Vladimir_Nesov comments on AI Sentience and Welfare Misalignment Risk