This is a linkpost for my submission to the ALTER AI Alignment Prize. As such, it's 8 and a half pages of pretty unapologetically dense math and/or philosophy the whole way through. It represents my own work, and I stand by the results and thinking in it. As such, I welcome actionable feedback, including checking of my reasoning and typography. I'm also happy to answer questions in the comments here or on Discord.


New to LessWrong?

New Comment