Pretty good overall. My favorite posts are about the theory of the human mind that helps me build a model of my own mind and the minds of others, especially in how it can go wrong (mental illness, ADHD, et. al.)
The AI stuff is way over my head, to the point where my brain just bounces off of the titles alone, but that’s fine—not everything is for everyone. Also reading the acronyms EDT and CDT always make me think of the timezones, not the decision theories.
About the only complaint I have is that the comments can get pretty dense and recursively meta, which can be a bit hard to follow. Zvi will occasionally talk about a survey of AI safety experts giving predictions about stuff and it just feels like a person talking about people talking about predictions about risks associated with AI. But this is more of a me thing and probably people who can keep up find these things very useful.
Pretty good overall. My favorite posts are about the theory of the human mind that helps me build a model of my own mind and the minds of others, especially in how it can go wrong (mental illness, ADHD, et. al.)
The AI stuff is way over my head, to the point where my brain just bounces off of the titles alone, but that’s fine—not everything is for everyone. Also reading the acronyms EDT and CDT always make me think of the timezones, not the decision theories.
About the only complaint I have is that the comments can get pretty dense and recursively meta, which can be a bit hard to follow. Zvi will occasionally talk about a survey of AI safety experts giving predictions about stuff and it just feels like a person talking about people talking about predictions about risks associated with AI. But this is more of a me thing and probably people who can keep up find these things very useful.