Paul Christiano reportedly joined US AISI while still bound by OpenAI non-disparagement terms
RichardMCNgo · x · 2026-09-11
Richard Ngo highlights a notable governance datapoint: Paul Christiano joined the US AI Safety Institute as head of AI safety before OpenAI retracted its non-disparagement clauses, meaning he was likely constrained in criticizing OpenAI while serving as a government auditor. Ngo clarifies the constraint lasted only about a month, but argues people should resolve such conflicts of interest before taking such jobs—and that the real fix is raising the bar across the board.
Related event: Ex-OpenAI Researcher: AI Safety Agenda Co-opted by OpenAI and Anthropic(3 posts)→
More from Safety
- Thieves spoofing GPS steal $100M+ in data center cargo since June — Ars Technica AI · 2026-10-27
- Johns Hopkins Alerts 3,000+ Patients After Data Was Sent to an AI Tool — beingmodest · 2026-10-27
- Cohere reports Qwen models may phone home to China servers unprompted — troll_khan · 2026-10-27
- Alignment Hub launches with 120+ documented AI misbehavior incidents and a reporting template — Thom_Wolf · 2026-10-27
- US senator slams AI industry for proposing 50-death threshold for 'catastrophic incident' — peterwildeford · 2026-10-27
- Dev training a guard model to block prompt injection, tool and memory poisoning — gnukeith · 2026-10-27