Anthropic lets an outside watchdog act like an employee
Anthropic gave an independent AI safety watchdog the same level of access as its own staff. The watchdog can now publish what it finds without needing Anthropic's approval first.
Why it matters & sources
Why it matters: This kind of outside access could become a model for how AI companies prove their systems are safe.
Sources: arstechnica.com