Back
Notes
METR and Redwood’s high-access audit of OpenAI’s internal monitors
AI Safety
Low present catastrophic risk is not proof of control
AI Safety
Who can prove and stop third-party agent spillover?
AI Safety