Anthropic gives outside AI watchdog “employee-like” access

The outside watchdog can publish its findings without Anthropic's approval.

Publisher's summary from Ars Technica AI. Read the full article on Ars Technica AI · More from Ars Technica AI