Proofs, Implications, and Models — AI Alignment Forum