Bottom line: Versa patents a zero-trust control mechanism for AI agents that validates planned actions and can escalate to manual approval.
The SASE provider Versa has patented a zero-trust architecture specifically developed for the Model Context Protocol (MCP). It validates actions planned by AI agents against security policies before they are executed and can enforce human approval when needed.
Versa, a provider of Secure Access Service Edge (SASE), has patented a zero-trust architecture for the Model Context Protocol. The MCP is a standardized protocol for controlling AI agents and their interactions with external systems.
The patented solution implements a validation system that automatically checks actions generated or planned by AI agents before they are executed. The system compares these actions against configured security policies and blocks potentially risky operations. Depending on configuration and rule sets, critical operations can be subject to manual approval by administrators.
For CISOs, this means a control mechanism over autonomous AI agents that was previously not standardized available. The zero-trust approach demands continuous verification of all actions — regardless of whether they are initiated by humans or by agents. The solution addresses the growing challenge of integrating AI agents into existing security governance while preserving their autonomy.
Source: www.security-insider.de · Published 23 July 2026
Lumi AI News — AI-assisted curation pursuant to Art. 50 EU AI Act. Paraphrase and classification by Lumi News Pipeline v1.7.3.