Vendor Strategy
Important
High
90% Confidence
Cisco Launches Agentic AI Zero Trust with Identity and Behavior Monitoring
Summary
Cisco extends its zero trust architecture with a security solution for AI agents, featuring agent discovery and identity management, granular access control, and real-time behavior monitoring, using identity-aware and semantic analysis for non-human entity security.
Key Takeaways
Cisco announced the 'Agentic AI Zero Trust' solution, integrating identity, access, and behavior monitoring. Key features include automated discovery and registration of AI agents, MCP servers, and tools, creating a centralized inventory and mapping to human owners for unified authentication; enforcement of least privilege policies via MCP gateway, controlling accessible services and operations with identity-aware, time-limited credentials; real-time evaluation of agent interactions in APIs, MCP servers, and enterprise systems, using semantic analysis to detect anomalies, policy violations, or sensitive data access, and blocking before propagation. Aims to address fragmented ecosystems, inconsistent policy enforcement, and dynamic non-deterministic operations in AI agents, with design focus on identity and intent-based contextual control at access points via SSE platform.
Why It Matters
Cisco extends zero trust to AI agents, enhancing non-human entity security control, potentially driving industry toward identity-centric security architecture and impacting enterprise AI deployment and competitive landscape....