Walt, I saw the headline in the Financial Times and immediately I thought of the new Beetlejuice movie. The Waymo Uber divorce went public and, as in the movie, it appears that it's getting messy. And then today, Zoox called up Willy Wonka and they got the golden ticket. Holy cow, after 12 years — Zoox is 12 years old — they can finally charge for a ride. And then Aurora reported earnings, and I got a lot of questions out of that. But let's get into the messy divorce, because if you want to pull up Michael Keaton, here's Beetlejuice. Or here's Johnny if you want to go to The Shining. We can get into it. So, start with the messy, messy divorce.
That was definitely a very intense opening. Well done. [laughter] A better cold opening than what I had to endure with some of Brandon Ross's opening songs on the LightShed podcast. But that should wake up our listeners right from the get-go. And everyone's eyes kind of popped out on Friday. This is after we had recorded last week's episode, with the Financial Times giving some more details about this topic that you and I have been talking about for the past year, which is the Waymo and Uber relationship. And while many people will say, 'Oh, you know, expected,' the stock was off 5%. Not expected ultimately by investors to that degree. It was certainly another leg down in the stock, and we've talked about how every kind of plus and minus of this relationship has resulted in some movement in the stock. But I think the key data here is that Waymo is going to have its own app in Austin and Atlanta in January of 2028. The entire contract ends in May of 2028. I think we just have to assume that it will end by then. So we have timelines, and so when we go to Austin and Atlanta in January 2028, we won't have to hope that we get a Waymo when we use the app and try to figure out ways to get the Waymo there. So, I guess the first question I have for you is: do you think this really ends up lasting until January of 2028? Do you think we will not see a Waymo One app in Austin and Atlanta before January of 2028?
It's my belief that the game of autonomous roulette will end sooner than the date that the FAA reported. Now, if you want to say the cat's out of the bag and now that it's been publicly documented, publicly reported and confirmed by the companies in that article — I want to point out it wasn't speculation, it was confirmed by the companies in that FT report — I say, and I'll say this for the record, this relationship is going to terminate sooner rather than later. The question is, does Uber try and hold on until they have a partner to replace it that is ready for it? That is the thing that I'm watching.
In this report, and it's citing the companies, you know, they're talking about complaints on both sides, right? Waymo talking about dirty cars, which I know is one of your hot-button items. But also bad routing when you think about that. Very high-profile and embarrassing. You know, the mainstream media loves to pick up on these Waymos circling around a cul-de-sac and make the autonomous vehicle look bad, but apparently those cars were sent there by Uber. On the flip side, Uber's talking about, you know, how it doesn't work in weather, okay, or weather no-shows and unsustainable financial terms. Again, according to this press report. I guess let's start on the Waymo side of things because they're the ones that seem more anxious to get out of this agreement. It feels like it's Uber that's holding them there till January of 2028. You know, if Uber is in fact routing their cars in this way where the cars are not getting maintained the way Waymo is expected, wouldn't Google have good enough lawyers to have had out clauses to get out of these contracts prior to January of 2028?
Alphabet has some of the finest lawyers in the world. And the one thing with autonomy, as we've seen in the various different incidents that have happened with individuals sabotaging these vehicles, there's cameras and there's data. So perhaps these lawyers are going through the data and comparing it to the data they have in-house to potentially find — and now I don't know, I'm just assuming because you and I both know and our audience knows these vehicles gather a lot, a lot of data.
Again, you know, let's get a specific answer out of you. When will we see in Austin or Atlanta a Waymo One app?
I will go on the record and I will say by June of 2027. I'm going to take an under on that because I just think that this relationship is at this point publicly, you know, torn. So like, what's the point of extending this? And I think, you know, certainly Uber can operate. It could be a quick switch, right? I don't think Uber can obviously continue to operate in that market without any problems. And you know, I'm not sure how long it really takes Waymo to stick them on their own app. It can't — I wouldn't think it takes that long. So, I'm going to take the under on your June 27th.
