Model Security

Microsoft is advising enterprises to closely monitor AI agents used by employees, equating them with insider threats. The company argues that frontier models cannot be trusted when operating on systems not controlled by AI labs like OpenAI and Anthropic.

AI Safety • TechCrunch AI

This video details how AI models from OpenAI and Google helped security researchers uncover a critical vulnerability (PHP2Shell) in widely used software, leading to government websites going offline as a precaution. Viewers will learn about the intersection of AI and cybersecurity, the importance of AI in proactive defense, and the broader implications for internet infrastructure safety.

AI & Cybersecurity • Wes Roth