OpenAI's evaluator 'off the record' is one-way — and that's likely by design
lxrjl · x · 2026-11-19
Analyzing OpenAI's employee-discussion terms for external assessors, the author flags a key asymmetry: conversations are 'off the record' only in one direction — nothing prevents OpenAI leadership from demanding everything shared with evaluators.
- Given private evaluator–SSC communication was explicitly mentioned, this likely isn't an oversight.
- A genuine intent to foster candid conversations would give staff two-way off-the-record options.
- The most plausible rationale: OpenAI doesn't trust staff not to leak commercially sensitive material, and considers leaks to 3P evaluators significantly more likely.
- The author still welcomes the voluntary pilot and hopes other companies adopt similar schemes.
More from Companies & People
- Shervin pitches a new NVIDIA board for Jensen's return, with Musk and Nadella on the list — inductionheads · 2026-11-19
- Anthropic Employees Debates Slowing AI as Historic IPO Nears, NYT Reports — JeffLadish · 2026-11-19
- NVIDIA CFO's cold expression while addressing Jensen Huang's exit sparks buzz — BecauseCulture · 2026-11-19
- NYT: Anthropic Shows Organizational Fractures Ahead of Historic IPO — kylekabasares · 2026-11-19
- NVIDIA C-suite shakeup: Rene Haas promoted and locked in with a big multiyear package — firstadopter · 2026-11-19
- Cross-coalition 'MisAnthropic Statement' signed ahead of Anthropic's IPO, RSI gates proposed — AndyMasley · 2026-11-19