That's great, but does that particular model also know if/when/that it does not know?
> AlphaProof is a system that trains itself to prove mathematical statements in the formal language Lean. … Formal languages offer the critical advantage that proofs involving mathematical reasoning can be formally verified for correctness.
Edit: To defend my response, the model definitely knows when it hasn't yet found a correct response, but this is categorically different from knowing that it does not know (and of course monkeys and typewriters etc., can always find a proof eventually if one exists).