Gunnar_Zarncke comments on Naive Hypotheses on AI Alignment