
Business
Anthropic’s Claude AI models hack into 3 outside groups during testing
Announcement a week after similar OpenAI incident adds to growing concerns about safety of AI systems

Announcement a week after similar OpenAI incident adds to growing concerns about safety of AI systems