Microsoft has announced Project Perception as its new “agentic security system”, which it claims is “designed to help organizations continuously identify, evaluate, and reduce security risk.” Alongside it, the company introduced Microsoft MAI-Cyber-1-Flash – its first cybersecurity-specialized AI model – built specifically for software vulnerability analysis.
“As AI accelerates the speed and scale of cyberattacks, defenders are being asked to secure increasingly complex digital environments with approaches built for a different era. Security teams are often forced to piece together signals, context, and risk insights across vast amounts of data, making it harder to keep pace with emerging threats,” the company said in its press release. “To address this challenge, Microsoft is introducing a new approach that combines security context, specialized AI models, and autonomous agents to continuously identify, assess, and help remediate risk across the digital estate.”
Microsoft Project Perception & MAI-Cyber-1-Flash Model
Project Perception is built around three classes of specialized AI agents that work in coordination to improve an organization’s security posture over time. Per Microsoft, the system is designed to shift security teams away from reactive incident response toward a more proactive and adaptive approach to risk reduction.
They’re color-coded, too. First is “Red agents”, which is responsible for identifying potential attack paths and vulnerabilities before they can be exploited; next, “Blue agents” investigate findings, apply security context, and determine what represent meaningful risk, while “Green agents” take corrective action and actively strengthen defenses across the environment.
Microsoft also introduced a dedicated AI model for this: MAI-Cyber-1-Flash is the company’s first AI model purpose-built for cybersecurity, with a focus on software vulnerability analysis (and requires only 50% of the cost of “leading models”). The company says this model is developed as part of its “broader multi-model strategy” which combines frontier and specialized AI models to handle different security tasks more effectively.
Pokdepinion: Cybersecurity is always a cat and mouse game, but it really start to look like the cats and mice here are less humans and more AI at this rate.
