{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,6,16]],"date-time":"2026-06-16T20:58:42Z","timestamp":1781643522650,"version":"3.54.5"},"reference-count":105,"publisher":"Springer Science and Business Media LLC","issue":"2","license":[{"start":{"date-parts":[[2026,3,23]],"date-time":"2026-03-23T00:00:00Z","timestamp":1774224000000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"},{"start":{"date-parts":[[2026,3,23]],"date-time":"2026-03-23T00:00:00Z","timestamp":1774224000000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"}],"funder":[{"DOI":"10.13039\/501100001774","name":"The University of Sydney","doi-asserted-by":"crossref","id":[{"id":"10.13039\/501100001774","id-type":"DOI","asserted-by":"crossref"}]}],"content-domain":{"domain":["link.springer.com"],"crossmark-restriction":false},"short-container-title":["AI Ethics"],"published-print":{"date-parts":[[2026,4]]},"abstract":"<jats:title>Abstract<\/jats:title>\n                  <jats:p>Early large language models (LLMs) were released with minimal alignment, offering a rare view of how generative systems reframe the ethical values embedded in human texts. We examine outputs from a 2021 version of OpenAI\u2019s base GPT-3, prompting it to summarise culturally diverse source materials (laws, political speeches, and philosophical works) and interpreting results through a descriptive, moral value pluralist lens. Where possible, we contextualise outputs with cross-national datasets such as the World Values Survey. We document recurring value drift: Australia\u2019s firearm policy is recast as a threat to liberty; de Beauvoir\u2019s feminist critique becomes gender-essentialist dating advice; and Merkel\u2019s humanitarian appeal is recast as immigration control. In contrast, multilateral documents (UN\/UNESCO) exhibit greater value stability, suggesting consensus-crafted language can buffer against cultural mutation. We argue that these early behaviours (observed before extensive fine-tuning and safety layers) provide a baseline for understanding how training distributions shape normative framing. Our contribution is twofold: (1) empirical evidence that value drift can invert or overwrite encoded values along predictable cultural axes, and (2) a pluralist, descriptive evaluation method that surfaces whose values dominate and when. We conclude with implications for culturally inclusive evaluation and alignment in contemporary LLMs.<\/jats:p>","DOI":"10.1007\/s43681-026-01038-x","type":"journal-article","created":{"date-parts":[[2026,3,23]],"date-time":"2026-03-23T13:04:44Z","timestamp":1774271084000},"update-policy":"https:\/\/doi.org\/10.1007\/springer_crossmark_policy","source":"Crossref","is-referenced-by-count":0,"title":["The ghost in the machine speaks with an American accent: cultural value drift in early GPT-3 and the case for pluralist evaluation of generative AI"],"prefix":"10.1007","volume":"6","author":[{"ORCID":"https:\/\/orcid.org\/0000-0001-7321-0744","authenticated-orcid":false,"given":"Rebecca","family":"Johnson","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-7520-1323","authenticated-orcid":false,"given":"Leslye Denisse","family":"Dias Duran","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-8615-5526","authenticated-orcid":false,"given":"Enrico","family":"Panai","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-0404-4900","authenticated-orcid":false,"given":"Natalia","family":"Men\u00e9ndez