The reason I'm saying June 27th — so today's July 30th. These are all hypothetical, made-up numbers. Waymo's going to have to get a depot. They're going to have to get charging infrastructure in the ground. They're going to have to get permits. That takes time. So, you're looking at months of that. And then on the Uber side, I'm also taking the hedge of how long until AV ride is ready to go in Austin. Perhaps AV ride will be the new partner that replaces Waymo. That's just an assumption there, but based on the presence that we've seen from the Texas Department of Motor Vehicles where the AV ride vehicles are clearly ramping up in Texas.
I mean, that's some optimism right there by Grayson on behalf of Uber with one of their partners in AV ride. And you know, I put out a report on this earlier this week. I went through each of the partners — Motional, AV ride, and our experiences that we both have had with them. Nuro, Rivian, Bobby, Zoox, right in the US, May Mobility even has a partnership. I mean, you look at the most optimistic scenarios for these companies. And you know, a year from today, what in aggregate of cars autonomous will they have on the road? I'm going to say autonomous, Grayson. I'm going to say even with a safety driver. I'm going to give you the safety driver. How many do you have on the road in a month in aggregate? Our guess is less than a thousand. And then compare that to where will Waymo be in a year's time. You know what the number is? 10,000. 10,000 in 30 markets. And how many are the Uber partners going to have in a year? Now look, we don't judge this industry over the course of a year. You know, it's really where are we going to be in year three and year five and what have you. But like, think about that and the catalyst over the next year and how that looks. Fault supervised is not autonomous. Supervised doesn't count. Waymo has set the bar. It is not supervised. They've set the bar in terms of vehicles, which are roughly around 3,000 according to Bloomberg. They set the vehicle in terms of odd size, the global benchmark that everybody is striving to meet. And not to be Debbie Downer, but the companies that are on their heels, as you see in the Road to Autonomy Robotaxi Index, it's BU keeps ramping up. And so it is supervised, doesn't count. And if you look at Uber's domestic US partners to get to the scale of Waymo, where they are just clearly based in the US, I will give you a number: they are probably at least three years behind to get to where Waymo's at least. And that's from a technology standpoint and that's from a regulatory standpoint, because if you look at Uber's partnership with Nuro in California, the regulatory time frame, Uber's three years behind Waymo in California anyway. You slice it, you start multiplying that around, you say 10,000, I'm going to go 15,000 cars by the time that Uber gets to some sort of meaningful competition. And the cars are going to be sub — so for all practical purposes, we'll have a 14,000-car lead. That's a big instrumental amount. And we're coming to the fall. So, is that your beloved Philadelphia Eagles walking their way into the Super Bowl? Is that what Waymo is doing?
Well, there's a lot to unpack there. But let's start with you. You're taking — even I was trying to show the disparity with where Waymo is going to be versus the Uber partners, and you even made that disparity deeper. Fine. But in terms of insurmountable, I don't want to say anything's insurmountable. If anything that AI has taught us in terms of these models keep leapfrogging themselves and open models, and the fact that like at the end of the day, 10,000 or 15,000 in my mind is not scale, right? In the US alone, Uber and Lyft have 2 million, 2 million drivers there. So nothing is insurmountable. You know, you get things, you get a good OEM relationship, you start cranking these things off the line. All I'm saying is that, you know, what is the — from an investor standpoint, we look at things three months, six months, whatever, a year over the next year, you know, where are these two companies going to be with their partner? And by the way, if for example Nuro, which I've said many times in this podcast, has a couple hundred cars in the market — that's in one market, that is huge for them. They can raise off of that, they can, you know, build a better, you know, be a stronger partner. But again, you have to view this in the context of where Waymo is at that time. And the other thing that's going to be happening at Waymo at that time is we'll see where the Hyundai relationship is, where maybe initially you've got some upfitting, but maybe Hyundai evolves, you know, into more inline manufacturing, scaling going up even further. And then secondly, you know, what other new OEM that Waymo starts to clock on, which wouldn't necessarily be driving any volume by the middle of next year, but at least like, you know, what are we going to be doing on this podcast? In another, you know, 50 more episodes from now, we're going to be saying, what's happening in the middle of 28, right? You just continue to roll the clock forward. So, it's not that the car number itself is insurmountable. It's kind of where the momentum of Waymo is going to be at that time.
