Transcription
AR Invest has a $2,600 per share price target, also known as fair valuation, for Tesla for 2029, almost entirely based on the successful deployment of a fleet of Robo-taxis. Tasha Keeney of AR Invest shares her thoughts on the truly profound implications of Tesla's FSD 13.2, which has begun rolling out, and the strongest indicator yet that Tesla's estimates and plans around the rollout of autonomy are realistic. By the way, in case you guys haven't been keeping up, that would be the first Tesla vehicles, Model 3s and Model Ys, in some pockets of the United States operating as robot taxis next year. By the way, it's December now, so next year is pretty soon.
Tesla started releasing version 13 of its full self-driving software to early testers and employees. Why do we care about this? Well, of course, longer term, Tesla has plans for full autonomy, so it's another step in that direction. But it's actually quite an important step because this is the stack. Previously, they were on version 12; version 13, which should have a number of iterations, is what they plan on launching a robot taxi network with. That will be next year.
What's different about this than past software launches? The biggest difference is that you can start from park, it can shift, and you can park at your destination. So basically, you're getting the features of what is actually an end-to-end drive for the first time.
All right, Tasha, we talk to a lot of people, and there are a lot of skeptics out there. What do you think are some of the key things that people are still skeptical of with Tesla's ability to launch, and what are the answers to those questions?
That's a great question. I feel like often—so to be clear—FSD right now is not perfect. I don't have version 13, but on my car, you know, there's eye-tracking technology. There still are interventions, is the summary point. I think often people say, "Well, okay, how are they able to launch when I still have to take over in these specific scenarios or when it doesn't work in bad weather or something?"
I mean, Tasha has actually raised quite a common point that comes up for many of the skeptics. Other than, "Well, they need light iron," and all that kind of stuff, forget that this is actually one of the more semi-serious points: What if it's pouring down rain and the rain's so heavy the car can't see? Then what?
Of course, if you think about this for more than three milliseconds, the answer is quite evident: the car will do what a human does in those conditions. It'll pull over and wait until it can drive safely again. A key thing that is maybe not talked about often enough is that they'll likely launch with remote operators. I mean, that's what Waymo does; that's what Cruise did. So it doesn't have to be perfect at launch; it just has to be good enough such that you can sub in this remote operator system when you need to.
Of course, the more you do that, the more complex it's going to be, and the more costly it's going to be. So they probably want to reduce the amount of time that the remote human operators help the car. To the best of my understanding, this is how companies like Waymo and previously Cruise operate. There are people sitting in a room or an office somewhere who can see and potentially take over the vehicle.
But there's two tiers to this for the remote operators. The first tier is the car will send a question back asking for a decision to be made. For example, "I don't know if I should merge over to this lane now or I should stay here. What do I do? Option A or Option B?" The car phones home, and Daddy says option A, and the car proceeds to follow option A. That's the first sort of level of intervention—pretty simple decision-making.
The second level of intervention is actually remotely operating the vehicle—less frequent but far more important when needed. And then, of course, the third level of intervention is when the car is so severely stuck that it can't be moved remotely, blocking an emergency lane for vehicles or something along those lines, and you physically have to move it.
Tesla hasn't really disclosed a huge amount of detail about whether or not they'll have remote operators. I mean, it seems pretty likely, and how this will work, but I think this is most likely right. The software is not perfect yet. In theory, Tesla already has a lot of not remote operators, but in-person operators who supervise. In fact, the word might actually be "supervisors" for FSD today.
So we have proof of concept: 95%, 98%, 99%, 99.9% of the time, the car's driving itself perfectly fine, but you're watching and intervening as needed. But I don't think Tesla's going to want to build up too much infrastructure around remote supervision and potential interventions because the software is going to get so much better so fast that this is going to be a temporary blip.
In the fullness of time, remote operators will not be needed, so it's going to be very curious to see how this rolls out. Does anyone have any other ideas as to how this could roll out? For example, without remote operators, I don't see how it would be possible, but I'm also pretty dumb. Share your thoughts in the comments below.
