Forget the AI Slowdown—the Vulnerability Explosion Is Already Happening
AI-assisted bug hunting drives record vulnerability disclosures, with 66,401 CVEs logged in 2026 so far, nearly double last year's pace, straining defenders and open-source maintainers.
WIRED's Kernel Panic newsletter reports that AI-enhanced vulnerability discovery is producing record CVE volumes: 66,401 CVEs had been recorded in 2026 as of mid-September, versus 33,512 by the same date in 2025, according to cve.icu's Jerry Gamblin. Microsoft has patched 974 CVEs this month, Oracle shipped 1,448 patches in July versus 309 in July 2025, and Chrome's two June releases included 1,072 patches, more than the prior 23 major releases combined. Mozilla found 271 Firefox vulnerabilities in a single bug-hunting sprint using Anthropic's Mythos model. Experts warn that discovery now scales with compute while remediation scales with people, risking developers and under-resourced security teams being outpaced.
- Microsoft issued a record 974 CVE patches this month; Oracle shipped 1,448 in July versus 309 a year earlier
- Chrome's two June releases carried 1,072 patches, more than the prior 23 major releases combined
- Mozilla found 271 Firefox vulnerabilities in one sprint using Anthropic's Mythos model
- 66,401 CVEs logged in 2026 versus 33,512 by mid-September 2025, per cve.icu
- Core concern: AI scales discovery while human-driven patching lags, widening the attacker advantage
- Countries
- United Kingdom
- AI models
- Mythos
From the original article
3 min read at wired.com
AI labs are toying with an industry-wide pact to slow development. Meanwhile, widely available AI chatbots are already helping uncover a tidal wave of security flaws.
By Matt Burgess, Lily Hay Newman · wired.com
A short excerpt; the article and its images belong to the publisher. Summary, key facts and assessment above are AI-written from the article and can be wrong.