Jason Hall

And before I kind of get to my take here, let's talk about really the difference in what Tesla is doing from anybody else.

And that's really what it comes down to.

It's kind of sensor focused.

It's mostly camera-based, optical sensor focused.

Most of the rest of the industry is using LiDAR, other technologies.

Radar is pretty popular across pretty much everybody.

But what we're finding is that the benefit of doing that is that it's

from a hardware perspective, should be substantially cheaper, even though LiDAR has gotten much, much lower cost.

But I think, again, as much as Musk's had this long history of promising things are going to happen next year, they eventually happen, but it's four or five years later.

I think as much as anything that might be promoting, but also, I think it's largely just how aggressive Musk is.

Eventually, a lot of these things do happen.

But that technical approach that Tesla is taking, we're seeing the challenges of it in real-time.

How much longer it's taken to get outside of Austin is much, much longer than Musk initially promised.

The decision to not use technologies like LiDAR requires the software and the hardware that's doing the processing to do a lot more heavy lifting.

It's clearly been the biggest challenge to scaling up autonomous taxis compared to Waymo, which is definitely the leader in terms of, they have 11 markets they're in, they do half a million paid rides a week.

Now, they're geofenced, they're only in very specific areas.

And Tesla says, look, our goal is to try to get into more places quicker over the long-term.

And we have billions of miles of cars driving autonomously in the wild, and that data is helping inform the decisions that we're making.

But the reality is, it's been a much, much harder process.

The slow expansion of the auto taxi business for Tesla does concern me that its technical approach might be, if not a failure, certainly ends up putting it far behind other competitors that just have a substantial lead.