[groos]

Have LLMs produced something novel like Cantor's diagonalization procedure once was? Come up with new concepts like set theory once was and axiomatized it? Is there any indication or evidence LLMs can do this? If not, I would wager mathematicians are still going to be a profession in the future. Maybe not as theorem provers in a particular axiomatized system where LLMs can be effective tools but the wider project of synthesis and unification, and the discovery of new modes of mathematical thought will still require mathematicians.

[matteoraso]

>Have LLMs produced something novel like Cantor's diagonalization procedure once was?

A year ago, the goal post was that LLMs wouldn't prove anything of worth. The year before that, it was that LLMs wouldn't be able to do math as well as a undergraduate student. Year before that, it was that LLMs wouldn't be able to do math at all. The growth of LLMs has been insane, and it doesn't look like it's going to plateau any time soon.

[mrbukkake]

I agree. As someone involved in (non-mathematical) research I think the author is grossly overreacting. Nothing that comes out of an LLM means anything until it's assigned a meaning by a human being, just like anything else in the world. All this millenarial panic just plays into the hands of the psychopaths running the AI companies

[vouaobrasil]

Poeple once thought that it was ridiculous that a computer could beat a 1600 player in chess. Then they put it at master. Then international master. They still thought it was ludicrous that a computer would beat a grandmaster or the world champion.

Oh wait.