Zoom's new guardrails catch deepfakes and runaway AI agents
Zoom just launched a suite of enterprise AI governance and security tools at its annual Zoomtopia conference. The platform adds centralized controls to oversee AI agents, custom guardrails to block sensitive data leaks in outputs, and real-time deepfake risk detection to flag synthetic audio and video during calls. Why it matters: As AI agents take on real work, enterprises need visibility to prevent data leaks and fake interviewees.