Pfft 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: Pfft 18 December 2010 05:34:20PM 2 points [-]

The source code verifier for friendliness is basically the weakest possible thing a friendliness researcher could hope for.

This is a good observation -- I think the article would be stronger if you added it to the intro paragraph. As it is, I was a bit amused when reading it when you say "this creates a problem", and then mention the problem of the box, while the problem of the verifier seems immensely harder...