AI Floods Microsoft with Security Holes Faster Than It Can Fix Them

Microsoft is facing a daunting challenge as AI-powered tools are discovering security holes in Windows faster than the company can fix them. The issue has arisen from the launch of Anthropic's Mythos AI model, designed to uncover vulnerabilities quickly.

The testing of Mythos revealed that it can surpass all but the most skilled humans at finding and exploiting software vulnerabilities. Within 31 minutes, Mythos generated its first proof-of-concept exploit for a Windows kernel vulnerability.

As a result, Anthropic decided to launch Project Glasswing, an effort to put these capabilities to work for defensive purposes. Many companies beyond Anthropic are part of the project, including Microsoft, Google, Amazon, Nvidia, Apple, and others.

However, access to Mythos is available to anyone, and it has been finding security holes and bugs far faster than companies can patch them. For now, Microsoft is only patching the most dangerous bugs and holes, with plans to eventually address 'moderate'-severity flaws uncovered by Mythos.