to 20,000, isn’t it, over time, um, as to how many vehicles could roll off the production line into those, um, uh, into that type of model. Most of the others have creating some early partnerships to, I guess, explore what’s possible, but not really giving any kind of firm commitments around volumes yet. And I think that’s fair, given it’s quite early stage as to what the, you know, what the opportunity would be and equally. It’s less known how they, how much integration work could be done in the factory to bring by the car companies and others to bring that, um, in-house versus retrofitting. Um, but I think ultimately it’s gonna become more, more of an important business model for car companies in future. I’m personally a little bit more skeptical on the personally owned robot Taxii front, um, because two things. One is the. More the maintenance requirements. I could see this more akin to the fleet. You know, if you’ve got a fleet of them, both from the financing standpoint, but also the, you know, more rigorous maintenance and upkeep of the vehicles to manage downtime. If it’s you or me owning our own personally owned av, and then I decide to maintain mine and you don’t maintain yours, and then there’s an accident, it starts to unravel it. A number of different liability questions. And so I think it, it could work where maybe you are the owner and then it’s operating in a fleet that then you are paying for the maintenance when it’s operating in that fleet potentially. But I think the easier business model to, to, envisage for now is where if, if you’re operating in a commercial sense, it’ll probably be owned and operated by larger fleets. Um, now. To your question on licensing, I think the business models a lot of these tech companies are looking for is that no one in the long term is really gonna wanna own and vertically integrate all of that technology. I think because of, well, the balance sheet implications being one, if you’ve suddenly got a million, 2 million, 10 million vehicles on, on the balance sheet, but also the, um, again, all of that kind of fleet management, maintenance charging and who’s gonna have to do some of that individual parts of it are already started to be outsourced to different partners. And I think that’s gonna probably make sense in the longer term. And but to, to be able to license that te technology reliably and safely, um, you know, there’s still some engineering hurdles and challenges that the, the companies need to, to be able to certify that, that they can do it across 1000, 10,000, a hundred thousand vehicles. Um, and obviously that’s the stage that, that they’re trying to work towards, but it certainly makes more sense for the tech companies that are involved here. That’s the kind of model that they’re used to. Um, and would, you know, certainly makes sense for, for the tech companies involved.

Grayson Brulte: I said this the other day to a friend, I said, at some point in the future, not gonna give a timeline. When and is my belief that Alphabet would like Waymo to operate in a very similar model to Android with a, with a caveat where they’re getting a licensing fee. Waymo licenses the, the Waymo driver to X manufacturer. They build it, they own the asset, and then Waymo gets. A licensing that gets really real interesting if that happens. And if you look at the cell phone tower business, do we get for companies that are formed or call ’em a Thomas vehicle, reach real estate investment trust that own the real estate and own the vehicles and hold them because you’re looking at billions upon billions of dollars of own those assets. And we do have some clear indications. Bloomberg has reported that Move IO out of Africa is looking to raise a billion dollars to own Waymo vehicles and their balance sheet. So it seems like there’s a potential early Kindles in that market for something that could emerge.

Martyn Briggs: Yeah, I think there’s, you know, as the, as the market opportunity gets bigger, you’re gonna have different asset classes want to get involved in the, in the AV ecosystem, right? Whether it’s in owning the vehicles themself in the fleet management, even charging, cleaning certain aspects in the same way that you have a huge business behind car rental and even ride hailing now, right? And car leasing companies that, that could be involved in that conversation as well. Um, so I, I think we are starting to see signs of, you know, companies that, that are looking to take on a specific part of the value chain. Um, and yeah, I think ultimately that’s gonna become a bigger investable asset class when it comes to the real estate and other side of it. I think, you know, that. Ultimately these cars were always gonna need somewhere to be financed clean, parked where they’re not in utilization and, and so there’s always gonna be opportunities there. Most of the conversations I’ve had have been with institutional investors, right? And the first question is, you know, what stocks can you look at what companies to, to play the value chain? And that, that has been quite limited until recently, where you’ve suddenly now got a number of, of companies that are either pure play or developing the, the, the AV technology in the first cl in the first place, but maybe less so on some of the fleet management side of it, which is at an earlier stage, uh, at least on the listed company side. Many of those that you’ve mentioned are still private, but it will become more of a, uh, of an investible asset class over time as the AV market grows, and not just in cars, as we’ve discussed. Right.

