DragonGod comments on AI alignment researchers don’t (seem to) stack

[DragonGod]

AGI Safety Fundamentals is trying to do something that is somewhat similar I think.