Rebasing Ethics — AI Alignment Forum