Grayson Brulte: I remember, I think it was seven, eight years ago, an individual came up to me and. And this is not investment advice and said, how do I invest in this sector? And I can only think of two companies. I won’t, I won’t, I won’t name ’em for privacy purposes, but I can only think of two companies. And now we have seen that the market is getting much larger. Today is Thursday, September 25th, and Kodiak went public today on nasdaq. Don and team, congratulations. I watched you ring the bell on nasdaq. Well done. Congratulations on that. So now we have two pure publicly traded autonomous trucking companies with a, with Aurora and Kodiak. From your point, from a research perspective and, and, and looking at this market and speaking with institutional investors, is this a sign of, of, of things to come where a, where the market from, let’s say investible, lots from a publicly traded standpoint starts to get larger, where more companies potentially might look to go public?

Martyn Briggs: I, I think so. I think, you know, you see whether, I mean, whether companies need to be public or not, I think at this stage doesn’t matter. I think it’s more. Looking at getting a, um, the technology readiness, right. The, and, and also, you know, we joked earlier on that it’s no longer a moonshot. You’ve got some commercial activities. Most of the companies that are now, um, you know, either raising capital, both private and public, are starting to be at or close to commercialization stage. I think that’s the, that’s the important difference now is it’s no longer a kind of vision. It’s showing what the technology can do in real circumstances. Right. And you’re seeing that applied in, on the car and robotaxi side of the businesses, the trucking companies that you just mentioned. You know, and you’ve got others that are still private, but are also doing driver out operations and planning some this year. I know you’ve hosted some on the podcast, right. Waabi, Bot Auto, others, so, and others. So I think there’s, there is an opportunity now where companies can show a specific. Commercial opportunity, a business case, a route, um, you know, in the case of Kodiak and Plus and others that they’re looking to, to commercialize and the business model. And then people can take a view as to whether that’s, uh, you know, whether that that’s the type of investment they’re comfortable with. Um, so I think the first stage now is the, the commercial activities. We’re no longer waiting five, 10 years to see what that might look like. Um, still maybe a bit early to see what the, um, I guess what the type of technology integration’s gonna look like inside the vehicles because everyone has a slightly different approach to that and a slightly different business model as to what, whether they’re proposing pay per mile driver as a service, license, direct licensing. So that still may be an earlier stage of development. Um, but we ourselves, you know, seeing a, across cars, trucks, delivery, warehouse ports, logistics, all these areas where you’ve got a number of companies starting to, to have real things you can test and, and, um, and analyze.

Grayson Brulte: I think the largest data set we can analyze is Southwestern Australia with the mine data, and look at how long that the mining’s going on. So they’re seeing a 30% increase in productivity, a 50% increase in labor, which which you documented there in the report. How do you see those numbers growing over time?

Martyn Briggs: Well, I think so far that most of the, the things we’ve seen have still been in pilot stage, right? So I think that the, ultimately it’s around either the, the cost saving that you can bring from removing the, the labor costs, same as what you’d have in removing from a, from a taxi, et cetera. Um, on the productivity side, it gets more interesting because, you know, it’s not only is it freeing up labor to do other things, but you’re starting to see evidence from those, uh, those examples in mining and agriculture where it’s maybe like less tire wear or less maintenance, or it’s safer, right? You know, less accidents and loss of life, or reduced exposure to dangerous areas for, for, for humans to go into in the first place. So, you know, then putting a number on that can, could be larger than 20, 30% as the pilots are showing now. Um, but it’s not just about the. The core cost reduction, that is maybe the easier thing to discuss in right? Hailing or robo taxis. It’s kind of a, you know, a change of, uh, the functionality of, of human roles in those businesses. Um, and you start to get, uh, the other benefits around longevity of the vehicles, as I say, maintenance, tire wear and things that might not be the obvious ones. So I think, yeah, the, you will start to see more potential benefits from those, uh, for, for those sectors, uh, beyond just the, the cost, the labor cost argument.

Grayson Brulte: what you’re seeing too, you’re seeing health cost savings. I spoke to Tim Bucher on, on a previous podcast. He’s the, the founder and CEO of Agtonomy doing autonomous trucks in California, sits on the, on the Callo advisory board to, to change that role. And I, I was talking to Mr. Butcher. I said, so why are driverless tractors banned in California? And he told me an answer and actually shocked me. It goes, because he goes, you’d put a little brick in the old days, in the seventies, how’d the tractor go? You jump off the back, throw your habell on to keep going from protective. It was actually done to save lives. And he said, so the law was done for the regulation was done for a good reason. And he goes, what we really need to figure out, he goes, it’s not just enabling autonomy on farms today. He goes, it’s the back. He goes, it’s the buttocks. And I said, what do you mean? He goes, the number of, and he told me the exact number of health complaints and medical costs of sitting in the tract world, they go up, up, up, up, has huge, huge medical costs. And he said, it actually has been scientifically proven at u uc, Davis to potentially shorten lives. So think about autonomy. You’re, you’re getting rid of the, the critical dangerous elements, but you’re also increasing somebody’s likelihood to live a longer, healthier life. That’s impressive. I think we should talk about it more.

