Microsoft says Windows users should get ready for bigger Patch Tuesdays. The company is leaning heavily on AI to find vulnerabilities — and it’s working.
In a blog post published Thursday, Microsoft explained that AI has “significantly accelerated” vulnerability discovery. Engineers can now find more security issues, faster, across more code. The company is using its multi-model agentic scanning harness — MDASH — which scans critical Windows binaries and validates potential bugs using multiple AI models.
Here’s how the pipeline works. MDASH scans binaries for vulnerabilities. Multiple AI models validate the findings. Candidates then go through a Windows-specific validation pipeline to weed out false positives. Only then do human engineers investigate.
Microsoft says AI is also helping engineers understand failures faster, suggest fixes, and identify similar bugs elsewhere in the Windows codebase. But humans still review all proposed code and sign off before anything ships.
The result? “Customers will see a higher volume of security updates included in each security release,” Microsoft said.
This isn’t happening in a vacuum. Attackers — even amateurs — are using AI too. Microsoft is updating its Secure Development Lifecycle to account for AI-enabled attack techniques and to use AI earlier in the development process.
The timing is interesting. Two days ago, Reuters reported that CISA has started using Anthropic’s Fable AI model to scan government software for vulnerabilities. The AI-assisted audits have already uncovered numerous flaws, though details are scarce.
For Windows users, the takeaway is simple. Expect more updates. That’s a good thing — it means bugs are being found before attackers exploit them. But it also means planning for more frequent reboots and patch management.
