Zscaler Reaffirms Fiscal 2027 Outlook as CEO Urges Narrow AI-Agent Access

Zscaler reaffirmed its fiscal 2027 financial outlook as CEO, chairman and founder Jay Chaudhry urged companies to restrict AI agents to specific applications and services instead of granting broad access across corporate networks.

“Don’t trust AI agents,” Chaudhry said. “Give them this much trust for certain applications and services. And that’s it.”

Chaudhry warned that AI agents operating inside corporate networks can scan websites, firewalls, virtual private networks and load balancers for vulnerabilities. Unlike ordinary chatbots, agents can query databases, call application programming interfaces, trigger workflows and move data between systems.

The comments came during Zscaler’s Oct. 6, 2026, investor day, when the company reaffirmed first-quarter fiscal 2027 revenue guidance of $935 million to $939 million, representing approximately 19% year-over-year growth.

For the full fiscal year, Zscaler reaffirmed annual recurring revenue guidance of $4.396 billion to $4.426 billion, implying growth of approximately 16.6% to 17.4%. Revenue guidance remained approximately $3.908 billion to $3.938 billion, or 16.6% to 17.5% growth.

Zscaler also reaffirmed its forecast for first-quarter fiscal 2027 non-GAAP earnings per share of approximately $1.15 to $1.16. The forecast assumes about 170 million fully diluted shares and a 21% non-GAAP tax rate.

The outlook implies slower growth than Zscaler recorded in fiscal 2026. Revenue rose to $3.352523 billion from $2.673115 billion in fiscal 2025, an increase of approximately 25.4%.

Organizations reported varying levels of governance for agentic AI. A survey of 3,235 business and information-technology leaders across 24 countries found that 21% described their organizations as having mature governance. Approximately 80% lacked mature capabilities such as decision boundaries, real-time monitoring and audit trails.

Zscaler is positioning its Zero Trust platform as infrastructure for controlling AI-agent activity. The company says its Zero Trust Exchange processes 750 billion transactions per day and includes tools for AI-agent access controls, agent registries and AI data-access mapping.

Its announced products include an AI Broker for agent communications, an AI Access Graph for mapping identities and data access, and endpoint controls for AI applications.

“The agentic AI era is quickly becoming the most significant opportunity in our company’s history,” Chaudhry said.