Martyn Briggs: I think there’s, there’s other benefits like that, that will come to the fore. I mean, obviously the, the reason why people jump straight to the commercial ones, certainly on, on the, in the investment case, it’s obvious, but is because you can start to make more of a tangible reason why this, this technology might work. But let’s assume that in any of these segments, autonomy becomes more, takes more of a meaningful share, then these benefits are the ones that will start to be talked about, right? Whether it’s on the, uh, on the health side, uh, and also freeing, creating a role in, in the case of agriculture, right? You’ve got a number, a thousand other different jobs that, that the farmers might wanna do, rather than driving a tractor up and down every day. And, and also it can then extend the, it could be used beyond the. the. usual hours that they, they’d wanna work on that it could, you know, be utilized in two, three shifts and for maybe more than one farm, right? So the, the business case might change overall. So I think a number of things like that are starting to, to, and finally is, and we mentioned this a little bit earlier on, is that some of these segments are maybe easier to automate because, you know, they’re not on roads, they’re not, um, working in operation with pedestrians or at least not as often, um, maybe easier from a regulation standpoint. Um, and maybe from an accuracy standpoint as well. So there’s some of those areas might be far or will be probably faster to automate, even if it’s not the kind of $700 billion tam that you might get in cars in the midterm.

Grayson Brulte: I still think trucking is, is a great opportunity because if you, if you look at the, the, the growing labor shortage as you highlighted Japan, and you’re right there, we have it here in the United States. You, you look at the, the rising cost of shipping, but then the other element that I talked to investors about. Consumer demand. Consumer want. Oh, you go on Amazon to order, I want it an hour. Why do I have to wait? I want it now. No. Oh, I have to wait two days. That’s not good. Next, let’s go to the next one. It’s that consumer per, it’s what consumers, consumers want. So I think that’s, I think consumers are going to push the adoption of autonomy in logistics and especially trucking. Going back to the report again, this was fascinating, you did really good research on this, Martyn and the team, I gotta give you credit. 60 autonomous trucking pilots running around the world. Wow. How did you uncover that? Where did you put, you’re in London, so did you put your Sherlock Holmes hat on and, and go through it? How did you find that?

Martyn Briggs: No, you are very much the Sherlock Holmes in this one. And Grayson, I know you, some of your detective work is, uh, second to none. I mean, on, on that one, I mean, that was a, a particular one was a Bloomberg study that, that, uh, that we, we saw and met the team there that have done some work around, uh, their views on autonomy. So, um, actually they had two, two reports. One was on the, on the robo, uh, taxi and shuttle front, and one on the commercial vehicle pilots. Um, so as you said, there’s 60 or so on the, on the commercial vehicle side, I think about 120 or so on robo taxis and shuttles. So, you know, you had, you know, close to, uh, with Del if you include delivery trucks as well. I think it was almost over 200 operational pilots that are going on around the world, which was more than I thought when I first started looking into this. Right. And so, uh. Even if, you know, back to your point around how quickly will we see things commercialize. We do have on the robo taxi side, at least about seven or nine cities now that are, uh, 24 7. Any weather operating, um, uh, commercial services, less so on the truck and delivery, uh, delivery services at the moment. But given you’ve got those kind of 60 to, well, I think almost 90, if you include the delivery vans as well, the pilots that are going on, um, even if 10% of those come to market in the next couple of years, you’ll then start to have the evidence base that we’re starting to see in robo taxis and the excitement and how they work and what, um, uh, you know, what type of business models are behind them. Um, but yeah, so we’ve, we’ve did look at a couple of studies, mainly the Bloomberg one in that instance, but there’s a few others that, that we quoted in the report on just the, the scale, which is actually larger than a lot of people think in terms of the operational pilots that are going on around the world.

Grayson Brulte: If you look at, say, I’ll say this, 2024 to me was the really true kickoff for robotaxis. ’cause Waymo just started scaling. They started making data publicly available and as announcement after announcement, that market really started to take off in, in 2024 is 2027, the year that autonomous trucking real off ’cause Torc is gonna start operating driver route that year. Wbi has said they’re gonna operate driver route that year plus as they’re gonna operate driver route. So it seems like you have in the front, in the front of the pack right now, you have Kodiak that’s operating driver route in the Permian Basin. You have Aurora that’s going from Dallas to Houston with a, with a safety driver, but they’re doing it. You have Bot Auto. This completed a 40 mile run. It seems like all this momentum is now leading up to 2027. What do you think?

