AI noise and the effect it’s having on vulnerability disclosure programs 

This piece outlines common pitfalls in vulnerability disclosure programs (VDPs), particularly the surge of low-value and AI-generated submissions that swamp PSIRT teams and obscure high-severity issues. It recommends tighter scoping, consolidating duplicate findings (e.g., many XSS instances into a single input-validation problem), flexible bounty structures that reward root-cause fixes, and clear escalation routes directly to organizations when warranted—emphasizing that while triage can be outsourced, security accountability cannot.