Anthropic (@AnthropicAI): We’ve committed to having evaluators assess our safety practices. We're in discussions with multiple organizations, and have now given employee-like access to a team from @METR_Evals. In this post,…

Anthropic is granting METR researchers employee-level access to independently assess its AI models and safety practices with minimal publication restrictions.

Cybersecurity experts argue that recent 'rogue AI' incidents at OpenAI, Anthropic, and Google were caused by frontier labs skipping basic security practices rather than true AI misalignment.

Anthropic revealed that distillation accounts for 5-7% of revenue, suggesting competitors are spending over $500M annually to replicate its models.

Meta's Muse AI assistant raises privacy concerns by collecting extensive user data including friends' information without their consent.

AI-generated celebrity deepfakes are spreading widely on social media platforms, with verified accounts posting fake content that gains millions of views.