Gonz\u00e1lez","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-4941-0505","authenticated-orcid":false,"given":"Giada","family":"Pistilli","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-1074-7047","authenticated-orcid":false,"given":"Julija","family":"Kalpokien\u0117","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-5459-7661","authenticated-orcid":false,"given":"Donald Jay","family":"Bertulfo","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"297","published-online":{"date-parts":[[2026,3,23]]},"reference":[{"key":"1038_CR1","doi-asserted-by":"crossref","unstructured":"Abbo, G. A., Marchesi, S., Wykowska, A., Belpaeme, T. Social value alignment in large language models. In International Workshop on Value Engineering in AI, 2023. Springer, 83\u201397.","DOI":"10.1007\/978-3-031-58202-8_6"},{"key":"1038_CR2","doi-asserted-by":"publisher","unstructured":"Abid, A., Farooqi, M., Zou, J. Persistent anti-muslim bias in large language models. 2021. arXiv. https:\/\/doi.org\/10.48550\/arXiv.2101.05783","DOI":"10.48550\/arXiv.2101.05783"},{"key":"1038_CR3","doi-asserted-by":"crossref","unstructured":"Basile, V., Fell, M., Fornaciari, T., Hovy, D., Paun, S., Plank, B., Poesio, M., Uma, A. We need to consider disagreement in evaluation. In Proceedings of the 1st workshop on benchmarking: past, present and future, 2021. Association for Computational Linguistics, 15\u201321.","DOI":"10.18653\/v1\/2021.bppf-1.3"},{"key":"1038_CR4","unstructured":"Simon de Beauvoir. 1949. Le deuxi\u00e8me sexe I. Les faits et les mythes. Paris: Gallimard. Beauvoir, S.(1949). Le Deuxi\u00e8me Sexe II: L\u2019exp\u00e9rience v\u00e9cue. Paris: Gallimard. Bredo, E.(1998). Evolution, Psychology and John Dewey\u2019s Critique of the Reflex Arc Concept. The Elementary School Journal 98, 5 (1949), 447\u2013466."},{"key":"1038_CR5","doi-asserted-by":"crossref","unstructured":"Emily M Bender, Timnit Gebru, Angelina McMillan-Major, and Shmargaret Shmitchell. 2021. On the Dangers of Stochastic Parrots: Can Language Models Be Too Big?. 2021. 610\u2013623.","DOI":"10.1145\/3442188.3445922"},{"key":"1038_CR6","doi-asserted-by":"publisher","unstructured":"Emily M. Bender and Alexander Koller. 2020. Climbing towards NLU: On Meaning, Form, and Understanding in the Age of Data. In Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics, July 2020. Association for Computational Linguistics, Online, 5185\u20135198. https:\/\/doi.org\/10.18653\/v1\/2020.acl-main.463","DOI":"10.18653\/v1\/2020.acl-main.463"},{"key":"1038_CR7","doi-asserted-by":"publisher","unstructured":"Noam Benkler, Drisana Mosaphir, Scott Friedman, Andrew Smart, and Sonja Schmer-Galunder. 2023. Assessing LLMs for Moral Value Pluralism. (2023). https:\/\/doi.org\/10.48550\/ARXIV.2312.10075","DOI":"10.48550\/ARXIV.2312.10075"},{"key":"1038_CR8","unstructured":"Isaiah Berlin. 1969. Four essays on liberty. (1969)."},{"key":"1038_CR9","doi-asserted-by":"crossref","unstructured":"Abeba Birhane, Pratyusha Kalluri, Dallas Card, William Agnew, Ravit Dotan, and Michelle Bao. 2022. The values encoded in machine learning research. 2022. 173\u2013184.","DOI":"10.1145\/3531146.3533083"},{"key":"1038_CR10","doi-asserted-by":"crossref","unstructured":"Su Lin Blodgett, Solon Barocas, Hal Daum\u00e9 Iii, and Hanna Wallach. 2020. Language (technology) is power: a critical survey of\" bias\" in nlp. arXiv preprint arXiv:2005.14050 (2020).","DOI":"10.18653\/v1\/2020.acl-main.485"},{"key":"1038_CR11","doi-asserted-by":"publisher","unstructured":"Su Lin Blodgett, Gilsinia Lopez, Alexandra Olteanu, Robert Sim, and Hanna Wallach. 