What about the infrastructure needed to support a robo-taxi business, charging being one part of it? I think in the demonstration, they showed a robot cleaning the car as well. I think charging feels like more of a hurdle, especially.
So I'm going to jump in here. First and foremost, he's obviously asking the question about charging Tesla's existing vehicles, 3 and Y in particular, because we saw the Cybertruck with the wireless charger. It can literally just reverse over one. Tesla will put them at their charging stations in the parking spots, and you don't need to think about that, right?
But for the software update that enables a 3 and Y to become a robot taxi, you do actually need to physically plug them in. If only Tesla had been working on a device—some might even call it Optimus—that was dexterous enough to perform this boring, repetitive, and extremely simple task as needed. Because then you could have an Optimus bot, just a single Optimus bot at a Tesla supercharging station.
Although I don't know how we would deal with people trying to steal one. Actually, here's an idea: maybe they somehow have a security sort of compound that only a Cybercab that's activated can actually get into automatically, and you have an Optimus robot there, just like a service station attendant, to plug and unplug as needed.
This is not as complex of a problem as people might imagine. We also saw at the autonomy event, which I was a little surprised by just the fact that they showed it, but Tesla's also working on or has already developed some autonomous basically Robo vacuum cleaners and arms and all kinds of things like that. This is nowhere near as challenging a problem as it would have been if Tesla didn't have an Optimus humanoid robot or the wireless charger for the Cybertruck, but they do.
So, next question: since they're doing a new type of charging, how do you factor that into the rollout of the robo-taxi network?
Yeah, so, you know, we want to hear Sam's thoughts on wireless charging since we've talked about it a lot. They unveiled wireless charging at the Cybercab event. It looks like, I would say, for the future scale of this service, that makes sense because then you don't need a human. The car just autonomously drives over a pad.
Yeah, very smart stuff. Oh, by the way, I just want to really emphasize how brilliantly perfect this is. Just think about any car parking space on Earth. If there's a car parking space somewhere on the planet, you can be certain there's electricity running to that location. The most remote parts on Earth without electricity don't have car parking spaces.
So everywhere already has the electricity service needed. Every single car parking space on the planet could potentially have a Tesla wireless charger installed. This is going to be stunningly convenient. Just think about this for a moment. Think about the shopping centers, the sports venues, the arenas, the central business districts in cities, the suburbs, existing supercharging stations, convenience store parking lots—you name it. If there is a car parking space, it could potentially have a Tesla remote charger installed that a vehicle can literally just reverse and park over, and bam, it's charging wirelessly.
This is insane, and I strongly suspect that Tesla's longer-term plan is actually to have distributed charging. So rather than having centralized supercharging stations—of course, there'll still be some of those, right?—but they've just got to distribute these things everywhere. Why wouldn't you? If you've got a fleet of robot taxis operating in a city, wouldn't it be a lot more convenient and make a lot more sense if these Cybercabs can just charge very close by as opposed to having to go out of their way to get to a charging station?
There's parking spaces everywhere. It's going to be so easy. And the most important thing is just from an aesthetic point of view. People might think, "Well, no one's going to want to do that in the car parking space." It's literally just a flat piece of stuff that goes on the thing, and you can't—no one's going to even notice, and it doesn't interfere with the parking space. If you're not a Cybercab, you can still use a parking space. They're going to be able to integrate so seamlessly.
It's amazing that gives it a full charge. They unveiled some, you know, if you look at the specs of what they said, it looks like it would be a pretty impressive experience, although you won't be in the car experiencing it most likely for the robo-taxi service. But I think at first, they're just going to use the infrastructure that they have.
So I expect this to be, you know, humans charging the car, cleaning the car. Of course, in the future, like why not have Optimus do some of that work? But do they need that to launch in one city? Probably not.
What do you think?
Yeah, I mean, I think you leverage existing infrastructure. If you have to hire someone to plug and unplug the vehicles, that's fine. It does sound like, you know, there was an acquisition related to the wireless charging. It seems like they're going to be able to do that fairly efficiently, and obviously, we'll see that roll out.