Martyn Briggs: It does feel like the, the announce the years announced by some of those companies, as you say, four or five have said commercial operations in 2027. Now I think that’s, but on the other hand, that doesn’t suddenly mean that that becomes a banner year where. Everything scales in just that year. Right? I guess it’s a a, a first test bed for what commercial activities all these companies are gonna be able to deploy. Um, and you might actually see some come sooner. Right. You know, the companies you mentioned, there’s already a few, um, examples where certainly the ones that are off highway like Kodiak and others that you mentioned, um, that are operating in, in certain use cases already. So you might start to see a few activities sooner than that. But I think from then, um, once you’ve started to see demonstration from these companies on exactly what they’re proposing, what routes, um, you might see the same start to happen if we’ve seen in robo taxis and on the car base services. Right. You’ve got. Then a number of different regulators and cities and areas will look to look into more detail as to how the, the these services can work under what type of operations, what type of regulations, um, and start to usher in a new era on the commercial vehicle side as well. Um, and the other thing I guess is the type of operating areas might in, in the short term or probably more highway based than urban deployment, at least for the trucks, you know, maybe different on the delivery vans. Um, but you’ve got a number of point to point type of applications on highways that, um, had already been tested that obviously have meaningful revenue, uh, meaningful cost savings potential from a labor standpoint, safety fuel, and all the areas that the truck companies are looking to exploit here. Um, but it doesn’t need, again, like we said, in, in cars, doesn’t need to be a hundred percent of the market to make it interesting. If you just have those kind of high repeatable routes that, um, uh, that could be the profitable ones with a, with a select number of large partners.

Grayson Brulte: And I mean, that comes on the partnership because, but staying on the, in the ecosystem, in, in the report Bank of America is projecting by removing the cost of the driver in a robotaxi can reduce taxi costs by 52% a mile. That’s huge. What goes into that? Is it, is it the labor cost? Is it the the start stop? Is it the fuel efficiency? What goes into that number?

Martyn Briggs: No, it’s a bit more, more simple than that. I mean, our autos team did that, did some analysis. Um, you know, mainly looking at the, the Tesla FSD launch a few months ago. And, and the build up to that. Um, and the work that they did was looking at the current ride hailing cost per mile, which I think was three and a half dollars per mile, if I’m not mistaken. Um, and estimating about 52% of that is driver costs. So the remainder is the vehicle infrastructure and of course, the profit that the, the, the platforms make. So that would almost entirely be removed in an, in AV operations, the 52%. Um, not entirely, right. You’ve still got some remote operator, some customer service costs. The fleet management stuff that we just discussed, um, maybe that’s less certain exactly what the cost per mile of that’s gonna be for now, but it will add some incremental cost back in. So you might not get a total 52% reduction, but by and large, a big part of the, the current taxi or ride hail costs could be removed. So what’s required to make that reality? Um, you know. Beyond the removal of human drivers. I think optimizing and developing, uh, AV operations and infrastructure directly could become key to AV economics in the longer term. I mean, I don’t think all of that’s gonna become vertically integrated, of course, but there’s specific parts of that value chain now, as we just discussed, with companies looking to do that. Maintenance, charging, insurance, fleet management, et cetera. Um, and of course. Cost discipline in all those areas over time will dictate who wins as well as the cost of the vehicle tech. Right? That is coming down significantly as we discussed earlier on, partially because of, of ai, but partially because of the, the cheapening sensor costs as well that are required to do this in the first place. But yeah, long story short, it’s uh, the 52% per mile predicated on removing the cost of a, of a, driver from the right hailing side of it. And then you’ve got a lot of opport. Now, not all of that will be passed on. Clearly. There’s a, there’s a, a profit taking opportunity there and you’re not seeing, you know, right. Uh, robo taxi companies undercut platforms in most parts the world for now. Uh, ’cause they don’t need to. Right. They’re kind of supply constrained and, you know, people like you or me will turn up and pay whatever ’cause we wanna test it for now. That won’t happen forever. But if you do get this significant cost reduction per mile. If any of that is passed on to consumers, you increase the addressable market for, you know, vehicle miles. Travel could increase. Um, and yeah, you increase the opportunities and maybe start to compete against other, uh, well even car ownership in the long term.

