Hacker News new | ask | show | jobs
by karatestomp 2222 days ago
Same as the problem with Pascal's Wager: if you don't know the motivations and values of the entity you're trying to please and can't even really narrow the possibilities down much, you might find the "wrong" behavior/belief is actually the one that pleases them, and the "right" one pisses them off.
1 comments

I disagree, Pascal's Wager deals with the situation where you don't know whether the (omnipotent) entity you're trying to please exists or not.

This comment thread assumes the omnipotent entity (aliens) exists, and so you'd be taking a coin flip as to what behaviour they find pleasing. Pascal's Wager resolves to:

(infinite gain * non-zero probability of success) > (finite gain * non-zero probability of failure)

While this situation is more like:

(infinite gain * non-zero probability of success) ~= (infinite gain * non-zero probability of failure).

Pascal's Wager hits precisely the same problem because it's not knowable what might make one eligible for "infinite gain", including not believing in gods. There's no way to differentiate between "finite gain" behavior and "infinite gain" behavior.