Tools AI-Found Vulnerability Register aifv-0067
Register entry
Anthropic says Project Glasswing partners found more than 10,000 high- or critical-severity flaws
- Project
- Glasswing partner codebases (about 200 organisations)
- Finder
- Claude Mythos Preview and Mythos 5 (Project Glasswing)
- Autonomy
- partner organisations running the gated model over their own code
- Evidence tier
- Aggregate claim
- Bounty
- None published
- Confidence
- CONFIRMED
What happened
On 2 June 2026 Anthropic said it was extending Glasswing to about 150 more organisations and that partners had by then found more than 10,000 high- or critical-severity flaws. No identifier list accompanies the figure.
Credit as recorded: Anthropic post of 2 June 2026, as quoted in the YFarmX report of 21 August 2026. Date basis: Anthropic post extending Glasswing, 2 June 2026.
Sources
- YFarmX, Claude Security now scans on Mythos 5yfarmx.com/claude-security-scans-run-on-mythos-5/
One record from the AI-Found Vulnerability Register, maintained by the Security Desk. Data: CSV · JSON ·RSS · CC BY 4.0 with attribution to YFarmX.Tracker updated · 19 September 2026