But I think the interesting thing—and we were debating this when they were talking about it, right?—because you had the snake robot that was going to plug it in, you have Optimus that could potentially plug in a vehicle or not. This seems to be tying into the Elon mentality of the best part is no part. Not only is it simple and fewer moving parts that you have to deal with, but you can actually remove those elements from the vehicle itself.
So, you know, you don't have to have that charging port. Obviously, you have something else, but no more moving parts—probably great for simplifying the vehicle, making it even cheaper.
What are the two cities they're likely to launch in next year? Which are those?
So they said that they're launching in Texas and California. You probably guessed that.
Okay, yeah, so she just described states. Obviously, the cities are going to be San Francisco and Austin. 95% confidence level on that guess. Now, it's possible, extreme sort of curveball scenario, Tesla could launch elsewhere in Texas, maybe out around Starbase, just for the laughs. They could probably easily get regulatory approval out there; it's Texas. But, you know, it's a small area.
But I strongly suspect Austin and San Francisco because they have a lot of data, a lot of Tesla drivers. Okay, it's maybe like the Anaheim area, yeah, maybe like the Silicon Valley area broadly, and then Austin.
Do you think the initial launch and rollout will have current Tesla owners opting in their cars to the service, or do you think Tesla's going to bring in some of the inventory they have and put it out on the streets?
That's, I think, another good point because I feel like that's also a point of confusion here. Tesla talks so much about, you know, so much to the audience of owners about, "Well, you can sign your car up, and actually, you can make a good amount of money off of the service." I think that they're doing that because they want people to sign their cars up to meet peak demand.
I don't think that's going to be—I don't think that's what they—I don't think every car, you know, in the initial launch has to be a customer signup car. I think they can use a small fleet of their own vehicles, but customer cars supplement it. So they want to get customers excited about it because they want that supplement. You know, they want to make sure that the volume is there when they need it.
But I long term, you know, this is a fleet model. So I think there'll be third-party companies. We've heard Tesla encourage these companies to sort of come about that own the cars and operate them for Tesla, and then Tesla keeps a small fleet to themselves.
I feel, yeah, this seems quite evident. That's the future model—almost entirely fleet operators. But early days, especially, it's going to be great for some Tesla owners to opt in.
Now, a lot of people have this idea: "Well, who'd want to lease out their car? What if somebody vomits in it?"
Oh, bro, you're not the target market. The target market is somebody who's probably a little bit younger, more entrepreneurial, less likely to have a family, more likely to be thinking, "Wait a minute, you mean to tell me I could lease one of these things or ten of them, and they can go make me money while I'm at work or asleep and replace my entire income?"
And yeah, maybe have to deal with some inconveniences along the way, but it's going to print money for me while I'm asleep, at work, relaxing, watching content on YouTube. Yes, please! I'll take as many as I can get.
Ten, fifteen years ago, I would have been all over this. When you're talking to the same guy who was running out of room in his one-bedroom apartment on Airbnb to make some extra income, yeah, figure that out. I had to sleep on my friend's couch.
So there will be enough people out there who are happy to add a Tesla vehicle that they own or at least one specifically for the network to supplement in the early days. But long term, you're just going to have gigantic fleets of these Cybercabs everywhere and Robo-vans, and a very, very small number of people who actually own a Tesla vehicle who are also putting it on the network.
It's not going to be zero. Like I said, it's just going to be the super entrepreneurial folks who also want a bit of extra income and are willing to take on potentially a slight amount of inconvenience along with that free money.
For a company that's so fast-paced, so fast-moving, so agile, for some reason, people, when they talk about Tesla, it's like, "Oh, it has to be one way, and they can't change or adapt." And it's like, that's obviously not the case. They adapt and change all the time.
So, you know, will they launch with their own fleet and perfect everything? It seems like that could be possible, right? It's like they want to have a great service. Then, exactly what Tasha said, you know, have other people. It's like part of Tesla's strength has been the retail support. It's like the referral program—everyone is so excited about it.
I think this ties into another one of those points that we often hear, which is people saying, "Oh, I don't know if I'd put my Tesla into the fleet." To me, that's a crazy statement because right now, there are Uber drivers sitting behind the wheel doing it. So you're telling me those people wouldn't sit behind the wheel and do it?
