What's the deal with RLHF?

Getting here: Enter the lobby at 100 University Ave (right next to St Andrew subway station), and message Giles Edkins on the meetup app or call him on 647-823-4865 to be let up to room 6H.

Reinforcement Learning through Human Feedback (RLHF), and related techniques, have been very important in aligning large language models with the goals of the companies that are creating them, and presumably with wider human values.

But how does this technique work, and what are its limitations? Is alignment "solved"? Find out this week!