{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,4,29]],"date-time":"2026-04-29T09:44:13Z","timestamp":1777455853140,"version":"3.51.4"},"reference-count":18,"publisher":"SAGE Publications","issue":"1","license":[{"start":{"date-parts":[[2023,1,1]],"date-time":"2023-01-01T00:00:00Z","timestamp":1672531200000},"content-version":"unspecified","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by-nc\/4.0\/"}],"content-domain":{"domain":["journals.sagepub.com"],"crossmark-restriction":true},"short-container-title":["Big Data &amp; Society"],"published-print":{"date-parts":[[2023,1]]},"abstract":"<jats:p>The right to artificial intelligence (AI) explainability has consolidated as a consensus in the research community and policy-making. However, a key component of explainability has been missing: extrapolation, which can reveal whether a model is making inferences beyond the boundaries of its training. We report that AI models extrapolate outside their range of familiar data, frequently and without notifying the users and stakeholders. Knowing whether a model has extrapolated or not is a fundamental insight that should be included in explaining AI models in favor of transparency, accountability, and fairness. Instead of dwelling on the negatives, we offer ways to clear the roadblocks in promoting AI transparency. Our commentary accompanies practical clauses useful to include in AI regulations such as the AI Bill of Rights, the National AI Initiative Act in the United States, and the AI Act by the European Commission.<\/jats:p>","DOI":"10.1177\/20539517231169731","type":"journal-article","created":{"date-parts":[[2023,4,21]],"date-time":"2023-04-21T05:47:08Z","timestamp":1682056028000},"update-policy":"https:\/\/doi.org\/10.1177\/sage-journals-update-policy","source":"Crossref","is-referenced-by-count":24,"title":["Extrapolation and AI transparency: Why machine learning models should reveal when they make decisions beyond their training"],"prefix":"10.1177","volume":"10","author":[{"ORCID":"https:\/\/orcid.org\/0000-0001-7830-8708","authenticated-orcid":false,"given":"Xuenan","family":"Cao","sequence":"first","affiliation":[{"name":"Department of Cultural and Religious Studies, The Chinese University of Hong Kong, Hong Kong"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-4551-5342","authenticated-orcid":false,"given":"Roozbeh","family":"Yousefzadeh","sequence":"additional","affiliation":[{"name":"Yale Center for Medical Informatics, Yale University, New Haven, CT, USA"},{"name":"VA Connecticut Healthcare System, West Haven, CT, USA"}]}],"member":"179","published-online":{"date-parts":[[2023,4,21]]},"reference":[{"key":"bibr1-20539517231169731","unstructured":"Balestriero R, Pesenti J, LeCun Y (2021) Learning in high dimension always amounts to extrapolation.\n                      arXiv preprint arXiv:2110.09485\n                      ."},{"key":"bibr2-20539517231169731","unstructured":"Barocas S, Hardt M, Narayanan A (2019)\n                      Fairness and Machine Learning: Limitations and Opportunities\n                      . fairmlbook.org. http:\/\/www.fairmlbook.org."},{"key":"bibr3-20539517231169731","doi-asserted-by":"publisher","DOI":"10.1073\/pnas.1903070116"},{"key":"bibr4-20539517231169731","volume-title":"Extrapolation Methods: Theory and Practice","author":"Brezinski C","year":"2013"},{"key":"bibr5-20539517231169731","doi-asserted-by":"publisher","DOI":"10.1126\/science.aba9647"},{"key":"bibr6-20539517231169731","unstructured":"Dheeru D, Karra Taniskidou E (2017) UCI machine learning repository. URL http:\/\/archive.ics.uci.edu\/ml."},{"key":"bibr7-20539517231169731","doi-asserted-by":"publisher","DOI":"10.1126\/science.abi5052"},{"key":"bibr8-20539517231169731","doi-asserted-by":"publisher","DOI":"10.1089\/aid.2019.0136"},{"key":"bibr9-20539517231169731","volume-title":"The Mathematical Papers of Isaac Newton","author":"Newton I","year":"2008"},{"key":"bibr10-20539517231169731","doi-asserted-by":"publisher","DOI":"10.1038\/s42256-019-0048-x"},{"key":"bibr11-20539517231169731","doi-asserted-by":"publisher","DOI":"10.1038\/s43586-022-00172-0"},{"key":"bibr12-20539517231169731","unstructured":"The European Commission (2021) The AI act. URL https: \/\/artificialintelligenceact.eu\/the-act\/."},{"key":"bibr13-20539517231169731","unstructured":"The White House Office of Science and Technology Policy (2022) Blueprint for an AI bill of rights: Making automated systems work for the American people. URL https:\/\/www. whitehouse.gov\/ostp\/ai-bill-of-rights."},{"issue":"2","key":"bibr14-20539517231169731","first-page":"494","volume":"2019","author":"Wachter S","year":"2019","journal-title":"Columbia Business Law Review"},{"issue":"2","key":"bibr15-20539517231169731","first-page":"841","volume":"31","author":"Wachter S","year":"2018","journal-title":"Harvard Journal of Law & Technology"},{"key":"bibr16-20539517231169731","unstructured":"Webb T, Dulberg Z, Frankland S, et al. (2020) Learning representations that support extrapolation. In:\n                      International Conference on Machine Learning\n                      . PMLR, pp. 10136\u201310146."},{"key":"bibr17-20539517231169731","unstructured":"Yousefzadeh R, Mollick JA (2021) Extrapolation frameworks in cognitive psychology suitable for study of image classification models.\n                      Workshop on Human and Machine Decisions at NeurIPS\n                      ."},{"key":"bibr18-20539517231169731","doi-asserted-by":"publisher","DOI":"10.1007\/s44007-021-00003-w"}],"container-title":["Big Data &amp; Society"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/journals.sagepub.com\/doi\/pdf\/10.1177\/20539517231169731","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/journals.sagepub.com\/doi\/full-xml\/10.1177\/20539517231169731","content-type":"application\/xml","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/journals.sagepub.com\/doi\/pdf\/10.1177\/20539517231169731","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,4,28]],"date-time":"2026-04-28T13:00:24Z","timestamp":1777381224000},"score":1,"resource":{"primary":{"URL":"https:\/\/journals.sagepub.com\/doi\/10.1177\/20539517231169731"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2023,1]]},"references-count":18,"journal-issue":{"issue":"1","published-print":{"date-parts":[[2023,1]]}},"alternative-id":["10.1177\/20539517231169731"],"URL":"https:\/\/doi.org\/10.1177\/20539517231169731","relation":{},"ISSN":["2053-9517","2053-9517"],"issn-type":[{"value":"2053-9517","type":"print"},{"value":"2053-9517","type":"electronic"}],"subject":[],"published":{"date-parts":[[2023,1]]},"article-number":"20539517231169731"}}