Description
Project Perception is an innovative agentic system developed by Microsoft Security, designed to enhance cybersecurity through the use of specialized AI agents. These agents operate within a multi-agent framework, enabling coordinated defense strategies that are both proactive and continuous. The system includes red agents that simulate attacker behavior, blue agents that investigate incidents, and green agents that focus on remediation and hardening. This orchestrated approach ensures that findings are swiftly transformed into actionable fixes without the need for manual hand-offs at every stage.
Read More
The agents within Project Perception are equipped with comprehensive organizational context, drawing from past incidents, policy decisions, and real-time signals across various platforms including endpoints, identities, clouds, and applications. This allows them to begin their tasks with insight rather than starting from data collection, providing evidence-grade reasoning instead of mere pattern matching.
Human oversight remains a critical component of Project Perception. While the AI agents carry out the bulk of the work, human defenders set objectives and guardrails, ensuring that every high-impact action is subject to human sign-off. This approach is supported by enterprise-grade governance and responsible AI practices, making every decision traceable and replayable.
Project Perception is integrated directly into Microsoft Defender at launch and is expected to extend across other Microsoft Security products over time. The system operates on a consumption-based, pay-as-you-go pricing model, where users pay only for the resources they utilize, with consumption rates varying based on task intensity.
- CategoryAI Cybersecurity Tools
- Pricing ModelPay-as-you-go
- DeploymentCloud
- PlatformsWeb
Project Perception's Core Features
Agentic security with AI agents
Multi-agent workflows
Red, blue, and green agents
Orchestrated defense strategies
Integration with Microsoft Defender
Consumption-based pricing model
Enterprise-grade governance
Responsible AI practices
How to use Project Perception?
Configure: Set up organizational context and policies
Use: Deploy agents for proactive defense
Monitor: Oversee agent actions with human sign-off
Optimize: Adjust settings based on insights and performance
Project Perception's Use Cases
- Proactive Defense
- Incident Investigation
- Security Hardening
- Organizational Context Analysis
- Human Oversight