What we do know on the OEM front: the chief product officer of Waymo about two months ago gave a speech in Japan in Japanese media. Omega uncovered it and translated it for us, where the chief product officer of Waymo in this speech talked about a quote-unquote deepening relationship with Toyota. The speech was given in Tokyo, not covered by American press but covered by the Japanese media. Deepening relationship. And then four days ago, the chairman of Hyundai gave a speech in Silicon Valley, talked about Hyundai's deepening relationship. That's going to get very interesting. Let's just — this is hypothetical. So the chief product officer of Waymo says deepening relationship with Toyota. Then the chairman of Hyundai says deepening relationship with Waymo. Waymo has two world-class global OEMs potentially with deepening relationships — Hyundai and Toyota. That gets very interesting very quickly if we're reading the signals correctly there.
Time will tell. Now look, in this note, you want to read it, just self-register with a company employer's email: www.lightshedtmt.com. We also delved into kind of, you know, where is [laughter] our friends at Nvidia in all this, right? This was the one thing that gave Uber stock a bit of a rally at the start of the year. Where are they? And I delve into that and provide four recommendations for what Uber should do. So check that out on our website, and maybe we'll get to those four recommendations over the next couple of weeks. Should we move on to NHTSA, or do you have anything more to say on this Waymo-Uber battle?
I will just give a plug. Walt wrote a beautiful piece, very elegant, and he gave it a little Mark Twain twist to it. That sounds like a Trump. Very beautiful, very elegant.
Well, if you don't like the people that read it, you can tell them to self-report out of the article, but that's a whole different story. So, that's that. Read the article. But on the golden front, Zoox got the golden ticket. NHTSA says, 'Here you go. Here's your golden ticket with a cap: 2,500 vehicles.' But most importantly, they can charge. Andy Jassy must be doing cartwheels, backflips. Now that I said that, now they can charge. Do we get any statement from Andy Jassy on the Amazon earnings call about Zoox, so they can actually, after 12 years and billions of dollars, they can finally make $1?
I hope that they're going to get at least some type of shout-out, because you're right, this is a momentous occasion. You know, thanks to the NHTSA administrator Morrison, who's going to push this forward. I think it's 2,500 vehicles per year for two years, so they can get up to 5,000. Which again, that would be a huge win for Zoox. You and I both have our questions about where that technology is in its capabilities right now. You may be able to charge. I'm not sure how many people are going to be willing to pay if you're only doing like six, seven, eight stops in a given market. So we'll see. We'll have to see how the technology advances. I don't think they were waiting on the NHTSA administrator before they're going to unleash, you know, true flexible, let's call it, autonomous service rather than just, you know, these fixed locations. But yeah, positive. And what is the read-through in terms of Tesla? Because I actually kind of in a way view it a bit negatively in that I don't like this 2,500 cap for Zoox. I don't care because like I think 2,500 would be a win for them. For Tesla, where it's like, okay, the theory is when it's on, it's on and it should be going gangbusters, then I'd get worried if there's any type of limitation of the Cybercab of less than 2,500 a year. So is that in fact a bad signal for Tesla if there's some type of cap that's associated with these, you know, wheel-and-pedal-out vehicles? I don't necessarily think it's a bad thing for Zoox, and I don't think it's a bad thing for Tesla. I think it's actually the opposite. It's a very positive movement for the entire robotaxi industry. Zoox, it benefits more because their vehicles clearly cannot scale as fast as Cybercab. But 2,500 vehicles — you said 5,000 for two years — that is a healthy amount of vehicles out there. Look at Waymo's most publicly reported number by Bloomberg: 3,800. So, in two years, 5,000 Cybercabs is quite a big number. And then perhaps — we don't know, we have to look to Mr. Administrator Morrison for clarity on there — perhaps after that two years, that cap goes away. Perhaps it's the, if you want to use the spring training or the minor league. So, I view it as a positive because Tesla apparently, or clearly based on the ruling, has a very viable path to deploying 2,500 Cybercabs a year.
