Microsoft Unveils Its First Cybersecurity Model and Perception Agent System
At a special event in San Francisco, Microsoft announced its first specialized cybersecurity model and a new AI platform. This step is seen as a serious competitive move against the company's main security rivals Anthropic, Google, and OpenAI. This is reported by Techcrunch.com reports .
According to ixbt.com, the newly introduced MAI-Cyber-1-Flash model is designed to find complex vulnerabilities in intricate codebases. The MDASH system, dedicated to identifying and resolving software risks, operates precisely on this model.
New AI-Based Defense Platform
Another important innovation presented by Microsoft is the Perception platform, which launches teams of AI agents to automate security processes. This system directly assists in finding and fixing errors and successfully integrates with MDASH.Company representatives stated that the MAI-Cyber-1-Flash model has shown significantly higher power and cost-efficiency compared to rival models in industry cybersecurity benchmarks. According to Microsoft AI CEO Mustafa Suleyman, the new development recorded leading results in the Cyber Gym main testing ground and is being deployed immediately into production.
AI Against Hackers
Recently, cybercriminals have been observed actively using AI technologies in their attacks. Microsoft Vice President of Security Hayete Gallot noted that the Perception platform allows enterprise defenders to combat AI with AI at the speed and scale of attackers.The Perception platform uses agents in three main directions during operation:
- Red Teams — provide detailed simulations of potential threats and vulnerabilities they might exploit.
- Blue Teams — handle the detection and prioritization of existing bugs.
- Green Teams — take necessary remediation measures against identified bugs.
Comments 0
…