Less Wrong is a community blog devoted to refining the art of human rationality. Please visit our About page for more information.

Stuart_Armstrong comments on JFK was not assassinated: prior probability zero events - Less Wrong

20 Post author: Stuart_Armstrong 27 April 2016 11:47AM

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

Comments (47)

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

Comment author: Stuart_Armstrong 03 May 2016 05:21:32PM *  0 points [-]

Let V be the set of worlds in which X happens. Let W be the set of worlds in which X and Y happens. Since Y is very unlikely, P(W)<<P(V) (however, P(W|message read) is roughly P(V|message read)). The AI gets utility u' = u|V (the utility in the non-V worlds is constant, which we may as well set to zero).

Then if the AI is motivated to maximise u' (assume for the moment that it can't affect the probability of X), it will assume it is in the set V, and essentially ignore W. To use your terminology, u(Z|X) is low or negative, u(Z|X,Y) is high, but P(Y|X)*u(Z|X,Y) is low, so it likely won't do Z.

Then, after it notices the message is read, it shifts to assuming Y happened - equivalently, that it is in the world set W. When doing so, it knows that it is almost certainly wrong - that it's more likely in a world outside of V entirely where neither X nor Y happened - but it still tries, on the off-chance that it's in W.

However, since it's an oracle, we turn it off before that point. Or we use corrigibility to change its motivations.

Comment author: gjm 03 May 2016 05:46:06PM -1 points [-]

Again, maybe I'm misunderstanding something -- but it sounds as if you're agreeing with me: once the AI observes evidence suggesting that its message has somehow been read, it will infer (or at least act as if it has inferred) Y and do Z.

I thought we were exploring a disagreement here; is there still one?

Comment author: Stuart_Armstrong 04 May 2016 09:21:43AM 0 points [-]

I think there is no remaining disagreement - I just want to emphasise that before the AI observes such evidence, it will behave the way we want.