Zscaler Unveils Zero Trust Platform for Agentic AI

Zscaler’s Zero Trust Agentic AI Platform Unveiled as AI Security Market Nears $8 Billion

LAS VEGAS, USA — June 10, 2026 — At its annual Zenith Live conference, cybersecurity leader Zscaler announced the industry’s first comprehensive zero trust security platform tailored specifically for autonomous, agentic AI ecosystems. The development marks a definitive milestone in addressing the security blind spots created by automated machine-speed workflows.

Securing Autonomous Agents at Machine Speed

Traditional enterprise security perimeters are poorly equipped for ecosystems where autonomous AI entities seamlessly exchange data without human intervention. The newly expanded Zscaler Zero Trust Exchange addresses this exposure by independently managing how AI agents connect, execute tasks, and communicate across hybrid corporate networks.

"Traditional security was never designed for millions of autonomous agents that act and reach sensitive data at machine speed," stated Jay Chaudhry, Chairman and CEO of Zscaler.

The platform architecture focuses on four critical defense parameters:

• Zscaler AI Broker: Secures Model Context Protocol (MCP) and Agent-to-Agent (A2A) pathways to govern data exchanges.

• Zscaler Endpoint AI Security: Identifies and mitigates malicious or misconfigured local AI actions on employee hardware.

• Zscaler AI Access Graph: Maps data pathways and identity interactions across systems to trace data lineage in real time.

• Expanded AI Protect: Discovers shadowed AI tools within internet traffic and actively scans agentic codebases for underlying vulnerabilities.

Market Size Reaches Critical Inflection Point

The rise of agentic networks is rapidly altering enterprise defense priorities. Organizations are transitioning away from perimeter-based security architectures in favor of highly granular control maps designed explicitly for automated tool workflows.

"The introduction of dedicated framework security architectures like Zscaler's platform highlights an immense shift in global deployment strategies," noted an analyst from Next Move Strategy Consulting. "According to NMSC analysts, the foundational Agentic AI systems security market is scaling from near-negligible spending to an estimated $8 billion milestone by 2030, driven by the immediate requirement for cross-platform data visibility and active runtime guardrails."

Redefining the Frontier of Machine Identity

As software agents assume greater administrative autonomy to trigger corporate workflows, security parameters must shift to evaluate the machine identity itself. Industry experts indicate that malicious actors are increasingly looking to leverage unmonitored internal AI pipelines to escalate administrative privileges and compromise sensitive intellectual property.

By isolating agent identities and enforcing continuous verification protocols, the newly introduced architecture ensures that a security compromise within an individual agent framework remains contained, preventing devastating lateral network intrusion.

Source: Network World

Prepared By: Prakhyat Chowdhury