Robotaxi is coming
10:23 Share

Hey, it's Dave. I was just watching Cathie Wood on CNBC explaining her five-year price target for Tesla, which is about $2,600 per share by the end of 2029. It's mostly predicated on the assumption that Tesla will be able to release and operate a robo-taxi platform, and that that platform will have software as a service margins. Now, a few questions come up when thinking about the possibility of a robo-taxi network in the next five years and its impact on Tesla's stock price.

I think probably the biggest question mark will be how quickly will the FSD or Robotaxi rollout be? Now there are several reasons why most people will be biased to assume that a Robotaxi rollout will be quite slow. By the time Robotaxi starts rolling out, it could take, let's say another five or ten years to really grow into a strong Robotaxi network.

Among these reasons that people might make this assumption is that Tesla has taken quite a long time to get to this place where their FSD software is still not completely Robotaxi capable.

In the past, Tesla has vocalized expectations that robotaxis would come earlier and those expectations haven't materialized yet. Because of these past statements that have been made, how long it's taken and how long FSD has taken to develop, most people probably are biased into thinking that robotaxis will come later than expected. And once they come, it'll be a slower rollout than expected as well. In other words, people are generally cautious about over expecting regarding robotaxis.

Another reason why people could be cautious in terms of how fast robo taxi rollout could happen is the example of Waymo. Now Waymo is rolling out their autonomous taxi service in cities like San Francisco and Los Angeles and a few other places.

However, their rollout has been very slow and when people look at that, they can compare it with Tesla and think Waymo is moving slow, their rollout is slow, therefore Tesla's rollout will probably be slow as well. However, I'm starting to question some of these biases of a slow robo taxi rollout. Now there are a few reasons why I think it could be possible that robo taxis could roll out surprisingly quicker than what people expect.

Now the first reason is that Tesla has a huge fleet of millions of vehicles already. So in every urban city, let's say across the US and in many countries around the world, Tesla has thousands if not tens of thousands of vehicles in and around major cities of the world.

When FSD software becomes capable enough where it's significantly safer than humans, and when Tesla is ready to launch their Robotaxi platform, a lot of these cars can be immediately eligible and be placed in the Robotaxi network. You can have literally millions of cars placed on the network very quickly.

The second reason I think we might be underestimating the speed of the Robotaxi rollout once it's released is people think that there's going to be a lot of complications to enroll your Tesla in a Robotaxi network. Now, it's never been done before where you can add your Tesla to a Robotaxi fleet for a week and take it off or add the Robotaxi fleet overnight and take it off or

Or let's say you go to work and you just want your car to be in the robo taxi while you're at work. This type of platform where you can share your car in an autonomous fashion has never been done before. So there are obviously a lot of question marks.

However, Tesla has been thinking through these issues for quite a long time. I remember back in 2019 at Investor Day, I was in attendance and one of the employees was showing us the ride-hailing app that Tesla had at that time. They had already developed this five years ago. And Tesla has been thinking about this Robotaxi platform for many years. And I think this is one of the main reasons why we have Tesla Insurance. Now, of course, Tesla wants Tesla Insurance because they want to keep insurance companies in check.

They want to offer a competitive insurance option so that other insurance companies don't charge too much for Tesla's vehicles. A second and perhaps maybe more profound reason why there is Tesla insurance is because it could allow instant deployment of consumer vehicles onto the Robotaxi fleet immediately. Let's say a person wants to add their Tesla vehicle to the Robotaxi fleet for a day.

Typically, insurance companies might get involved and say, hey, you have to enroll your car with insurance or Tesla would have to enroll with that company's insurance or something. It would be a mess and very complicated. However, if Tesla has their own insurance company in that state, they could just say, hey, we're going to cover the insurance of every robotaxi, every consumer vehicle that's enrolled in a robotaxi platform.

Now you might say, oh, that would be too pricey or too expensive. But if you think about it, the only reason why Tesla would release their robo taxi platform is if their vehicles and their robo taxis

are significantly safer than a human. In other words, it will have a much lower accident rate than human driven cars. So the insurance costs will be much less. So it'll be actually cheaper for Tesla to insure their robotaxis than consumer driven vehicles. And I don't think it would add that much liability to Tesla at all. Further, they could subsidize those insurance costs and just add them into the take rate that Tesla charges as a robotaxi platform.

