I would still trust a mathematical argument I can understand over the output of some automated theorem prover. Of course these aren't mutually exclusive, provi

I would still trust a mathematical argument I can understand over the output of some automated theorem prover. Of course these aren't mutually exclusive, provided the theorem prover is simple enough to understand.