Grayson Brulte: ‘ costs continue to go down, B of A is publicly projecting. By 2035, we hit peak car. With 20% fewer cars on the road doing 50% of the miles driven by 2050. Walk us through that.

Martyn Briggs: So yeah, that was actually some assumptions quoted from, well, a few different studies that we looked at. Main one being, um, uh, uh, Bloomberg and the F1. I think on the, uh, the potential reduction in the extreme most, uh, high AV scenario was I think minus 20 or minus 23% vehicles on the road by 2050, but importantly, doing 50% more miles now. Thesis here is not a new one. Right? If you remember going back to the days of car sharing, ride hailing and the proliferation of a lot of mobility services, um, you know, include people, including me, you got quite excited that, well, if the cost of this comes down significantly enough, you can remove the need for private car ownership as the default, or at least as many number of cars that people would have in a household. Because you can start to rely on these different alternatives in a more mobility as a service type of world where you access stuff on demand and don’t need to own it. Um, but of course there’s a reason why private cars are still, um, you know, very relevant in, in the market that we have is the safety net, the certainty, it’s your, your own asset. And of course the cost. There’s a very compelling argument as to even with rising costs of insurance and all the other areas, car ownership is still tends to be cheaper on a cost per mile basis than ride heading. All those other things we’ve just discussed now in a robo taxi and autonomous world, the, the scenarios that we looked at that well. Peak car or the, the number of cars on the road being sold was probably gonna flatline from about 2035 out anyway. Um, and maybe even before that. So you’ve got, you know, the, the kind of growth markets for automotive have, um, are not gonna continue forever for a number of reasons. But if you start to add in a number of high a, uh, autonomous vehicle penetration scenarios, you could reduce the number of cars in the fleet or car park as it’s called, but still assume that number of miles are gonna increase because of economic growth, ’cause of different demographic factors. And finally, if the, the cost per mile that AVS can enable, um, uh, you know, makes the cost come down, then you increase the addressable market over time. So I think there’s a number of things there that, uh, that, that are important, but the TAM could actually increase even if the number of cars decrease in the next 20, 30 years.

Grayson Brulte: And when you. Of Tam and investors could have interested, I wanna give some numbers that you publicly put in the report here. During the period of 2023 to 2024, 9 billion was invested in a Thomas vehicle companies. Let’s zoom out to 2010 to 2024 time period. Over 200 billion. That’s right, over 200 billion. Was invested in over 600 autonomous vehicle companies. It’s now 2025. As we head into 2026, where do you see this trend heading?

Martyn Briggs: Well, I wouldn’t wanna give you a number for next year as a direct one. We haven’t made any kind of projection or forecast, but what you did see from those numbers we quoted and historically is you saw every year since, uh, 15 years ago, the kind of funding was increasing for autonomous vehicles. Then you’ve had a kind of decrease from 2021 to 2020. To 2024, um, uh, to 2023, sorry, because of a number of areas, consolidation, number of players that have either disappeared from the market or been acquired and they’ve moved to more commercialization phase. And then this resurgence in funding has been concentrated into those companies that are now starting to launch things commercially. Right. Um, so I think two things now is that. You’ve had that initial Wayve of, of a number of different companies that are looking to, to do autonomous vehicle technology development. Now you’ve got a smaller list of players that are ready to go to com commercialize. Um, and what do they need that money for? It is to scale operations, right? It’s not just to develop the AV tech anymore. It is about investing in the underlying infrastructure and things we’ve discussed a bit already. Um, so, uh, whether the number goes up or down, I think it’s not as much the important trend anymore. It’s are the companies that are looking to scale having sufficient runway and sufficient funding to be able to do that. Um, so far, you know, we’ve, we’ve seen three or four companies that are doing so we’ll probably see many more next year that are raising more money around that. Um, but I think it’s, we’re at that stage of commercialization phase. And the other thing I’d mention is, and we wrote about this in the report, is, you know. Whilst the 200 billion or so is the, what, what’s been invested in the kind of autonomy related companies. If you scale it out or look at all the areas related to shared mobility, obviously electric vehicles, um, and a few related areas, um, you know, you’ve then suddenly got a trillion dollars that’s been invested in that same kind of timeframe across 4,000 companies. So yeah, the 200 or just over 200 billion is, is a big number for autonomy. But actually there are related areas in software development, in mobility services, in obviously electrification that also will touch upon the autonomy related ecosystem. Um, but yeah, the, the figures have, have always been large of investing in autonomy. We have seen a re resurgence for, uh, for 2024, I think for the next couple of years. It’ll be interesting of the companies that are actually receiving that money and what they need it for. Right. Which is probably more the commercialization phase than just building out a team of, of software engineers to do this.

