The PoC Pollution Problem: How AI-Generated Exploits Are Poisoning Detection Engineering

This report warns that AI-assisted generation of superficially plausible but functionally broken PoCs is flooding public repositories and security blogs, creating wasted analyst effort and the risk of ineffective or polluted detection rules. Using examples tied to Cisco CVEs (CVE-2025-20281, CVE-2025-20337, CVE-2025-20188) and a representative pcap, it recommends source reputation tracking, multi-source validation, rapid functional testing, traffic-first analysis, automated PoC validation infrastructure, and training to reduce false leads and detection blind spots.