Grant Demaree comments on Why so little AI risk on rationalist-adjacent blogs?