Shoshannah Tekofsky comments on Naive Hypotheses on AI Alignment