The security team used cutting-edge AI to simulate attacks on 150 Bitcoin core projects, uncovering an average of one critical vulnerability per person per hour.

August 9th, a voluntary security team recently used cutting-edge AI models to scan approximately 150 Bitcoin Core-related project code repositories and has discovered over a dozen vulnerabilities involving wallet, cryptography library, and infrastructure projects. It is reported that the team used Kimi K3, OpenAI's GPT Sol, Anthropic's Claude Fable and Opus models, and Z.ai's GLM 5.2 to identify vulnerabilities and generate support documentation.

Team members stated that currently, on average, each person can discover about one critical vulnerability per hour. In the past 12 hours, security reports have been submitted to multiple projects, but the specific affected projects have not been disclosed. Recent security incidents such as Coldcard and Boltz have also shown that AI is being simultaneously used by security researchers and attackers to quickly discover software vulnerabilities.