HAProxy lead developer: AI bug hunting finds issues but struggles to land them
Understand this story
The HAProxy lead developer documented months of experience using AI tools to hunt for vulnerabilities in open-source code: Anthropic's Mythos once found real defects of relatively high complexity, but the generated reports lacked reproduction steps and patches, and developers had to use Opus 4.8 to decode them before they could be handled, with about one-third of the fixes eventually merged. This practice shows that AI can indeed locate some highly complex vulnerabilities in open-source software, but incomplete reports make it difficult to turn findings directly into mergeable fixes, and developers still need to step in.
AI overview based on source reports · Updated 34 min ago
Coverage timeline
Follow the coverage to explore different perspectives.
- Hacker News AIThe Use of AI-Based Tools to Find Bugs in Open Source Software
HAProxy lead developer Willy Tarreau describes receiving hundreds of AI-assisted security reports that lack reproducers or patches, and details how access restrictions to Anthropic's most capable models, frequent safeguards blocking defensive work, and unusable Mythos-generated reports burden open-source maintainers. He outlines what developers actually need from these tools—self-scanning, reproducers, fix candidates, and regression tests—and criticizes Anthropic's Cyber Verification Program and the CLA required by its new OSS scanner project.
Activity for this story
- Current comparable activity
- 7
- Peak comparable activity
- 10Nov 20 08:00
- Change in the past 24 hours
- –
Trends compare the same fully observed participants, which may cover fewer participants than current activity. Move the pointer or tap the chart to inspect hourly activity; use the left and right arrow keys with a keyboard.