Have you heard about the AI box experiment?
Eliezer won it twice, but ISTR he said he felt terrible afterwards because he wasn’t sure the other party’s consent was informed enough.
The fact that Eliezer won a zero-sum game isomorphic to one where the opponent declares a winner after talking to him for a period of time is evidence that at least some people do have that level of power.
I don’t get the point of this comment.
Jonil asked for ways to test his manipulation skills, and I offered an example of such a test.
Have you heard about the AI box experiment?
Eliezer won it twice, but ISTR he said he felt terrible afterwards because he wasn’t sure the other party’s consent was informed enough.
The fact that Eliezer won a zero-sum game isomorphic to one where the opponent declares a winner after talking to him for a period of time is evidence that at least some people do have that level of power.
I don’t get the point of this comment.
Jonil asked for ways to test his manipulation skills, and I offered an example of such a test.