Four: Brian Christian on artificial intelligence
Effective Altruism: Ten Global Problems – 80000 Hours
00:00
Climate Change Is More Analogous Than the Same
In inverse reinforcement learning, you operationalize some definition of what you want. It doesn't perfectly contain everything you care about, but it's pretty good. So i think that's basically were at as a civilization. And there's, there's hope in, this is a little bit a dreamy here, but there's hope that something like inverse reinforcement learning could help us get out of that trap.
Transcript
Play full episode
