Lex Fridman

I think cellular automata is one of the most incredible, complicated, fascinating.

It feels like an open door into a world we have not quite yet explored.

And it's just a beautiful illustration of that world, the Game of Life, but calling it a game.

Maybe life balances it, because that's a powerful word.

but it's not quite a game.

It's a fascinating invitation to an incredibly complicated and fascinating mathematical world.

I think every time I see cellular automata and the fact that we don't quite have tools, mathematical tools, to make sense of that world, it fills me with awe.

Speaking of a thousand years from now, it feels like that is a world we might make some progress on.

Nevertheless, when you sit back and look and visualize the thing, some little mini cellular automata civilizations are born and die quickly, and some are very predictable and boring, but some have this rich, incredible complexity.

Maybe that speaks to a thing I wanted to ask on the halting problem and decidability.

You've mentioned this thing where if you understand the program deeply, you might be able to say something.

So can we say something interesting about maybe, statistically, how many programs we know something about in terms of whether they halt or not, or what does it mean to understand a program deeply enough to be able to make a prediction?

So 13% of programs, you can say.

So statistically, in the limit, you solve the halting problem.

Computably solve it.

What do we take from that?

Because you didn't solve the halting problem.

That's pretty cool.

That's kind of, I mean, I don't know.

The amount of fields and topics you worked on is truly incredible.