There is an alternative way (but not a replacement) to go about it as well. Rather than reverifying proofs (I mean now ones that have been proved in the literat

There is an alternative way (but not a replacement) to go about it as well. Rather than reverifying proofs (I mean now ones that have been proved in the literature) one could study central features. And whenever a feature breaks, then one can stop to think whether the theorems themselves hold water or not.