LLM-generated advisories can poison the feeds humans and automation trust, so…

LLM-generated advisories can poison the feeds humans and automation trust, so provenance tiers and cross-source confirmation must gate any ‘critical’ action. The JFrog audit of fabricated SQLite CVEs shows plausible-sounding reports can travel into NVD and ADP pipelines and trigger scanners before anyone reproduces a PoC. Require vendor corroboration or a reproduced exploit before opening incidents, blocking deploys, or auto-upgrading; treat single-source AI advisories as intelligence, not incident triggers.