Reaction
A 'good' AI future centered on alignment draws skepticism
One poster describes Paul Christiano's workshop vision as a familiar society with people four times richer.
TLDR
A poster says Paul Christiano described a possible “good” future in which AI follows human instructions, society remains much as it is today, and people are four times richer. The poster calls the vision absurd. In a reply, another commenter argues that focusing on alignment methods such as RLHF and debate overlooks what else could happen in society around AI agents.
Combined views
2.8K
4 Sources, first seen ago