{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,8,17]],"date-time":"2026-08-17T05:24:30Z","timestamp":1786944270039,"version":"build-2736575974"},"reference-count":29,"publisher":"Frontiers Media SA","license":[{"start":{"date-parts":[[2026,8,17]],"date-time":"2026-08-17T00:00:00Z","timestamp":1786924800000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":["frontiersin.org"],"crossmark-restriction":true},"short-container-title":["Front. Med."],"abstract":"<jats:p>\n                    The U. S. Food and Drug Administration has published several types of public regulatory summaries of artificial intelligence-enabled medical devices, including 510(k) summaries,\n                    <jats:italic>De Novo<\/jats:italic>\n                    decision summaries, and premarket approval summaries of safety and effectiveness data. These documents are among the few regulatory texts routinely accessible to clinical institutions, patient advocates, educators, and researchers. Existing empirical audits have documented reporting deficits in areas such as demographic representation, model development details, and clinical validation data. This Perspective draws on these empirical findings and sociotechnical scholarship to formulate the saturation-gap paradox as a preliminary conceptual hypothesis requiring future empirical validation. It proposes that public summaries may be technically and procedurally detailed while providing only limited entry points for assessing fairness, clinical equity, and accountability. Rather than presenting a new systematic audit or inferring what FDA reviewers evaluated internally, this perspective focuses on the accessibility, structure, and accountability function of public disclosures. It considers three ways in which these disclosure genres may limit fairness scrutiny. These involve semantic abstraction, metricization and procedural governance, and accountability redistribution. Finally, building on this prioritized conceptual framework, this Perspective offers preliminary design propositions for disclosure reform and an illustrative TPACK-based educational outline as directions for future research, stakeholder consultation, and prospective development.\n                  <\/jats:p>","DOI":"10.3389\/fmed.2026.1894905","type":"journal-article","created":{"date-parts":[[2026,8,17]],"date-time":"2026-08-17T04:33:32Z","timestamp":1786941212000},"update-policy":"https:\/\/doi.org\/10.3389\/crossmark-policy","source":"Crossref","is-referenced-by-count":0,"title":["The fairness paradox in public regulatory summaries of FDA-authorized artificial intelligence-enabled medical devices"],"prefix":"10.3389","volume":"13","author":[{"given":"Chun-Wang","family":"Lui","sequence":"first","affiliation":[{"name":"Tsinghua Shenzhen International Graduate School, Tsinghua University","place":["Shenzhen, China"]},{"name":"Technological and Higher Education Institute of Hong Kong","place":["Hong Kong SAR, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Pusheng","family":"Wang","sequence":"additional","affiliation":[{"name":"Tsinghua Shenzhen International Graduate School, Tsinghua University","place":["Shenzhen, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1965","published-online":{"date-parts":[[2026,8,17]]},"reference":[{"key":"ref1","doi-asserted-by":"publisher","first-page":"e253172","DOI":"10.1001\/jamahealthforum.2025.3172","article-title":"Early recalls and clinical validation gaps in artificial intelligence\u2013enabled medical devices","volume":"6","author":"Lee","year":"2025","journal-title":"JAMA Health Forum"},{"key":"ref2","doi-asserted-by":"publisher","first-page":"388","DOI":"10.1038\/s41746-025-01800-1","article-title":"How AI is used in FDA-authorized medical devices: a taxonomy across 1,016 authorizations","volume":"8","author":"Singh","year":"2025","journal-title":"NPJ Digit Med"},{"key":"ref3","doi-asserted-by":"publisher","first-page":"973","DOI":"10.1177\/1461444816676645","article-title":"Seeing without knowing: limitations of the transparency ideal and its application to algorithmic accountability","volume":"20","author":"Ananny","year":"2018","journal-title":"New Media Soc"},{"key":"ref4","author":"Raji","year":"2022"},{"key":"ref5","doi-asserted-by":"publisher","first-page":"273","DOI":"10.1038\/s41746-024-01270-x","article-title":"A scoping review of reporting gaps in FDA-approved AI medical devices","volume":"7","author":"Muralidharan","year":"2024","journal-title":"NPJ Digit Med"},{"key":"ref6","author":"Dignum","year":"2025"},{"key":"ref7","doi-asserted-by":"publisher","first-page":"613","DOI":"10.1007\/s11023-022-09611-z","article-title":"Contestable AI by design: towards a framework","volume":"33","author":"Alfrink","year":"2023","journal-title":"Mind Mach"},{"key":"ref8","doi-asserted-by":"publisher","first-page":"673","DOI":"10.1038\/s41746-025-02052-9","article-title":"Evaluating