Oliver Daniels comments on [Linkpost] Theory and AI Alignment (Scott Aaronson)