Microsoft’s MDASH AI System Finds 16 Windows Flaws Fixed in Patch Tuesday
In order to enable vulnerability detection and remediation at scale, Microsoft has introduced MDASH, a new multi-model AI-driven system that select clients are testing as part of a limited private preview.
Multi-model agentic scanning harness, or MDASH for short, is a model-agnostic solution that leverages custom AI agents for various vulnerability classes to independently find, validate, and demonstrate exploitable flaws in intricate codebases such as Windows.
According to Taesoo Kim, vice president of agentic security at Microsoft, the harness coordinates over 100 specialized AI agents across an ensemble of frontier and distilled models to find, discuss, and verify exploitable defects end-to-end, in contrast to single-model techniques.
MDASH is designed as a "structured pipelin...