Robotaxis Are More Than a Decade Away, Says Luminar’s CEO Austin Russell

(To receive weekly emails of conversations with the world’s top CEOs and business decisionmakers, click here.)

Austin Russell became the world’s youngest self-made billionaire back in 2020 at age 25. And he did it by betting against the hype.

The $2.8 billion company he founded, Luminar Technologies, makes light detection and ranging systems (LiDAR) for cars—basically high-tech laser arrays that help computers “see” the world around them in 3D, which many technologists believe will be essential for making cars that can drive themselves. The promise of driverless cars generated an enormous amount of hype over the past decade, with investors pouring billions of dollars into companies that promised to make steering wheels a thing of the past. Russell, however, says he thought getting rid of drivers would be harder than people realized, and he focused his company, which he founded when he was just 17, on deals and technology focused on augmenting human drivers, rather than replacing them.

He may have been right. The autonomous vehicle industry is facing rough headlines about its failure to deliver on its optimistic promises. Russell, meanwhile, says he’s sitting pretty, with deals to deploy his LiDAR technology commercially with partners like Volvo and Nissan. TIME sat down with Russell, who is a TIME 100 Next honoree, to talk about the state of the industry, driver safety, and another notable driverless car contrarian, Elon Musk.

This interview has been condensed and edited for clarity.

(For coverage of the future of work, visit TIME.com/charter and sign up for the free Charter newsletter.)

You’ve gotten a lot of attention your age. What’s it been like running Luminar at such a young age? Have you learned anything in doing that that you think an older person might not?

Well sometimes it’s just as important what you don’t know or aren’t jaded with, as what you know. But that said, Luminar wasn’t always a sizable company. It obviously started out quite small when we first founded it. But at the same time, there’s no question that building a technology that can be fundamentally differentiated and seeing that 10 or 100 times improvement on what’s existing is very, very challenging. But frankly maybe even more challenging is actually being able to build a great business. It takes a lot more than a great technology to build a great business at the end of the day. The kind of operational skill set that you have to have is not one that can be taught in any traditional degree program, or really from anyone or anything. It’s trial by fire. It’s either sink or swim. Unfortunately, the vast majority of new companies end up sinking. It’s the few that end up swimming. And you have to continue to do so time and time and time again, as you scale it up.

You’ve said that you’re an autonomous vehicle skeptic. What’s the state of the industry from your perspective?

It’s maybe ironic-sounding because I’m in the industry, but I did describe myself as the chief autonomous vehicle skeptic. I’ll try not to take credit for things that are mostly luck, but I don’t think this one is. The huge contrarian bet that I made early on was that [autonomous driving] was less so going to be realized in any reasonable timeframe via what many had assumed would be robotaxi-type urban autonomous vehicles without any driver at all. That problem was just massively complex, and I saw that people were just dramatically underestimating the significance of that challenge by at least an order of magnitude—realistically, multiple orders of magnitude. Even today, it’s still not within grasp. There’s no commercially viable business that’s centered around all that. Not that it can’t exist —it can and ultimately will—but there was this assumption that it was going to all be solved in a couple of years. Everyone had said that by 2021, we’re gonna welcome our robot overlords, and they’re gonna be driving us around everywhere, and car ownership will be a thing of the past. Obviously, that didn’t happen.

That was the whole reason why, in the beginning, we were focused on the existing multi trillion [dollar] a year consumer vehicle industry. The objective isn’t just about replacing the driver. It’s about enhancing the driver. It’s about actually making the vehicle safer and saving the driver time. That’s what can make all the difference. From a market standpoint, we made the right bet, and that’s how we ended up where we are today. We effectively have more major commercial wins than any other LiDAR company, and probably any other autonomous vehicle industry company. We won most of the game.

Read More: Meet the 2022 TIME100 Next

Elon Musk is another notable contrarian. He’s been pretty skeptical about the need for LiDAR. What’s your counterargument to that? And do you think that Tesla’s is on the path to full self-driving?

There’s nothing wrong with the fundamentals of what they’re trying to do. The only discrepancy is the advertising of what it actually is, and some of the manipulation [of] consumers, which is pretty questionable. Calling it ‘self-driving’ at all, much less ‘full self-driving’ is a fundamentally inaccurate statement. They have a great assisted driving system. There’s so much more room for improvement from an assisted driving standpoint. From an autonomous standpoint, if you want to be able to get to a point where you’re truly autonomous, that’s where LiDAR comes into play.

The reality is that cars, including even the most advanced Teslas, the vast majority of the time will not come to a safe stop for an unknown object, or person, or whatever it is in front of you if you’re traveling at a reasonable speed. It can sometimes reduce the severity of the collision, but preventing it altogether is another story. That’s effectively what we’re doing, because we have a ground truth understanding of what’s going on. The whole concept of how you can utilize this technology is effectively moving toward the ‘uncrashable car.’ That’s something that no one is immune to, whether it be Tesla or Toyota or Volvo or whoever

So the reality is that the uncrashable car is the first benchmark you have to get to, and self driving comes after that?

Exactly. And the reality is that people have been thinking about it the other way around and trying to skip over steps when the capability just isn’t there. It’s not off by 20%. It’s off by a factor of 10,000. The hard part is autonomous systems have to be inherently perfect. When you have a human driving you only need to have the system take over when the human makes a mistake, but it’s not that frequent. Autonomous systems make silly mistakes all the time. Our LiDAR makes it a hundred to a thousand times easier, and it’s still hard, even with that. That’s why trying to do it without it is a joke.

So if you had to put a year on it, what’s your best guess of when people will be getting in self-driving robotaxis?

When it comes to [having] our technology on consumer vehicles, it’s literally right around the corner. We’re talking months. That will dramatically improve safety and enable certain autonomous capabilities starting on highways and expanding there on out. When it comes to robotaxis and truly full self-driving systems at any appreciable scale: well into the 2030s. If I were to pick a number I’d call it 2035. That would be my over-under on how that will materialize starting out. I think by 2050 we would see it more widespread. Super promising, it’s going to be hugely valuable. But the question is, what’s the right path to be able to get there and to be able to see the technology through?

You have talked about this “100, 100, 100” vision. Can you tell me about that?

The holistic goal behind everything that we do at Luminar is our 100-year vision, which is to save as many as 100 million lives and 100 trillion hours out on the road over the next 100 years. I think that this is going to be one of the most impactful things that we could do from a societal standpoint. For people between 1 and 44 by some metrics it’s the leading cause of death—car accidents. [That’s a] totally solvable problem, we just need the technology to see it through. Everything that we do ultimately all ties back to that big picture all around saving lives, and saving people time. And it also ties in to the overall notion that you don’t have to replace the driver from day one to be able to already start having a huge lifesaving impact. That’s the north star we’re centered around.