Anthropic Grants Outside AI Watchdog Employee-Like Access
Anthropic has granted an external artificial intelligence safety watchdog unprecedented internal access to evaluate its systems, including the unique right to publish findings without prior company approval. This move sets a new transparency standard for the tech industry as developers face mounting pressure to prove their advanced models are safe before public deployment.
Source: Ars Technica