Dagon comments on Why the concept of AI alignment as it is currently formulated is morally troubling