Grayson Brulte: it’s commercialization, paying for, for compute and, and data centers. We’ve all seen those public numbers as the industry consolidates and you’re getting, I, I don’t want call ’em ownership. I think that’s, that’s premature, but a, a smaller group of companies that have a better shot at getting into long-term profitability. How are institutional investors thinking about this? Are they starting to dip their toes in or explore the opportunities in autonomy?

Martyn Briggs: Yeah, absolutely. I think that’s one of the reasons we did the report now. Right. We find you, you sense that, I mean, on the, on the one hand we saw a number of different companies expanding, the technology’s getting better, all the stuff we discussed earlier. But you know, that was also translating into a number of, uh, inbound requests from investors as to, you know, how big’s the market, who are the key companies, what’s your view on this? Obviously that always gives you a signal that there’s interest of investors that are doing work around the topic. So I think as we discussed a bit earlier on, it’s no longer just looking at who are the top 2, 3, 4, 5 companies, but also how does the underlying tech work, how do the companies differ on that front? That’s why we went into a bit of a section around this emerging rules-based versus end-to-end approach to autonomy, which, you know, is, is obviously, uh, becoming more important. Um, and ultimately the the global expansion potential, right? And, and finally geopolitics, right? All of these things, investors are starting to look at not beyond just who are the top 10, 20, 30 companies to, to consider here also the bigger picture around, um, you know, all the, the market that they can address, the, the expansion plans, um, the implications of this as well as we’ve discussed a bit today.

Grayson Brulte: is there certain pockets of, is it say, is it Hong Kong, is it us and are European and UK investors coming on, especially with the UK with Wayve raised the billion dollars from SoftBank with Uber and Nvidia having participation. And then Jensen was recently in London. He signed them, the GPU for Alex and, and put out a statement where they’re gonna invest potentially ’cause it’s not done well. Were potentially an additional 500 million into Wayve Like, oh wow. Okay. This is getting interesting. So is it, where are you seeing this institutional interest from?

Martyn Briggs: I think across the, it’s hard to, a certain region is more interested than others, but I think that the key questions that we are getting is around. What different technology approaches these companies are taking and why that’s important. Right. So the example you just gave with say, Nvidia and Wayve or others, is that, you know, what, for some of the more end-to-end approach to autonomy for those type of companies to be successful, I guess the, the scalability comes down to data centers, GPUs, compute, um, that might become maybe more of the blocker versus beforehand. Um, if you’ve got a, a. Rules-based approach to autonomy that’s dependent on more and more sensors, more and more vehicles that you might need to invest in and deploy your on your own. Um, then it’s a game of who can raise the most money and, um, have the best technology. And so I think things are shifting a little bit to be a bit more dependent on the scalability of autonomy and how it can generalize to new markets. Um, all companies are starting to approach more AI in the, the way that they achieve the, um, uh, how they navigate and how they operate. Um, and that could be done much cheaper, right? Maybe with not as much requirement on HD maps and other, um, technologies that have before maybe been quite expensive. Um, and even, of course, the hardware front as well, right? The, the access to, um, GPUs doesn’t just sit in the data centers anymore. You have supercomputers on wheels effectively, right? And the number of chips that that cars are coming equipped with already in the factory and. Those that can be added for robotaxis with, you know, with GPUs from NVIDIA and others are starting to, uh, to come embedded within the vehicle. So the compute starting to be there. The, the training of the models via data centers and AI capabilities is getting faster. The simulation tools are getting better as well, of course, which precludes the need for as much on road testing that’s slow and expensive. Um, so I think all those things combined mean that, yeah, you can look at the scalability side of it, but to answer your question regionally, I think, you know, whilst you could, you could pick several different pockets, right? The US has always been at the forefront of developing the AV technology, you know, right. The way back to funding competitions from DARPA and the government and, and you know, and roots back to where a lot of these companies have started from. You’ve had a load of activity in China more recently as three or four companies have started operating commercially and now expanding internationally, which I think is important to note. But finally. You’ve got a lot of, uh, penetration of, um, level two, level two plus systems. Um, and, and the implication of that is a lot of the hardware is being embedded in automotive grade, uh, volumes, which is bringing the cost down. Doesn’t mean that all of that will be sufficient to operate a robotaxi service necessarily, but it is bringing the kind of volumes and cost down in sensors, lidar, radar, camera, and equipment, that that reduces the cost burden of the AV tech overall. And of course, finally, the GPUs and compute needed to do this.

