Anthropic’s Project Glasswing uncovers over 10,000 software vulnerabilities using AI
Anthropic's Project Glasswing has identified over 10,000 software vulnerabilities using its AI model, Claude Mythos Preview. This includes thousands of zero-day flaws that had gone undetected for years, with only a small number of patches implemented so far. The initiative highlights the disparity between discovering vulnerabilities and the industry's ability to fix them.
- ▪Project Glasswing uncovered over 10,000 high- and critical-severity software vulnerabilities.
- ▪Among the findings are a 27-year-old vulnerability in OpenBSD and a 16-year-old bug in FFmpeg.
- ▪Fewer than 100 confirmed patches have been deployed for the identified vulnerabilities.
Opening excerpt (first ~120 words) tap to expand
Anthropic’s Project Glasswing uncovers over 10,000 software vulnerabilities using AI Claude Mythos Preview found thousands of zero-day flaws in critical software, including bugs that went undetected for decades, but fewer than 100 have been patched so far. Share Add us on Google by Editorial Team May. 23, 2026 window.sevioads = window.sevioads || []; var sevioads_preferences = []; sevioads_preferences[0] = {}; sevioads_preferences[0].zone = "01f21ccf-2092-46b1-9ac7-8c44cc782e0f"; sevioads_preferences[0].adType = "native"; sevioads_preferences[0].inventoryId = "c5700508-581b-472c-8fdd-a931cdbfc8e1"; sevioads_preferences[0].accountId = "1e47efc1-ec2d-4fca-a8b9-354e249e5095"; sevioads.push(sevioads_preferences); Anthropic’s restricted AI model just did in weeks what human security…
Excerpt limited to ~120 words for fair-use compliance. The full article is at Crypto Briefing.
Discussion
0 commentsMore from Crypto Briefing
