Recent news has caused me to think through some questions about AI alignment, so I collected my thoughts here. While I'm sure a lot of this stuff isn't new, I haven't seen all these ideas presented together in one place. I think that some of the approaches that are used in designing decentralized systems can also be useful in constructing alignment systems, so I've tried to do that here. Anyway, I welcome feedback on my ideas.

New Comment
1 comment, sorted by Click to highlight new comments since:

FYI, these sorts of posts generally get more readership/responses if they copy over the text of the post here.