I mean, I could not disagree more in that like — yeah, 5,000 seems great now for a company in Tesla who in Austin has been in that market for a year and can't even get to our benchmark, the Autonomy Markets benchmark of 100 cars available for commercial service. Sure, in that context, it looks great. But in the context of the believers in Tesla — and I try to believe in whatever company, you know, is hoping to achieve — the theory is that, you know, once you get these miles done for the Cybercab, which last podcast we talked about, you know, maybe it's 250,000 miles, maybe it's the end of the year, it's like first you're walking, running, whatever it is, then you're sprinting. Like if you're good to go at the end of the year, you're telling me that Gigafactory that we visited can't crank out multiples of 2,500 Cybercabs a year? Of course they can. Of course they can. So if you're now getting handcuffed — you know, so is this an inkling of them potentially getting, you know, handcuffed, you know, by this new thing, or is this just kind of phase one and by the time Tesla gets ready to go and then the NHTSA administrator is getting the data that he wants and he's demanding, especially in terms of first responders, that that number is going to be removed altogether? My read, reading the team, believes it's phase one and then over time once the data, the safety case is validated, that number evaporates.
Let's move on to something else that was in this letter from the administrator, which is this. I'm not even sure how to pronounce this, but it's an A2 send consortium with SAE, who you have done some work with, explicitly aimed at a single national standard. Is this what can relieve us of these 2,500-car caps, or what is that part of this all about?
Yes. And so first and foremost, Pete Dodds, sir, job well done, sir. Job well done. So this is going to run through SAE ITC. It's a consortium. It's a best practice, not a standard. I want to clarify there's a big difference. ITC issues best practices. SAE International through the working groups issue standards. So this group is going to come together and it will develop — industry will develop a best practice for safety to set a national safety standard. And it is my belief once the industry buys into the standard and then it is recognized by NHTSA and the federal government, the cap — the 2,500 cap per year — will go away at that point. So to me, what Administrator Morrison did was say, okay, we are going to give you a path to start charging, and once the best practice is done and the federal government signs off on it, that cap will go away. So watch the work that Pete Dodds and the wonderful team at ITC are going to do. It's going to be really, really important. It's going to have a frankly huge impact, huge impact on the future of robotaxis in America.
So maybe that's the answer I was looking for, which is that this 2,500 thing or the Zoox approval is just priming the pump. It's getting the federal government behind the industry while you have local municipalities fighting it, at least in some areas. But the longer term is a single national standard, which by the time hopefully that Tesla is ready to go — which I know people that are more bullish than I think that's like tomorrow, but to me it seems like more like end of year, early next — that this stuff will be resolved. How quickly do you think this consortium can work? How quickly can they deliver, do you think, on a single national standard? Obviously, the timeline here is the president's term.
Just to clarify, not to go in the weeds, it's a best practice, not a standard. It's really going to depend on how fast the consortium can move. I will give you a very optimistic, bullish number: 12 months. Realistically, you're probably in the 16-month range. It just depends on how fast the consortium and the members of the consortium want to move.
Well, you need to talk to your friends that are at the consortium and tell them they're going to need to work a little bit faster than that because this technology is moving faster than that. So, I don't think we want to wait for 12 to 16 months for a best-practices standard.
It's going to happen. And it's not the individuals and Pete Dodds's team and all the individuals that work at ITC running it. It is the companies themselves because this is a collaborative process. And how much resources are the companies going to put? How much effort and time are they going to put? And so I just think historically of looking at all the great work that ITC has done over the years, I'm sticking to that 12 to 16-month range. Would it be awesome to go faster? Yes. But the realistic likelihood of it happening is less than 10%.
