Jeffrey Ladish comments on Thoughts on the OpenAI alignment plan: will AI research assistants be net-positive for AI existential risk?