Grayson Brulte: And I wanna highlight something that we haven’t discussed yet that hasn’t really been discussed openly in this industry. The other day, I read this really long, interesting, fascinating report on American Express and the trial, I’ll, I’ll use the term, this is my own terminology, the trials and tribulations of when American Express tried to embrace AI for using it for customer service applications, American Express took the data set, dumped it into a large language model, and you know what happened, you and I know it didn’t work very well. So American Express had to spend billions of dollars to re to reformat the data to get into the proper format for the LLMs. And now it’s been a huge success for them on the customer service front for, for identifying issues. I think when you, when the reason I bring American Express up is when you look at the data from a lot of these autonomous driving companies, to me it comes down to the quality of the data. Uh, how was it labeled? Is it in the right format? And I could almost sit here and make an argument that if you’re starting today with really great format of data, you can accelerate one, you can make the, I believe you can make the car drive faster as we saw with Bot Auto, but two. The cost of training go down. When do investors start to really, truly pay attention to the cost of GPUs, the training, and if you want to call it the infamous cloud bills, when do they pay attention to those growing bills?

Martyn Briggs: I think for, if you’re talking about it directly in an autonomous vehicle sense, you’d, you’d need to see more, well, certainly more scale of operations, but secondly, more disclosure around what that entails. Right? And you know that, that, that you don’t have at the moment. Um, we did show some things in the report around the potential energy burden of autonomy, right? And, you know, if, but again, assuming that GPUs and the, the compute used for autonomous driving gets no better, you could if, uh, theoretically half of an electric vehicle’s battery would be used from the autonomous vehicle system. Now the reason that’s probably flawed is that you and I know that if that does become an issue, those companies will make power efficiency, uh, energy efficiency. The name of the game, and that will be, be given a lot more focus that that won’t become an issue. Right. So similarly to what you said on data centers and training, whilst we showed a couple of charts on Right. If we assume we scale up to a few million vehicles that need that kind of, uh, level of training and capacity on a per vehicle basis, we see today you could see a tenfold growth, a hundredfold growth that, that some of the numbers that Nvidia had quoted, I think at their conference earlier in the year. But again, if that does become a meaningful problem from a cost standpoint, then. There’s gonna be a lot more focus around the bringing that cost down, and that either means more efficiency of the hardware or that you’re gonna have, um, uh, well more efficiency in the cloud and access to the data centers. And also at the moment, a lot of the training of the, uh, autonomous vehicle models is ongoing and continuous, and we’re in early stage of updating those models as, you know, maybe that the, the need for the amount of times that the models are retrained, maybe that goes, reduces over time, right? At least per vehicle. So anyway, problems today that we can foresee getting a bit bigger in terms of access to data centers, the hardware, the training, the energy efficiency. But I think if, you know, like anything you see already the, the cost of a, of a chat, GPT search or inference for large language models was a problem three years ago. Now that is, is reducing orders of magnitude every, every model update these companies do. And I think autonomous driving will be no different in that sense.

Grayson Brulte: You’re right. At some point perhaps the autonomous driving has steep stigma restricts me to the, the Chinese market. In the report you said 84% of Chinese cars are expected to have some self-driving capability by 2030, and they’re gonna have twice as many Robbo taxi as by 2035. How close attention should investors be paying to the Chinese market? Because you’re seeing, we ride in Europe now you’re seeing Baidu now you’re seeing pony AI in Europe. So you’re seeing, I wanna say outside these, these growing global ambitions.

