AI-Generated Code Security: 78% of AI PRs We Scanned Shipped a Vulnerability
78% of 14,213 AI-assisted pull requests we scanned contained a vulnerability that survived human review. Here is how we measured it, the flaw classes, and what fixes it.
4 articles
78% of 14,213 AI-assisted pull requests we scanned contained a vulnerability that survived human review. Here is how we measured it, the flaw classes, and what fixes it.
AI coding tools are making developers faster — but faster development also requires better visibility, stronger review workflows, and more reliable remediation. This is a practical governance guide for teams adopting Codex, Claude Code, Cursor, Windsurf, and other AI coding agents.
Detection alone cannot keep up with AI-speed development. AI-native remediation is the next layer — helping teams fix, validate, and track vulnerabilities in AI-generated code at every stage of the SDLC.
AI coding tools are writing nearly half of all new code. And 45% of that code ships with at least one vulnerability. Vibe coding security is the practice of securing software created by AI — detecting, prioritizing, and remediating risks before they reach production.
Plexicus is Proof-Driven AppSec: validated findings, contextual understanding, and reviewed remediation — anchored in evidence, scoped with you.