Connor Leahy on AGI and Cognitive Emulation
Future of Life Institute Podcast
00:00
OpenAI: A Creative Engineering Approach
Gore: I don't think there's a theory here. It sounds like, ah. Once you do 7.5 micro alignments of RLA-TIF, then you get what you want. And this is mostly from the reinforcement learning from human feedback. If I didn't think these systems were extremely dangerous, I would be an absolute awe of OpenAI.
Transcript
Play full episode