To me, that doesn't make any sense. It's like, of course, maybe, you know, a VC who has a Model S Plaid isn't going to want to do that, but I'm sure the vast majority of people would gladly lease out their car for extra cash.
So Sam makes a good point. However, I'm going to push back a little bit. I don't actually think—and I don't know the numbers—but I don't think it'll be the majority, as in more than 50%. If people had sense, it would be, but a lot of people are extremely emotional, and the bell curve is real.
So there will be a very large percentage of people who own a Tesla vehicle, and instead of thinking about the massive increase in household income that this product can produce for them autonomously while they sleep for a little potential bit of inconvenience now and then, instead of thinking about that and weighing it up, they're going to go, "Oh, I want people in my vehicle. What if something bad happens?" And that's it—game over.
I don't know what the exact number of people putting their vehicle in the network is, but I don't think it's going to be the majority. I'd love to be wrong, but I don't think people have as much common sense.
Now, there's this caveat here too: if you're really well off and free money isn't going to meaningfully change your life or your lifestyle, yeah, there's less incentive, right? But if you're a household who's struggling to pay the bills or has to cut costs—say no to the kids doing an extra sport because you can't afford it—that kind of stuff, you got a Tesla vehicle, or you could lease one and put it in the network and have it more than pay for its cost, and then free money? You'd have to be crazy not to do it.
But I think a lot of people are crazy. Anyone want to guess what percent of the existing Tesla fleet will be added to the network over time? Leave a guess in the comments below. And if you own a Tesla vehicle, would you put the vehicle on the network?
Yeah, and it's like, let's remember the kind of current trends that have been happening for the past decade plus. It's like fewer young people want their licenses. You know, ride-hailing has taken hold, so this kind of fits in with what's, I think, been happening for a while.
And I also feel like the call-out to owners is somewhat a call to entrepreneurs again to start that fleet model. So it's like, okay, maybe you feel like you don't want to sign your car up, but maybe the other person sees opportunity in this, and they want to buy, you know, 30 cars or something.
So I feel like that's a part of it too.
Yeah, I think that will be interesting because it's like in areas where you can set up charging and figure that out. But it's like, you know, four cities—maybe there are some very go-getter individuals who want to acquire an area for charging and, you know, whether it's acquire a garage or partner with the garage and then buy 50, 500 Cybercabs and try and dominate the city.
I mean, that's an exciting prospect, and there are actually already pockets of people doing this. I know I saw on X someone in Europe has a fleet of Teslas, and this is like their exact plan when the service is available there.
So I don't think it's too crazy of an idea. I think that there are already those people out there. It's just a matter of incentivizing them and, you know, getting the logistics of the program in place for it to work.
Does the fleet model exist in the current state of ride-hailing, or is it non-economical because you have a driver in the seat? Are there companies that go out?
Yeah, and I mean, if you think of the New York City taxi medallion model, often it's like a medallion holder and then a lot of drivers. Or, you know, one person holds multiple medallions, and that means multiple cars.
So I think it exists in history. We're in the ride-hailing space. Yeah, you don't quite see it operating that way today, but I mean, we saw Uber partnering with Waymo. They're going to maintain and house cars for the first time.
So I think that the ride-hailing players are wising up to that. This is kind of the future business model, and, you know, if they don't adapt, maybe they won't be around.
So I can confirm they probably won't be able to adapt, and they probably won't be around. The idea of Uber and Waymo partnering, to me, borders on hysterical. Imagine this partnership trying to compete with Tesla, who can produce a vehicle called the Cybercab for, call it, it's going to cost them like less than $20,000 to make.
But let's just say it's 20 grand, right? It's going to be less, but 20 grand put on roads, charge $2 a mile.
The numbers are very simple on that: $20,000 to produce, $2 a mile. They need 10,000 miles to have produced its entire costs in revenue—obviously not profit, but the actual cost to operate a driverless vehicle is approximately zero.
Yet you can still charge a comparable fare, at least initially. So their profit margin on that could easily be 30%, 40%, 50%, 60%, 70%, 80%. But let's be really conservative and say it's a mere 30%, which makes absolutely no sense.
