Microsoft launched its first model built specifically for cybersecurity work on Monday, alongside a new platform that deploys teams of AI agents to automate security operations. The announcements, made at a small event in San Francisco, mark the company’s entry into a market where rivals like OpenAI and Anthropic have already released their own security-focused AI tools.

The model, called MAI-Cyber-1-Flash, is designed to identify hard-to-find vulnerabilities in complex codebases, according to Microsoft. It runs inside MDASH, the company’s existing harness for finding and fixing software flaws. Microsoft claims the model outperforms competitors on a widely used AI cybersecurity benchmark, though it did not name specific rivals in its public materials. Mustafa Suleyman, CEO of Microsoft AI, said the company is shipping the model into production immediately.

The new security platform, Perception, uses what Microsoft calls agentic red, blue, and green teams. Red teams simulate potential attacks, blue teams detect and prioritize existing bugs, and green teams take corrective action. The platform can integrate with MDASH and aims to reduce hours of manual work by multiple specialists to minutes. “We have a fix for all of this,” said Dave Weston, Perception’s lead engineer, adding that the system generates detection rules, posture fixes, and code patches. Hayete Gallot, Microsoft’s vice president for security, described Perception as a way for defenders to “defend against AI with AI at the scale and speed that the attackers have.”

The tools arrive as AI-powered cyberattacks grow more sophisticated, giving defenders new capabilities while also arming adversaries. Microsoft said Perception and MAI-Cyber-1-Flash will enter preview on November 3, joining a crowded field of AI security solutions. OpenAI and Anthropic have each launched similar platforms earlier this year, though the companies have not disclosed the number of customers using them.

Source: TechCrunch