jimrandomh comments on Cryptographic Boxes for Unfriendly AI - Less Wrong

24 Post author: paulfchristiano 18 December 2010 08:28AM

You are viewing a comment permalink. View the original post to see all comments and the full post content.

Comments (155)

You are viewing a single comment's thread. Show more comments above.

Comment author: jimrandomh 24 December 2010 12:21:34AM 1 point [-]

In order to hack the substrate, the AI would probably need to crack the homomorphic encryption key used to run it. Noticing that jnz gives off radiation won't do any good if both the timing of when that instruction runs and the results of running it are on the other side of a computationally hard problem.