#65 – Katja Grace on Slowing Down AI and Whether the X-Risk Case Holds Up
Hear This Idea
00:00
The Importance of Maximizing Faces
I know that this point got kind of some discussion in the comments, especially about what it means to be maximally faced like. So my claim was that the fact that current systems can make extremely accurate looking faces that many people can't distinguish from real faces is evidence that AI systems can also probably do relatively well at learning values as you're just discussing. And so if they're maximizing behavior looks very crazy, that's still a bad sign,. even if we're non maximizing, they could make a thing that was a lot like what you wanted.
Transcript
Play full episode