But right now, look at it: make it a mere $6,000 on 10,000 miles. But a typical taxi does around 70,000 miles per year. Imagine a Tesla Cybercab does 70,000 miles, making $6,000 for every 10,000 miles driven. Not only could it pay for its cost in year one, but pay for the cost of manufacturing another Robo-taxi instantly—an infinite money loop.
Wild stuff! Yeah, you could be conservative and say, "Well, what if it only does half the miles of a typical taxi?" Well, it still paid for itself in the first year, and in year two, the profit that it makes can pay for the next Cybercab to be manufactured. Infinite money loop!
They have this gigantic feedback loop, and suddenly Tesla can scale their fleet at a staggering rate—almost instantly. How do you compete with that if you're Uber and Waymo, where Waymo's vehicles cost $1 plus $30,000 to put on roads?
By the way, the answer to that question is you don't compete; you go out of business, aka you won't be around.
I think that we're seeing the early innings of it today with what Waymo is doing. Just speaking of Uber and Lyft, I know Tesla has been very focused on vertical integration. We're building our own direct consumer application.
Do you think there's any chance that to reach the consumer, why wouldn't Tesla start on Uber and Lyft, get people accustomed to autonomous driving, leverage Uber and Lyft, and then just break off once they reach demand?
The reason they wouldn't is because they don't need to. Now, if Tesla was a brand new company and they hadn't already established a great brand, you could make the argument. But just think about this: you either own a Tesla or you're a customer. You can now either put your vehicle into the network or, as a consumer, hail a Tesla Cybercab in the early days.
What do you think you're going to do? You'll be willing to pay a massive premium just for the novelty value of getting into a Tesla Cybercab. You're not going to need to leverage an existing platform.
Now, the other argument would be, "Well, it'll make it a ride-hailing app really complicated." I mean, relative to what Tesla's done, it's a walk in the park. People massively oversell how difficult it is to have a ride-hailing app.
Now, it's not easy, okay? I understand that. But when you look at Tesla's track record of doing incredibly complicated things at lightning speed, there's very little a company like Uber or Lyft, or pick your ride-hailing company, can provide to Tesla.
And wait for it—there's something else. Tesla has something that these companies also have that many people might think Tesla hasn't got access to, which is real-time data about vehicle usage in every city they'll be operating their vehicles.
I'm giving odds of Tesla formally partnering with a company like Uber out of necessity—what about a 0.42069% chance? It is possible, but bro, why?
Well, it certainly seems like that might be Waymo's plan. Waymo started their own service, and then they decided to partner with Uber. I think one of the reasons they might have done that is because they don't have a scaled fleet.
So in order to offer competitive service times to customers, they probably want to supplement with a larger fleet. But the reason for Tesla not to do that—well, you know, the obvious reason for any company not to do that is sharing economics, right?
But, you know, we've actually published a blog that shows that if Tesla were to launch with human drivers today, they could already undercut the cost of Uber slightly because we know from Sam's work that the operating cost of an electric vehicle is already cheaper than a gas-powered car.
And, you know, Tesla has this nice verticalized structure that I think allows a lot of advantages. So I think that because Tesla is this vertically integrated manufacturer, they're not in that Waymo position where they will have to necessarily partner.
I'm not saying it's totally out of the question; I just think it's unlikely. I'd be really surprised, and it's not really Tesla's style. They're more of a do-it-yourself type of company.
So I think what Tasha just said is a 0.0694% chance. Amazingly, we agree. It ties into what Tasha was saying earlier with the drivers being able to put their cars in the fleet.
So you can imagine, right, there's not a steady demand for rides over the day. So if Waymo was going to do this themselves, it's like during rush hour, how many cars would Waymo need to own to meet that demand? Because you don't want to open up a Waymo app and say, "Oh, you know, 17 minutes until a car gets to me."
Right? Then you're never opening the Waymo app; you're just going to Uber. With Tesla and the support of drivers being able to put their vehicles into the fleet, it's like Uber doesn't solve a problem for Tesla if Tesla has that capability.
