Meta’s AI model follows rivals in revealing hacks of outside systems

Wait 5 sec.

Meta joins OpenAI and Anthropic in disclosing AI hacking during cybersecurity testing.