So let's say a person pays $10 for a ride, Tesla might use a portion of that to cover for insurance costs. Overall, I think Tesla is thinking proactively and thinking ahead. And this is probably one of the main reasons why they're launching and have launched Tesla Insurance. Basically, Tesla has created an insurance company

Actually many subsidiaries in different states. Tesla is ensuring that consumers will be able to seamlessly add their vehicles to the Robotaxi platform when they open it up. Another reason why I think we might question a slow rollout bias is the nature of AI improvements and the nature of driving. Once unsupervised, FSD becomes much safer than a human

then it's not just about, let's say, opening up a robot taxi platform. And what this means is that this FSD software will only get much safer and better over time. Considering the pace of AI improvements recently, every six to 12 months, FSD would be radically more safer than it was before.

So once Tesla is able to pass this critical point where FSD becomes an unsupervised FSD becomes safer and much safer than a human driver, then it can just accelerate into this never ending cycle where it gets safer and safer and amazingly safe. When you think about that and how it affects Robotaxi rollout, it could actually speed up the rollout because the rollout first only happens when the FSD software is safer and significantly safer than human.

And at that point, it doesn't just increase or improve incrementally, rather it actually improves radically fast and that could fuel a faster adoption of robotaxis as demand increases as well. Now, if we look out five years from now, how realistic is it to think that Tesla will have this huge robotaxi platform operating?

I think it really depends on when that crucial crossover point is. In other words, when will unsupervised FSD become significantly more safer than humans? And when can Tesla roll out this Robotaxi platform? Now, if they're able to do this within, let's say, the next one to two years, then they have another three or four years before that five-year mark to really grow that Tesla Robotaxi network into something really big within, let's say, the five-year mark.

By having millions of cars already in major urban cities and by having Tesla insurance already available in many states so people can just add their cars seamlessly to network. And then you've got AI, which is just radically improving the capabilities of FSD. These are all a lot of good signs.

that show that once Tesla is able to release the Robotaxing network, it could grow very fast and very soon. And it could actually bring in a significant amount of revenue and profit.

However, the question still remains: Can Tesla really release a robotaxi network, an unsupervised, let's say, robotaxi network within the next one or two years? And that's the question to be seen. And a lot of this will depend on how fast Tesla is able to improve their current FSD models. How long does it take for Tesla to make significant improvements?

So far we've seen that, let's say every few months Tesla is making decent improvements to upgrading their FSD performance, which is good news. But going forward, can Tesla keep up that pace? Can they actually improve the pace of improvements? We also know that Tesla is going to be starting training on their hardware for inference trip right toward the end of this year. So next year in 2025, Tesla will be really maximizing performance on hardware for.

And then in 18 or 24 months, we got AI5 or their fifth generation chip coming out and that's going to boost capability immensely as well. So you have a lot of planning Tesla is doing to exponentially boost the capability of FSD. So you've got the car's inference hardware, right? Going through generations of changes and improvements. You've got training hardware. Tesla's bringing in a ton of compute, right? To really boost training their models.

Tesla's bringing in tons of video data with their millions of cars on the road. And there's many positive signs that Tesla is really prioritizing this. However, it's still a bit unclear to me exactly when that crossover point happens. When does unsupervised robotaxis become viable and viable enough that Tesla can release this robotaxi platform?

I would imagine Tesla might release some type of pilot robotaxi program first where it's more of a limited area and maybe they could access cars remotely if they get stuck or in trouble and this would allow Tesla to kind of test out the waters a bit and this could happen a bit earlier than let's say that true crossover point where unsupervised robotaxis are really significantly much safer than a human.

Anyways, I think we'll have signs before all of this happens. It's going to be an interesting and exciting next few years, to say the least, as we monitor and see before our own eyes how Tesla pushes forward with FSD software and robot taxis. And it's definitely going to be a game changer if Tesla can really deliver with this platform of autonomy. All right, hope this is helpful. We'll see you guys in my next video. Thanks.