# Microsoft brings AI vulnerability-hunting tool to government cloud > Source: > Published: 2026-09-08 14:35:00+00:00 # Microsoft brings AI vulnerability-hunting tool to government cloud ## Select agencies and partners can preview MDASH, which uses teams of AI agents to find software flaws and test whether attackers could exploit them. Microsoft is giving select government customers access to an artificial intelligence system designed to find exploitable software flaws, in an effort to bring automated vulnerability hunting into its cloud environment for sensitive government work. The tool, codenamed MDASH, is available in limited preview through Azure Government, a secure cloud computing platform built for U.S. government agencies, Microsoft [said Tuesday](https://www.microsoft.com/en-us/microsoft-cloud/blog/us-government/2026/09/08/codename-mdash-brings-agentic-ai-security-scanning-to-us-government/), adding that both defense and civilian offices are exploring its capabilities without identifying specific agencies. “The tool is uncovering previously unknown vulnerabilities and significantly improving the quality of the underlying code,” Steve Faehl, Microsoft’s vice president and chief technology officer for its U.S. public sector business, said in a statement. The company did not provide examples of those vulnerabilities or say how many had been discovered. The rollout offers agencies yet another way to examine the software supporting their day-to-day work as increasingly capable AI systems create opportunities for both defenders and attackers. Since April, a growing family of advanced AI systems focused on cyber capabilities have rapidly emerged across the tech industry. The expansion has prompted policymakers and intelligence agencies to aggressively rethink how they leverage private sector AI models for cyberdefense and offensive cyber operations. MDASH assigns more than 100 AI agents to examine the same software, with different agents looking for different types of weaknesses, according to Microsoft. A second group then reviews those findings, testing whether a suspected flaw can be reached and exploited. The government rollout comes amid broader interest in using advanced AI for national security work. Top NSA official Tim Kosiba said late last month that the agency [wants access](https://www.nextgov.com/artificial-intelligence/2026/08/nsa-wants-access-all-ai-models-top-official-says/415672/) to a wide range of models and is engaging with frontier AI companies.