Hey everybody, Rob Mauer here and today we're jumping back into our conversation with James

Dauma.

This is part two of this interview.

If you missed part one, I'll put the link to that down in the description.

This is from a couple months back.

So we're talking to all things neural networks here, all things FSD beta.

This is before the FSD version 9 release, but I think all the topics are still relevant

here, but just keep that in mind as you are listening today.

So with that said, we're just going to jump right in to where we left off with part

one.

Well, I have tons more questions on neural networks.

I think that's awesome background, but we probably don't need to talk about that all

night.

So getting to where we're at right now with like the FSD beta, I'd love to get a sense

of your sense of the progress, where it tells us at right now, how far they may have to

go, how things are looking, how happy you've been with the progress, those sorts of things.

I've been, so I did get a chance to ride an FSD beta car, and I watched a ton of FSD

beta videos, I'm sure everybody on this channel has, let's hope so, that's part of the homework.

Yeah, I was surprised at how good it was when it first came out, like when the first videos

I saw, I was, I was pretty impressed, and I mean, this is speculation on my part, right?

But I drove the early versions of autopilot, and they were pretty scary, right?

And then I remember when we got to AP 2, like AP 1, when you, when AP 2 first came out,

...