Mati_Roy comments on Problems in AI Alignment that philosophers could potentially contribute to