On September 21-22, 2026, security researchers tracked 225 vulnerabilities attributed to Project Glasswing, Anthropic's initiative providing select partners access to the Claude Mythos Preview model. Only one CVE showed confirmed exploitation in the wild, indicating attackers were largely unresponsive to the disclosed flaws.
No BS check run yet — press ⚖ to extract this story's claims and verify them against independent sources.
Of 225 flaws found by Glasswing and tracked by VulnCheck researcher, just one has confirmed exploitation in the wild
Anthropic-linked CVEs pile up, attackers mostly shrug https://www. theregister.com/security/2026/ 09/21/anthropic-linked-cves-pile-up-attackers-mostly-shrug/5298018
Garrity began tracking CVEs attributed to Project Glasswing, Anthropic’s initiative to give select partners access to its Claude Mythos Preview model, shortly after the AI company announced the program in April. https://www. theregister.com/security/2026/ 09/21/anthropic-linked-cves-pile-up-attackers-mostly-shrug/5298018
Anthropic-Linked CVEs Pile Up, Attackers Mostly Shrug https:// packetstorm.news/news/view/437 89 # news