qwant news | Versa applies zero-trust controls to AI agent actions with new MCP architecture - SiliconANGLE
AI generated summary, Read the full article for complete information.
Versa Networks announced a zero‑trust architecture for its Model Context Protocol (MCP) that validates every action taken by an AI agent within its Versa Verbo operations co‑pilot before execution. Integrated into the VersaONE Universal SASE platform, the system checks each AI‑generated step against user identity, role‑based access controls and pre‑defined policies, requiring explicit human approval for high‑risk actions and logging all activity for full attribution. Versa’s senior director of AI, Sridhar Iyer, says this approach is essential to make enterprise AI trustworthy, echoing Gartner’s warning that traditional SASE solutions aren’t built for non‑human identities. The zero‑trust MCP layer, now available in release 23.1.1, builds on Versa’s earlier open‑source MCP Server and adds AI‑driven event correlation, anomaly detection, and guided troubleshooting, enabling enterprises to run AI‑driven operations in production while meeting compliance and security requirements.








