[Linkpost] OpenAI's Interim CEO's views on AI x-risk — LessWrong