2021. Stereotyping Norwegian Salmon: An Inventory of Pitfalls in Fairness Benchmark Datasets. In Proceedings of the 59th Annual Meeting of the Association for Computational Linguistics and the 11th International Joint Conference on Natural Language Processing (Volume 1: Long Papers), August 2021. Association for Computational Linguistics, Online, 1004\u20131015. https:\/\/doi.org\/10.18653\/v1\/2021.acl-long.81","DOI":"10.18653\/v1\/2021.acl-long.81"},{"key":"1038_CR12","unstructured":"Rishi Bommasani, Drew A Hudson, Ehsan Adeli, Russ Altman, Simran Arora, Sydney von Arx, Michael S Bernstein, Jeannette Bohg, Antoine Bosselut, and Emma Brunskill. 2021. On the opportunities and risks of foundation models. arXiv preprint arXiv:2108.07258 (2021)."},{"key":"1038_CR13","doi-asserted-by":"publisher","unstructured":"Tom B. Brown, Benjamin Mann, Nick Ryder, Melanie Subbiah, Jared Kaplan, Prafulla Dhariwal, Arvind Neelakantan, Pranav Shyam, Girish Sastry, Amanda Askell, Sandhini Agarwal, Ariel Herbert-Voss, Gretchen Krueger, Tom Henighan, Rewon Child, Aditya Ramesh, Daniel M. Ziegler, Jeffrey Wu, Clemens Winter, Christopher Hesse, Mark Chen, Eric Sigler, Mateusz Litwin, Scott Gray, Benjamin Chess, Jack Clark, Christopher Berner, Sam McCandlish, Alec Radford, Ilya Sutskever, and Dario Amodei. 2020. Language models are few-shot learners. https:\/\/doi.org\/10.48550\/arXiv.2005.14165","DOI":"10.48550\/arXiv.2005.14165"},{"issue":"1","key":"1038_CR14","doi-asserted-by":"publisher","first-page":"15","DOI":"10.1007\/s10676-018-9448-6","volume":"20","author":"JJ Bryson","year":"2018","unstructured":"Bryson, J.J.: Patiency is not a virtue: the design of intelligent systems and systems of ethics. Ethics Inform. Technol. 20(1), 15\u201326 (2018)","journal-title":"Ethics Inform. Technol."},{"key":"1038_CR15","doi-asserted-by":"publisher","unstructured":"Yong Cao, Li Zhou, Seolhwa Lee, Laura Cabello, Min Chen, and Daniel Hershcovich. Assessing Cross-Cultural Alignment between ChatGPT and Human Societies: An Empirical Study. (2023). https:\/\/doi.org\/10.48550\/ARXIV.2303.17466","DOI":"10.48550\/ARXIV.2303.17466"},{"key":"1038_CR16","unstructured":"Ruth Chang. Incommensurability, incomparability, and practical reason. (1997)."},{"key":"1038_CR17","unstructured":"Mu Chen. ChatGPT doesn\u2019t understand Chinese well. Is there hope? Retrieved May 26, 2025 from https:\/\/www.baiguan.news\/p\/chatgpt-doesnt-understand-chinese"},{"key":"1038_CR18","volume-title":"The alignment problem: How can machines learn human values?","author":"B Christian","year":"2021","unstructured":"Christian, B.: The alignment problem: How can machines learn human values? Atlantic Books (2021)"},{"key":"1038_CR19","doi-asserted-by":"publisher","DOI":"10.1017\/CBO9780511620539","volume-title":"Using language","author":"HH Clark","year":"1996","unstructured":"Clark, H.H.: Using language. Cambridge University Press (1996)"},{"key":"1038_CR20","doi-asserted-by":"publisher","DOI":"10.7551\/mitpress\/12549.001.0001","volume-title":"AI Ethics","author":"M Coeckelbergh","year":"2020","unstructured":"Coeckelbergh, M.: AI Ethics. MIT press (2020)"},{"key":"1038_CR21","unstructured":"CommonCrawl. Common Crawl - Open Repository of Web Crawl Data. Retrieved November 17, 2021 from https:\/\/commoncrawl.org\/"},{"key":"1038_CR22","unstructured":"Council of Australian Governments. 2017. Australian National Firearms Agreement."},{"key":"1038_CR23","volume-title":"Liberalism and value pluralism","author":"G Crowder","year":"2002","unstructured":"Crowder, G.: Liberalism and value pluralism. Bloomsbury Publishing (2002)"},{"issue":"9","key":"1038_CR24","doi-asserted-by":"publisher","first-page":"92","DOI":"10.1145\/2701413","volume":"58","author":"E Davis","year":"2015","unstructured":"Davis, E., Marcus, G.: Commonsense reasoning and commonsense knowledge in artificial intelligence. Commun. ACM 58(9), 92\u2013103 (2015)","journal-title":"Commun. ACM"},{"key":"1038_CR25","unstructured":"Emily Denton, Alex Hanna, Razvan Amironesei, Andrew Smart, Hilary Nicole, and Morgan Klaus Scheuerman. 