Santa Clara, Calif. – May 21, 2026 — As enterprises race to deploy agentic AI, a fundamental trust problem has emerged. AI agents rarely execute single, transparent actions — a single user prompt can trigger multiple underlying steps across critical network and security systems, many of them invisible and unverifiable to the operator. Compounding the risk, AI systems can misinterpret intent or generate unintended actions. Without visibility and validation at every step, enterprises are forced to choose between AI productivity and operational security.
Versa, the global leader in unified security and networking, today introduced a patent-pending Zero Trust architecture for the Model Context Protocol (MCP), applying Zero Trust principles to AI execution. With this approach, no AI action is implicitly trusted – every agent-generated step is validated against user identity, role-based access controls, and system policies before execution, with explicit human validation based on administrator-defined policies.
As Gartner noted in a recent report, “AI has introduced a new, high-volume class of digital users in the form of agents that traditional SSE/SASE Platforms were not built to secure.”
The Zero Trust MCP architecture is delivered within Versa Verbo, the company’s AI-powered operations co-pilot, and is integrated with the VersaONE Universal SASE Platform. Together, they deliver Zero Trust AI execution with AI-powered event correlation, anomaly detection, and guided troubleshooting, enabling enterprises to adopt AI-driven operations in production environments without sacrificing control.
Administrators define policies in advance that determine which agent actions execute automatically, which require human approval, and which are blocked — based on user identity, role, system context, action type, and risk level. Every approved action is logged with full attribution, giving operators visibility, auditability, and policy-driven control without creating bottlenecks in AI-driven operations.