Researcher slams OpenAI's oversight as insufficient, rating it barely level 2
davidmanheim · x · 2026-07-22
Researcher David Manheim posted on LessWrong criticizing OpenAI's current oversight mechanisms as highly insufficient. He noted that while active monitoring with the ability to pause sessions is a positive step, it falls far short of providing meaningful oversight.
Referencing his own oversight paper, he explained what is actually required for proper governance. Based on his paper's criteria, he estimates that OpenAI's current practices are only operating at level 1.5 or 2.
More from Safety
- Prompt turns Claude Code into an iPhone backup forensics tool exposing 30-day 'deleted' photo copies — alex_verem · 2026-10-27
- Anthropic reveals it bans customers worth 5-7% of revenue every 9 months — Polymarket · 2026-10-27
- US chases AGI dominance while China, EU and ASEAN treat AI as public infrastructure — BubblyExtent1693 · 2026-10-27
- Agent Field Kit launches $29 security runbooks for production AI agents — alvelda · 2026-10-27
- DeepSeek Safety Eval Finds "Over-Attribution": Model Mistakes Ordinary Tasks for Tests — Much_Boysenberry9124 · 2026-10-27
- Anthropic IPO timing fuels claims anti-open-weights panic is self-serving — mark_k · 2026-10-27