Martyn Briggs: Yeah, I think a lot more focus is starting to be a paid, uh, to be, be paid to this, right? And the geopolitics of the AV race, and I think that’s something that investors need to pay more close attention to because, and I think I wrote about it in the report, is that, you know, China’s focus is starting to go a bit or switch a bit from EVs to avs, right? Everyone knows a bit the China EV story now of the, I’ve, I’ve got one myself in the uk. The, the, you know, the, the cost of it, the, the, we can talk about that maybe in a different conversation. But when it comes to avs, I think given the proliferation of EVs in China, how do you compete, right? And the, the car companies are now starting to compete on autonomy, right? The, the low, uh, L two, L two plus systems, I guess trying to compete against Tesla’s FSD and, and exceed the capabilities of that. So that’s one clear aspiration that they’re making. The other is, um, deploying some of the higher, uh, level equipment sensors, specifically lidar, uh. Partially as a marketing tool, but partially to be used for higher levels of autonomy. So you embed or maybe over equip the hardware today for what you’d need, um, on the premise that you increase the levels of autonomy over time. Um, and so our, our China Autos analyst and the team there did some assumptions, as you said, that believe 84% penetration of autonomous, um, vehicles in the China market by 2030. Most of that being level two and level, uh, level two plus and three systems only, I think about 4% of that being highly automated, level four robotaxis. So that’s, I guess, the important distinction, um, that there still might be a, you know, bit of a gap before we see higher levels of autonomy. But the important thing to note is that as we’re de deploy, deploying all of those sensors, the cost is coming down, the cost of lidars Hald last year and the year before. Um, so we’re, we’re seeing the same kind of things that happened in the EV transition, um, starting to be applied in the AV transition, as I mentioned earlier on. Some of the companies you mentioned, I think, uh, pony AI said right at our conference last week, the cost of deploying their next gen robax will be $40,000. Not of the AV tech all in the cost of the vehicle. Now, you know, whatever that number ends up being, um, obviously it’s, you know, depends what your, what’s gonna be included in that and the extra r and d cost and other things on top. But bottom line, the cost of developing AV tech is coming down for some of the Chinese players. Can that be replicated and adopted and deployed in Europe, the Middle East and others? Well, there’s about six markets that they’ve announced now internationally. Um, and we wrote a bit about this with some of the charts in the report, is that you’ve got some international aspirations from the likes of Waymo and others that, that are launching internationally, but you’ve had far more cities announced by the Chinese robotaxi players. Um, so. Exactly how they’re deployed and what type of vehicles remains to be seen. In Europe, you’re seeing a bit more of the, uh, the robo shuttles, right? So the deployments from those companies have been a bit more in the akin to public transport, which I guess is easier for, for regulators to, to see tests deploy and, um, you know, more, uh, there’s obviously more openness to maybe public transport and sharing and, um, and could be deployed in, in European cities in the Middle East. A few of the pilots have been more with the ro, the right haing players, the same way you’ve seen in the us. So yeah, it is becoming more of a competition globally between, you know, US and China AV companies and a number of others that, that will, uh, that will benefit from that. Um, but yeah, I think that’s, that’s also underpinning the, some of the regulation and policies that are, that are starting to play take place to have national frameworks and encourage that, uh, that expansion of companies internationally. So don’t miss out on this opportunity.

Grayson Brulte: and as the market grows, pay attention to to the geopolitics and the Martyn of the market. Martyn, putting this conversation into a mixing bowl. We’re gonna make a great cake. I’m gonna see you next week in London. I’m really excited for it. You ask the question in reports. I’m gonna ask you the question here. When do autonomous vehicles have their ChatGPT moment?

Martyn Briggs: Well, two answers to that. I think one is that we’re seeing some of that now, right? That’s why we did, uh, the report and a section around some of the technology and how it works partially to get. You know, my head into it a little bit more and what the differences are, but you could argue we are seeing that ChatGPT moment in automotive more generally. But in autonomous vehicles playing out at the moment because you are seeing the application of many different generative AI and, and simulation and AI tools to make this technology more generalizable and scale faster. I guess the bigger part of your question is when’s it gonna become a meaningful part of your life? My life, other, other people in cities and um, you know, for that I think we probably need to be a bit more patient. Sadly, I think it’s, uh. See the number of cities that, that, uh, that we’ve got, as we said, those 200 pilots that, that are going on around the world operationally, not all of those will come to market and you’ll see where are they gonna be deployed ultimately is where there’s gonna be commercial activity and opportunities, right? So I think in the robotaxi markets, it’s gonna be the big ride hail, or big taxi cities in the logistics and the trucking markets. It’s those bigger profitable routes that will have a tangible use case for deploying autonomy. So again, it’s not, it doesn’t go from zero to a hundred percent of any of these markets, sadly, for, you know, for some time. But we are seeing the so-called ChatGPT moment of bringing AI to automotive and certainly it’s speeding up the development of autonomous vehicles. And ultimately that’s why a lot of people are starting to, to do the work, look at the investment opportunities and getting excited about this opportunity again after a few years of maybe disillusionment and, and not as much commercial activity.

Grayson Brulte: The bottom line is autonomy is scaling. Autonomy is becoming a business. The future is bright. The future is autonomous. The future is the autonomy economy. Martyn, thank you so much as always for coming on the road to autonomy.

Martyn Briggs: Thanks, Grayson. Good to speak to you as always, and I’ll see you next week in London.

Grayson Brulte: See you next week. Take care. Bye-bye.