Video
Unknown · 0:00
In this short talk, an AI security researcher argues that securing agentic AI is the main thing still blocking real-world deployment. He proposes a three-part approach: red-team the agent to find risks, build safety i...
Unknown · 0:00
In this short talk, an AI security researcher argues that securing agentic AI is the main thing still blocking real-world deployment. He proposes a three-part approach: red-team the agent to find risks, build safety i...
Redirecting...