Zscaler (NASDAQ: ZS) on September 9, 2026 launched Zscaler Agentic SOC, a new AI-native security-operations platform available globally at launch. Announced at the Goldman Sachs Communacopia + Technology Conference in San Francisco, the offering deploys specialized agents that handle alert triage, root-cause investigation, verdict assignment and response workflow — the four choke-points where human SOCs get overrun by AI-driven attacks.
Anatomy Of An Agentic SOC
Agentic SOC fuses Zscaler's zero-trust telemetry — the Zero Trust Exchange inspects more than 500 billion transactions per day — with third-party security data, proprietary threat intelligence and models from Anthropic and OpenAI. Agents don't just summarize incidents — they investigate, correlate signals across identity, cloud and endpoint, and trigger contain-and-remediate workflows on their own, with human review for high-impact actions.
Beyond SASE
The launch marks Zscaler's most aggressive move beyond its zero-trust and SASE roots. AI-driven attackers now generate phishing, exploit chains and lateral-movement traffic faster than any human SOC can keep up. Zscaler is betting that the answer is symmetrical: put fast, cheap AI agents on the defenders' side too, wired directly into the same policy engine that already sits between users and applications. It's the same thesis Cisco followed when it bought WideField for Splunk in July and that CrowdStrike codified in SafeMind earlier this month.
Available Now, Priced For Enterprise
Agentic SOC is generally available worldwide at launch, sold as an add-on to Zscaler's existing security stack. Zscaler executives told analysts the platform is designed to "scale human expertise" rather than replace tier-1 analysts, but the operational math — millions of alerts per week per large enterprise — points squarely at consolidating SOC vendor spend into Zscaler's cloud.
Reporting based on coverage from Zscaler, GlobeNewswire, Help Net Security, Manila Times and StockTitan.