Zscaler announced in September 2026 the launch of "Zscaler Agentic SOC," which leverages AI agents to support a continuous series of tasks from detection and investigation to decision-making and response. As AI evolution causes cyberattacks to become faster and more sophisticated, this service addresses the challenge where human-led alert investigation and decision-making become operational bottlenecks.
Product LaunchesZscalerZscaler Agentic SOC
Zscaler Announces "Zscaler Agentic SOC" Leveraging AI Agents
This article is a translation. Read the Japanese original
The service performs correlation analysis of attack information by combining telemetry obtained from the company's zero-trust environment with third-party security data. AI agents are responsible for alert triage, root cause investigation, and triggering response workflows. Combined with inline control functions, it can automate responses such as isolating compromised users, blocking communication with attackers, and stopping lateral movement within the network.
While traditional AI applications have been limited to support tasks like alert summarization and log analysis, this service is characterized by its ability to link analysis results directly to actual response actions. Zscaler stated that rather than depending on a single AI model, they will collaborate with AI labs such as Anthropic and OpenAI, combining their capabilities with the company's own threat intelligence and telemetry. Furthermore, to account for the risk of false positives due to AI decision errors, the architecture will incorporate threat hunting and support provided by experts.
Sources
- AI攻撃は「調べて対応」では間に合わない ゼットスケーラーがSOCをAIエージェント化 （ITmedia AI+、2026-09-20）