Comment by juvvel

I don't see why these things should be mutually exclusive. You can start prescribing the drug (= working with the AI-generated math proofs) while still trying to understand how it works exactly. Because the explanation might prove invaluable. I don't think anyone would be object to using an AI-proven theorem to further their work if it helps, but a scientist would surely object to the notion that we should just stop working towards understanding. Asking "but why though" is literally a scientist's job.