Microsoft Launches New AI Agent Service 'Microsoft Perception' to Combat Cyber Attacks
November 21, 2026 · 2 min read
Also published in Indonesia, हिन्दी, Türkçe, Русский, עברית, العربية, Deutsch, Français, Italiano, Español, Norsk, Nederlands, Svenska, Português (Brasil)
Microsoft has announced the launch of a new service called 'Microsoft Perception' designed to combat AI-driven cyber attacks. This announcement was made during the annual event 'Microsoft Ignite 2026,' which began in the early hours of November 18, 2026, Japan time.
Microsoft Perception is now generally available and integrates red, blue, and green team agents through playbooks to simulate attacks, detect threats, and remediate issues, delivering security outcomes at machine speed while keeping human operators in control.
As AI-driven cyber attacks become increasingly automated, the time from discovery to attack has dramatically decreased, making it difficult for human operators to respond effectively. Therefore, defenders need to operate at machine speed using AI agents to counter these threats.
The service is structured around three distinct team agents: the red team agent, which identifies vulnerabilities from the attacker's perspective before an attack occurs; the blue team agent, responsible for detecting and investigating cyber attacks and managing incidents; and the green team agent, which focuses on fixing vulnerabilities, bugs, or reducing risks.
These three team agents work autonomously in coordination according to a set of procedures known as playbooks, aiming to prevent cyber attacks before they occur and enabling machine-speed responses to any incidents that arise. However, critical decisions will still require human judgment.
The playbook for the green team agent reportedly covers not only Microsoft's platform but also includes AWS and Google Cloud, showcasing a comprehensive approach to cloud-native and serverless environments.