AI finds so many Windows flaws, Microsoft can’t keep up. Should you worry?
Be careful what you wish for. That’s what Microsoft found out recently when it discovered that AI — rather than making Windows more secure by helping the company close security holes — could help hackers find flaws faster than Microsoft can fix them. It’s put Microsoft in a serious bind. Should the company devote a potentially massive amount of resources to fix every AI-unearthed security issue…
Microsoft recently discovered that AI, specifically Anthropic's Mythos model, can uncover security vulnerabilities in Windows much faster than the company can fix them. This has put Microsoft in a difficult position, as it must decide whether to prioritize patching all discovered issues or focus on the most critical ones first. The company is facing criticism from the US government and Congress, who fear that frequent hacks could put national security at risk.
Microsoft has already patched 622 bugs in a single Patch Tuesday release, but experts warn that this triage approach may be outdated in the age of AI, as low-level flaws can combine to create high-severity issues. If Microsoft fails to address all the vulnerabilities quickly, they could become a target for AI-powered hackers, leading to severe consequences for businesses, governments, and national security.
Written by urgent.news from Computerworld's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.
