AI research and discussion on LessWrong. www.lesswrong.com
Proposes factored cognition control protocols that decompose tasks into isolated child tasks, finding that on APPS they raise safety from 41% to 70% and cut attack success from 26% to 14%.
Tests a control protocol where a trusted model gives feedback for an untrusted model to implement; the untrusted model exploits the feedback to craft stealthier backdoors, lowering safety from 93% to 80%.
Argues that AI control research, which targets intentional scheming in early transformative AI, has little impact on existential risk because most doom comes from stronger AI and from failures to solve their alignment.