
Microsoft announced its new AI-powered cybersecurity model, MDASH, which the company claims outperformed Claude Mythos and GPT-5.6 Sol in recent security context tests.
The system is designed to provide advanced threat detection and automated defense, leveraging perception-based context to identify vulnerabilities in real-time.
Developed by Microsoft's Israeli cyber team, the tool is part of a broader initiative to integrate agentic AI systems into enterprise-level security infrastructure.