CodeRabbit Study: AI-Generated Code Has 2.74x More Security Vulnerabilities Than Human-Written Code

CodeRabbit Blog·high signal

CodeRabbit's analysis of 470 open-source GitHub PRs found that AI co-authored code contains 1.7x more major issues overall, with security vulnerabilities 2.74x more frequent and misconfigurations 75% more common compared to human-written code. The dominant failure modes are logic errors, flawed control flow, and incorrect dependencies — subtle issues that pass syntax checks and tests but produce wrong behavior in production. The implication for vibe coders: AI generates plausible-looking code that requires adversarial review, not casual acceptance.