Reaction
Do existing AI systems have behaviorally relevant long-term preferences over the state of the world?
A user asks about long-term AI preferences as a question of behavior, without proposing an answer.
TLDR
A user asks whether existing AI systems have long-term preferences over the state of the world in a behaviorally relevant sense. The post poses the question without offering a conclusion.
Combined views
1.6K
1 Source, first seen ago