AI Can Find Bugs, But Human Knowledge Still Proves Them

This article discusses how AI is accelerating discovery in offensive security but cautions that generated findings are hypotheses, not proven vulnerabilities; it urges human technical validation, offers a practical checklist for elevating leads to validated findings, and warns that overreliance on AI can produce noisy, unverified reports and degrade practitioner skills.