So... Doing a complete verification of a CPU in a strict sense is impossible. These days we are relying more and more on computational proofs for math and sci

So... Doing a complete verification of a CPU in a strict sense is impossible. These days we are relying more and more on computational proofs for math and science research. Therefore - what can we say about the solidity of our scientific knowledge going forward?