Shmi comments on Moral realism and AI alignment