EN Anthropic gives outside AI watchdog “employee-like” access The outside watchdog can publish its findings without Anthropic's approval. AIAI agentsAI oversightAnthropicsafety