Zscaler is building Zero Trust security for the AI-agent era

$ZS is expanding its AI security platform as enterprises move from traditional apps to autonomous AI systems.

AI agents create a different security problem. They access data, trigger actions, inherit permissions, and communicate across enterprise systems at scale.

Zscaler’s answer is Project AI-Guardian.

The initiative brings together Zscaler’s Zero Trust Everywhere framework and global system integrator partners to help enterprises secure AI agents, autonomous workloads, data flows, and enterprise applications.

The core product layer is AI Protect.

It helps organizations discover shadow AI, map AI usage, assess risk, and gain a 360-degree view across endpoints, traffic, SaaS apps, cloud services, and code repositories.

Zscaler is also adding access graph technology through its planned acquisition of Symmetry Systems.

The access graph maps how human identities, AI agents, applications, and data connect across the enterprise.

That matters because AI governance needs more than user permissions.

Security teams need to know what each agent accessed, why it acted, which identity it used, and which systems it touched.

Combined with Zero Trust Exchange, Zscaler can use access graph visibility to support least-privilege policies, trace data lineage, detect anomalies, calculate blast radius, and contain risky AI behavior in real time.

Zscaler is positioning itself as a control plane for enterprise AI security, where identity, data, applications, and autonomous agents all need continuous Zero Trust enforcement.