logoalt Hacker News

striking • today at 5:22 PM • 1 reply • view on HN

This comes up frequently in threads but it's actually a much less impactful issue than you might think. There are actually several entirely disjoint implementations of Lean, each with different bugs; it is incredibly unlikely an agent would be able to exercise the same bug in every single implementation.


Replies

tomjakubowski • today at 6:09 PM

Does a malicious proof need to exercise the same bug in all the engines necessarily? Why couldn't it exercise a different bug in each engine?

➕ show 1 reply