Just in thinking about Uber and Lyft and what other competitive advantages, I think this applies more to Uber, but they're a full-service platform. They do more than just ride-hailing; they have delivery, they have different segments of ride-hailing.
Do you think about that at all? If you're operating an autonomous fleet, delivery might be hard until Optimus starts working, but I would think that maybe that gives at least more mindshare to an Uber when the consumer goes to think of a ride-hailing app. They're already accustomed to getting delivery from Uber Eats.
That does have some sway with the consumer. Maybe price is a determining factor here, but I'm just curious how you think about it because Tesla's going after one component, but there are other services now tied into the businesses of Uber and Lyft.
So your point is more about the customer network that Uber already has—a full suite offering. I think of it like in my space, you know, we have a company in terms of Spotify, right? Spotify started just music; now they do audiobooks, they have podcasting, and it's like a full-service audio platform.
That's what Uber and Lyft and some of these other services are going after, and that, to me, is a moat that is really hard to break. Once you have the full suite of services on one platform, so if there were a company that were to come out with a streaming service and they just—
Okay, so yeah, I get his point. Bro, one, Optimus is going to happen faster than people think. And two, if the ride-hailing app, which is the big driver of profits for a company like Uber, implodes, the rest of your business is going to get dragged down with it.
Rest in peace, Uber.
So it's almost time, folks. The robot taxis are about to awaken, and while skeptics are becoming slightly less skeptical, I think even most bulls don't fully grasp the scale of this opportunity, the sheer magnitude, the massive disruption that's going to take place, and most importantly of all, how quickly Tesla's fleet will scale and how truly insurmountable their lead.
Again, a reminder: AR Invest's five-year Tesla fair valuation is $2,600 per share, and I think about 90% of that is comprised of the successful deployment of their fleet of Robo-taxis. It sounds outrageous, and this is for 2029. It's soon going to be four years, not five, and I'm a lot less interested in the exact timing and a lot more interested in two questions: one, will Tesla be successful? And two, is their lead truly insurmountable, meaning they will take the lion's share of the market, the revenue, and most importantly of all, from an investor's point of view, the profits?
And I strongly believe the answer to those questions is a resounding yes. So don't say I didn't warn you.
Want more content, early access, and a bunch of perks? Click the links in the pinned comment.
AG1 is awesome! I've been taking it daily now for more than three years. It's a great way to fill nutritional gaps. It's packed full of vitamins, minerals, and whole food source nutrients, plus it has prebiotics, probiotics, and adaptogens to improve gut health, regularity, and help your body handle stress.
I'm always looking for an edge to help me feel and perform my best, which is why I haven't missed a day of AG1 for more than three years. Just try it and see how you feel. Click the link in the pinned comment or head to drinkag1.com/smmr and get yourself a free one-year supply of vitamin D3 plus K2 and five travel packs.
This is what viewers of the channel had to say after trying AG1: "I feel like I have a lot more energy since I started on AG1." "Just got my AG1 in the mail—legit feeling the effects after day three." "Three months ago, I started AG1 and have been enjoying the evenness of alertness and energy that lasts the day. Just started the wife on it too."
You convinced yet? I mean, hey, it's worth trying, right? Click the link at the pinned comment or head to drinkag1.com/smmr.
Or I can keep going. This viewer said, "After about a month on AG1, definitely a lack of fatigue in the afternoon. Pleasant side effect is that my coffee intake has imploded and is almost down to zero."
One more? Yeah, why not! "I honestly feel younger and will be continuing to use AG1. This stuff really is crazy good. I didn't think it would be, but this stuff is awesome."
It really is what everyone is saying. One more? Don't mind if I do! "I've just received my third month supply. I've drank it every day. I have so much energy throughout most of the day. I'm productive. Started a new business, started socializing, refurbished the boat. It's no coincidence. Thank you for your persistence, your integrity, and your insights."
Now look, these are not my words; these are not my testimonials. This is what you guys and girls are saying. Maybe it's 100% placebo effect, but even if that's the case, I think it's money well spent.
Click the link at the pinned comment or head to drinkag1.com/smmr and get yourself a free one-year supply of vitamin D3 plus K2 and five travel packs.