Home Cyberlaw Ars Technica What happened to OpenAI’s long-term AI risk team?