"Looks Good to Me": A Practical Guide to Handling AI-Generated Code
summary
Open source projects see bots committing at scale (one bot hit nearly 100 repos); maintainers report up to 70% AI-assisted PRs in some flows and industry counts cite 17 million AI PRs by March 2026 and “60%+” of new code being AI-assisted. Practical steps: publish a short 60–100 word policy, run automated PQSS anti-slope checks (31 heuristics, ~15s), embed security skills (128-skill catalog, CodeGuard/Trapidust integrations) and triage AI reports with LLM-assisted analysis that flags real issues (demo rejects a PR for SQL injection).
AI-written summary · about
chapters (16)
- 0:00AI code challenges
- 3:21AI in open source
- 9:03Security impacts on bug bounties
- 13:19AI policies in projects
- 18:03AI threats and escalation
- 20:52AI coding readiness
- 22:17Handling AI contributions
- 23:22AI coding guidelines
- 27:39Secure AI coding practices
- 33:00AI security working groups
- 36:56Secure AI tools
- 39:28AI code review demo
- 46:45AI contribution quality control
- 51:51AI-era vulnerability triage
- 57:15Actionable steps for AI code
- 1:00:05AI in software development
related talks
43mWriting for Machines: How To Capture Your Project's "Vibe" and Survive AI Slop
23mAI-Generated Code: Who Is Responsible for the Bugs?
14mAI's impact on open-source contributions and mentoring
19mThe AI Code Review Bottleneck: Escaping the Velocity Trap
27m3 data-backed ideas to deter AI code quality problems
49mClean Code in the AI era
21mHow To Make AI-Assisted OSS Contributions Review-Ready
52mAre Your Engineering Teams Really Prepared for the GenAI Revolution?!