Tesla making some headway in its Texas Robo taxi initiative, now operating unsupervised in the evenings in Austin, according to the Robo taxi trackers, the EV maker competes with Waymo and now Uber. Our next guest says Tesla could be the big winner in the end. You know, when we talk about a a robo taxi, Adrian, what do we mean by that? How do we define robotaxi? And why, Adrian, are so many folks now so excited about this and and and racing to build one?

Robo taxis essentially uh um an Uber without a driver. So the car can be designed differently um because the ergonomics would be different. A lot of the things within the car, um that you you might otherwise want, for example, like a steering wheel and dashboards and stuff, kind of are irrelevant because there's no driver. You just get in, you can organize the car around exactly how you want to sit, want to interact with your family, um and off you go. That's the that's the difference between a car uh and a robo taxi.

And why so much excitement now, Adrian, about this? What's changed? You know, is it the technology, the investment, is it the regulation?

I think it's a gradual change that's been going on for about 10 years to be honest. Um, Tesla have been investing heavily in machine learning, um and they've got a machine learning computer in the car for over 10 years now. So they've been learning about driver behavior, ride and handling, motion control, road conditions, four-dimensional maps, etc for the for the last period of time. With the advent of AI now coming in, the acceleration on learning and the acceleration on the pace of change is getting faster and faster. So you're seeing the iteration cycles just decrease dramatically, uh allowing um Tesla to make a lot more progress in a faster manner.

You describe sort of two different broad ways to build a robotaxi. One with expensive sensors, another with cameras and a brain. Just let's break down those two models for us.

Sure. Well, there's two different approaches. If you look at Waymo, Waymo are using lidar technology. So essentially they buy in a Jag, they um fit about $100,000 or $80 to $100,000 of technology on the roof as you can see in the video. Um it then uses radar as a primary mechanism to identify what's happening on the road. It's a very secure way of doing things, but it basically puts the piece price at over $150,000 each. So those units are very expensive. Then you flip to the robo taxi approach that Tesla are using and they're using cameras as the primary interface point. No li- lidar technology. So essentially the idea or the construct behind that is similar to a human being, where you have sort of weak we we as humans have got weak eyes and a big brain. So throw the cameras as the weak eyes onto the car and use the big brain to process the data on the road and make sure that you can kind of identify what's happening and and go forward like that. So it's a very fundamental approach that's different to Waymo. Um, the benefit for that is you can do a $30,000 car with some $60 cameras, and as long as your tech is good, you're you're you're driving a self-driving car. So the unit economics are dramatically different.

So you you bottom line, Adrian, you'd make the bet that Tesla ultimately is going to be the ultimate winner here.

Yeah, primarily because of unit economics. Yeah? You're going to get more cars out on the road quicker, faster. Um, the cost per mile is going to be significantly lower than Waymo can afford to do. So if you look at the cost per mile for an Uber, it's probably what, $2.50 to $3 a mile. Um if you look at the future model on robo taxis, it could be around 20 to 30 cents a mile.