Kevin Hartnett

One is the same thing that like all fields are trying to do, which is like, what are the new rules for the road in the age of AI?

Like basically saying, if you use AI when writing a proof, you need to tell us about it.

Like the Math Archive, which is where proofs are posted online before they're published, recently issued a statement that if we see unedited use of AI in your documents,

PDF that you upload, like we see like the metadata from the AI prompt, like you copy and pasted it in there and you didn't even know it was there and you didn't review it.

We're banning you from the platform for a year.

So there's an effort to like set some new rules for the road.

That's one.

And then two, I think they worry that the types of math and incentives to do math, the types of math problems that LLMs are good at

are different than the kinds they care about, but like their own priorities are just gonna be like steamrolled by the rapid pace of progress and the particular kinds of problems that AI is good at at the moment, all the attention and kind of money that flows to them.

I think they're worried the field would kind of get squeezed off and they'll have no say in that direction.

They're trying to have a say.

So, I mean, is this just like a purely defensive thing?

I think there are definitely large elements of math that are worth preserving and that like AI could in a certain way kind of undermine those elements without fully replacing them.

And I think that is like a valid worry.

Like math is like a very rich discipline.

It's the kind of the flowering of like the human mind.

I think there are things to worry about that AI can kind of strip a lot of the incentive and value out of it without fully replacing it.

Oh, I think it's actually entirely that, like, if AI can generate proofs that are really good and we can read them and we can all be super impressed by them, then there'll be no more reason for us to, like, have jobs.

Like, we will be, like, hobbyists, like, great chess players.

I think the anxiety is having been in possession of something for a long time that was very special, that was rare in the human population, great ability at math, and now is generally available.