Just published a benchmark testing whether AI code reviewers catch security bugs nobody tells them to look for.
I ran 10 "harmless-looking" refactors past Claude Sonnet 5, GPT-5.5, Gemini 3.7 Flash, and DeepSeek-R1 — all four missed the exact same one-line
For further actions, you may consider blocking this person and/or reporting abuse
Top comments (0)