Bitcoin Red Team Scans Entire Ecosystem with Chinese AI Models

The Bitcoin Red Team is using Chinese AI models to scan the entire Bitcoin open-source ecosystem for security flaws. The team, led by pseudonymous developer Calle, combines AI tools with human review to examine wallets, Lightning applications, software libraries, and other projects. They have confirmed numerous critical and high-severity vulnerabilities in various projects.

Calle warned that unmaintained projects should not be trusted due to the risk of security flaws. The team is using models from Chinese startup Moonshot AI's Kimi K3 and Z.ai's $GLM 5.2, as well as OpenAI and Anthropic. However, they have faced restrictions when using American models.

The Red Team has made significant progress in their scans, with nearly the entire Bitcoin open-source ecosystem examined. They reported filing 4,962 findings across 390 projects, including 85 rated critical and 635 rated high severity. Calle noted that response speed varies across projects, indicating how healthy each project is.

Calle emphasized that relying on unmaintained projects can be hazardous, especially with the rise of AI-powered audits. He warned developers to act quickly in addressing vulnerabilities, as those who started AI audits months ago are at a significant advantage over those who didn't.