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

2 months ago

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