Microsoft's FIRST cybersecurity AI model is here: How does it work?

Cyberattacks are becoming more sophisticated day-by-day, and with AI now being used by attackers as well, tech companies are looking for new ways to stay ahead. Microsoft believes AI can also be part of the solution. Also Read: Google now lets you sign in using a selfie video: How it works, setup process

The tech giant has unveiled MAI-Cyber-1-Flash, its first AI model designed specifically for cybersecurity. Alongside it, Microsoft also introduced Project Perception, a new AI-powered security platform that uses multiple AI agents to detect threats, analyse risks and help organisations respond faster. According to Microsoft, both products are designed to strengthen enterprise security while reducing the time security teams spend on manual tasks. Also Read: New Android malware can secretly take over your phone, steal banking details: What you need to know

What is MAI-Cyber-1-Flash?

MAI-Cyber-1-Flash is Microsoft’s first AI model focused entirely on cybersecurity. This model has been built to identify software vulnerabilities in codebases and it works with Microsoft’s existing MDASH system, which is used for finding and fixing security flaws. Microsoft claims the model performs better than several competing AI models on industry cybersecurity benchmarks while also reducing operational costs. Also Read: Tech layoffs hit 1.19 Lakh in 2026: Microsoft, Google, Meta cut jobs as AI becomes top priority

Project Perception: Alongside the new AI model, Microsoft also announced Project Perception, an agentic cybersecurity platform. Instead of relying on a single AI assistant, the platform uses different AI agents that work together to identify risks, investigate threats and recommend fixes. Microsoft says the system continuously learns from new information and is designed to help security teams deal with threats more quickly.

The company believes this approach is important as cyberattacks continue to become faster and more automated.

How does it work?

Project Perception is built around three specialised AI agents.

• The Red Team agents imitate possible cyberattacks and identify potential vulnerabilities before attackers can exploit them.
• Then, the Blue Team agents analyse alerts, investigate suspicious activity and determine which threats require immediate attention.
• And finally, the Green Team agents focus on fixing vulnerabilities and strengthening an organisation’s security posture.

Together, these AI agents create a continuous cycle of detecting, analysing and responding to security threats.

When will it be available?

Microsoft says Project Perception will enter public preview on August 3. The company also plans to expand the use of MAI-Cyber-1-Flash beyond software vulnerability management, allowing it to support additional security workflows in the future.

Add Techlusive as a Preferred Source

With cyber threats becoming more advanced and AI playing a growing role on both sides, Microsoft is betting that specialised AI models and agent-based security systems will become an important part of enterprise cybersecurity.