transparency in AI\/ML model characteristics for FDA-reviewed medical devices","volume":"8","author":"Mehta","year":"2025","journal-title":"NPJ Digit Med"},{"key":"ref9","author":"Selbst","year":"2019"},{"key":"ref10","doi-asserted-by":"publisher","first-page":"e258052","DOI":"10.1001\/jamanetworkopen.2025.8052","article-title":"Generalizability of FDA-approved AI-enabled medical devices for clinical use","volume":"8","author":"Windecker","year":"2025","journal-title":"JAMA Netw Open"},{"key":"ref11","doi-asserted-by":"publisher","first-page":"197","DOI":"10.1111\/rego.12358","article-title":"Human judgment in algorithmic loops: individual justice and automated decision-making","volume":"16","author":"Binns","year":"2022","journal-title":"Regul Gov"},{"key":"ref12","doi-asserted-by":"publisher","first-page":"7236","DOI":"10.48550\/arXiv.1609.07236","article-title":"On the (im)possibility of fairness","volume":"2016","author":"Friedler","year":"2016","journal-title":"arXiv"},{"key":"ref13","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1145\/3449180","article-title":"Conceptualising contestability: perspectives on contesting algorithmic decisions","volume":"5","author":"Lyons","year":"2021","journal-title":"Proc ACM Hum-Comput Interact"},{"key":"ref14","doi-asserted-by":"publisher","first-page":"582","DOI":"10.1038\/s41591-021-01312-x","article-title":"How medical AI devices are evaluated: limitations and recommendations from an analysis of FDA approvals","volume":"27","author":"Wu","year":"2021","journal-title":"Nat Med"},{"key":"ref15","doi-asserted-by":"publisher","first-page":"e253351","DOI":"10.1001\/jamahealthforum.2025.3351","article-title":"Benefit-risk reporting for FDA-cleared artificial intelligence\u2212enabled medical devices","volume":"6","author":"Lin","year":"2025","journal-title":"JAMA Health Forum"},{"key":"ref16","doi-asserted-by":"publisher","first-page":"21","DOI":"10.1038\/s41746-023-00992-8","article-title":"Transparency of artificial intelligence\/machine learning-enabled medical devices","volume":"7","author":"Shick","year":"2024","journal-title":"NPJ Digit Med"},{"key":"ref17","doi-asserted-by":"publisher","first-page":"336","DOI":"10.1177\/1527476418796632","article-title":"Data colonialism: rethinking big data\u2019s relation to the contemporary subject","volume":"20","author":"Couldry","year":"2019","journal-title":"Telev New Media"},{"key":"ref18","doi-asserted-by":"publisher","first-page":"833","DOI":"10.1093\/ehjdh\/ztaf041","article-title":"Medicine, healthcare and the AI act: gaps, challenges and future implications","volume":"6","author":"Vardas","year":"2025","journal-title":"Eur Heart J"},{"key":"ref19","author":"Sloane","year":"2022"},{"key":"ref20","doi-asserted-by":"publisher","first-page":"11894","DOI":"10.2139\/ssrn.3311894","article-title":"Contestability and professionals: from explanations to engagement with algorithmic systems","volume":"2018","author":"Mulligan","year":"2018","journal-title":"SSRN Electron J"},{"key":"ref21","doi-asserted-by":"publisher","first-page":"449","DOI":"10.1353\/sor.2019.0022","article-title":"Owning ethics: corporate logics, Silicon Valley, and the institutionalization of ethics","volume":"86","author":"Metcalf","year":"2019","journal-title":"Soc Res"},{"key":"ref22","author":"Raji","year":"2020"},{"key":"ref23","doi-asserted-by":"publisher","first-page":"1017","DOI":"10.1111\/j.1467-9620.2006.00684.x","article-title":"Technological pedagogical content knowledge: a framework for teacher knowledge","volume":"108","author":"Mishra","year":"2006","journal-title":"Teach Coll Rec"},{"key":"ref24","doi-asserted-by":"publisher","first-page":"235","DOI":"10.1080\/21532974.2023.2247480","article-title":"TPACK in the age of ChatGPT and generative AI","volume":"39","author":"Mishra","year":"2023","journal-title":"J Digit Learn Teach Educ"},{"key":"ref25","doi-asserted-by":"publisher","first-page":"4","DOI":"10.3102\/0013189X015002004","article-title":"Those who understand: knowledge growth in teaching","volume":"15","author":"Shulman","year":"1986","journal-title":"Educ Res"},{"key":"ref26","year":"2024"},{"key":"ref27","year":"2025"},{"key":"ref28","year":"2021"},{"key":"ref29","year":"2024"}],"container-title":["Frontiers in Medicine"],"original-title":[],"link":[{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fmed.2026.1894905\/full","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,8,17]],"date-time":"2026-08-17T04:33:36Z","timestamp":1786941216000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fmed.2026.1894905\/full"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,8,17]]},"references-count":29,"alternative-id":["10.3389\/fmed.2026.1894905"],"URL":"https:\/\/doi.org\/10.3389\/fmed.2026.1894905","relation":{},"ISSN":["2296-858X"],"issn-type":[{"value":"2296-858X","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,8,17]]},"article-number":"1894905"}}