All right. So now let's move on to — we had a lot of earnings this week, are going to have a lot of earnings next week. Aurora, Qualcomm, a couple of the OEMs. Let's start with Aurora, you know, which we talked a little bit about last week because of some of the customer milestones and driver-out milestones that they were hitting. You know, they gave some updates. Things are a little, I think, a little slower than what I had hoped in terms of some of the near-term stuff, but it's really getting to next year, right? And the number of trucks that they're going to have at year-end next year, which is a thousand. The number of trucks they're going to do this year is 200. It's definitely very back-end loaded. But it's next year, a thousand, 300 I think of which are going to be Volvo, you know, with our friends over at Volvo Autonomous Solutions. So I think a lot of those or most of those will be driver-as-a-service for Aurora. So that's low, that's capital efficient, right? You know, because Aurora — excuse me — because those are Volvo trucks, but the majority is going to be, you know, whether it's Conti or their existing guy making, you know, upfitting existing trucks. Seems like an aggressive target to get to that number of a thousand trucks by the end of the year. What are your thoughts?
It seems aggressive. Are there enough international trucks to go around? I'm not trying to poo-poo this. I'm merely asking a question because you have to look at the fleets, the carriers and the trucks, and Plus has a public partnership with Plus. We don't know where that stands with Internationals. So, we have to look into that. Are there enough trucks to go around? I mean, that's really the truly the line. And then does Roush have enough spare capacity to hit that number? So, I think there's a lot of cooks in the kitchen in order to make this happen. I'll be watching. I'll be rooting for it, but there's a lot of things that have to go right in order to hit that number.
Well, it sounds like Roush has the capability. But the reason that they didn't get to this, you know, 20 a week that they were supposed to be getting to in Q3 based on the number, because they're only going to end Q2 — Q3, excuse me — with 25 trucks in service. Sounds like it wasn't really Roush, it was his Fibernet, I guess, or whoever is the one before Amoyo, which is the ex-Conti, is delivering them the stacks. Again, all this is like near-term noise, proof of concept. Even owning the trucks for this company is really proof of concept, right? That ultimately what you need — just like Uber needs to own the cars to get it out there and happen. But ultimately what's going to happen is they're going to want these trucking companies to buy the stack, you know, from effectively Continental or whatever it's called now, Mobileye, and then just charge them based on, you know, on a service plan.
At the end of the day, doesn't want to be asked that heavy, but when I look at all this and I still have questions and I can't get straight answers, is that when SASO and the team at Volvo Autonomous Solutions ramp up and they're running the Aurora driver and then Aurora ramps up, who goes where and who competes with what and how does the margin split work? Because I truly believe that Volvo Autonomous Solutions is going to be a major, major player in the future of autonomous trucking.
Yeah, look, I have a lot of questions about how these contracts work, how they're charged for miles, how it's accounted for. Not necessarily getting clear answers on all of them. I'm sure that's because it's a bit of a work in progress, and I'm guessing that they're going to agree to whatever contract terms the customer wants. But I think also the importance of the near-term stuff is it helps them raise capital, and you know, they have money out until early 2028. But you know, even in this quarter when the stock was at $7, which some may find expensive and some may find cheap, but they were willing to issue shares through this program that, you know, effectively diluted the stock additionally because you're basically bringing in cash. But that's what you do, right? You opportunistically bring in cash. So now as you approach 28, you want to have more cash. So in order to get more cash, you want to have that be as cheap as possible. So you want your stock to be higher. And by announcing new customers, by announcing more trucks that are getting upfit from Roush, whether they're International or whoever, that's going to help them, you know, get to this end state of, you know, whatever driver-as-a-service, Volvo cranking them out and the thousands of trucks that are there. So ultimately these near-term numbers become very important even though it's, you know, you can argue like, who gives a crap about the 20 — 20 trucks, sorry mom — the 20 trucks that they have with Paccar now that they're basically going to like not use by the end of the year. Well, because it just shows some use of the product and again enables the company to raise money. So that's where these — and by the way, I've seen this, I've done this 30 years. I've seen these types of near-term unit KPI metrics and how they, you know, have an outsized importance in the overall success of the company. And I think it's going to be no different. Aurora. Aurora needs to do a better job at communicating precisely how many trucks are going to be in the market, at what time and by who, and under what type of revenue structure, even if it's not what it might look like three years from now.
