
Counterarguments to the basic AI x-risk case
LessWrong (Curated & Popular)
00:00
Utility Maximization: A Concept of Goal-Directedness
One well-defined concept of goal-directedness is utility maximization, always doing what maximizes a particular utility function. If you want things to go a certain way, then you have reason to control anything which gives you any leverage over that. This is in serious conflict with anyone else with resource-sensitive goals. Call machines that push toward particular goals but are not utility maximizers, pseudo-agents.
Play episode from 05:52
Transcript


