Vladimir_Nesov comments on “Benevolent [ie, Ruler] AI is a bad idea” and a suggested alternative (not author)