Microsoft has unveiled a new artificial intelligence-powered cybersecurity platform designed to help organizations detect software vulnerabilities faster, investigate threats more efficiently, and automate security operations while reducing operational costs.
The announcement introduces Project Perception, a new multi-agent AI security system, alongside MAI-Cyber-1-Flash, Microsoft’s first in-house AI model built specifically for cybersecurity tasks.
Unlike traditional AI systems that rely on a single large language model, Project Perception intelligently selects the most suitable AI model for each security task based on its complexity, urgency, and cost.
Microsoft says this approach enables faster responses while lowering expenses by assigning routine work to lightweight specialized models and reserving more powerful AI systems for advanced investigations.
The platform is scheduled to enter public preview on August 3, beginning with Microsoft Defender before expanding to additional security products.
The company also claims that its updated MDASH vulnerability research system, powered by MAI-Cyber-1-Flash, achieved a 96% score on the CyberGym benchmark, outperforming several competing AI configurations.
Microsoft further says integrating the new model into MDASH can reduce vulnerability research costs by nearly 50%, making enterprise-scale security operations more affordable. However, the company acknowledges these results come from a coordinated multi-agent system rather than a single standalone AI model.
Project Perception organizes its AI into three specialized roles: Red Agents search for vulnerabilities, Blue Agents investigate threats and suspicious activity, while Green Agents focus on fixing security issues and strengthening system defenses.
These AI agents automatically share information and coordinate responses, allowing organizations to detect, analyze, and remediate cyber threats much faster than traditional workflows.
Microsoft says human administrators will remain responsible for approving critical security actions.
The launch comes as competition in AI-powered cybersecurity continues to intensify among major technology companies including Google, OpenAI, and Anthropic.
Microsoft believes future enterprise security will depend on teams of specialized AI agents working together rather than a single chatbot handling every task.
While the company promises significant improvements in speed, efficiency, and cost savings, experts note that organizations will still need strong governance, monitoring, and human oversight before granting AI systems greater autonomy in live production environments.