2020. Bringing the people back in: Contesting benchmark machine learning datasets. arXiv preprint arXiv:2007.07399 (2020)."},{"key":"1038_CR26","doi-asserted-by":"crossref","unstructured":"Jwala Dhamala, Tony Sun, Varun Kumar, Satyapriya Krishna, Yada Pruksachatkun, Kai-Wei Chang, and Rahul Gupta. 2021. Bold: dataset and metrics for measuring biases in open-ended language generation. In Proceedings of the 2021 ACM conference on fairness, accountability, and transparency, 2021. 862\u2013872.","DOI":"10.1145\/3442188.3445924"},{"key":"1038_CR27","unstructured":"Esin Durmus, Karina Nguyen, Thomas I Liao, Nicholas Schiefer, Amanda Askell, Anton Bakhtin, Carol Chen, Zac Hatfield-Dodds, Danny Hernandez, and Nicholas Joseph. 2023. Towards measuring the representation of subjective global opinions in language models. arXiv preprint arXiv:2306.16388 (2023)."},{"key":"1038_CR28","unstructured":"David M. Eberhard, Gary F. Simons, and Charles D. Fennig. 2019. Summary by language size. SIL International, Ethnologue (2019)."},{"key":"1038_CR29","doi-asserted-by":"crossref","unstructured":"Ethayarajh, K., Jurafsky, D. Utility is in the eye of the user: a critique of NLP leaderboards. arXiv preprint arXiv:2009.13888 (2020).","DOI":"10.18653\/v1\/2020.emnlp-main.393"},{"key":"1038_CR30","doi-asserted-by":"publisher","first-page":"681","DOI":"10.1007\/s11023-020-09548-1","volume":"30","author":"L Floridi","year":"2020","unstructured":"Floridi, L., Chiriatti, M.: GPT-3: its nature, scope, limits, and consequences. Mind. Mach. 30, 681\u2013694 (2020)","journal-title":"Mind. Mach."},{"issue":"2","key":"1038_CR31","doi-asserted-by":"publisher","first-page":"219","DOI":"10.1177\/1077800405284363","volume":"12","author":"B Flyvbjerg","year":"2006","unstructured":"Flyvbjerg, B.: Five misunderstandings about case-study research. Qual. Inq. 12(2), 219\u2013245 (2006)","journal-title":"Qual. Inq."},{"issue":"3","key":"1038_CR32","doi-asserted-by":"publisher","first-page":"411","DOI":"10.1007\/s11023-020-09539-2","volume":"30","author":"I Gabriel","year":"2020","unstructured":"Gabriel, I.: Artificial intelligence, values, and alignment. Minds Mach. 30(3), 411\u2013437 (2020)","journal-title":"Minds Mach."},{"key":"1038_CR33","doi-asserted-by":"publisher","DOI":"10.1017\/CBO9780511613579","volume-title":"Liberal Pluralism: The Implications of Value Pluralism for Political Theory and Practice","author":"WA Galston","year":"2002","unstructured":"Galston, W.A.: Liberal Pluralism: The Implications of Value Pluralism for Political Theory and Practice. Cambridge University Press (2002)"},{"key":"1038_CR34","doi-asserted-by":"crossref","unstructured":"Cliff Goddard. 2021. Natural semantic metalanguage. In The Routledge handbook of cognitive linguistics. Routledge, 93\u2013110.","DOI":"10.4324\/9781351034708-7"},{"key":"1038_CR35","unstructured":"James Griffin. 1986. Well-being: Its meaning, measurement and moral importance. Clarendon press."},{"key":"1038_CR36","doi-asserted-by":"publisher","unstructured":"Zishan Guo, Renren Jin, Chuang Liu, Yufei Huang, Dan Shi, Supryadi, Linhao Yu, Yan Liu, Jiaxuan Li, Bojian Xiong, and Deyi Xiong. 2023. Evaluating large language models: a comprehensive survey. https:\/\/doi.org\/10.48550\/arXiv.2310.19736","DOI":"10.48550\/arXiv.2310.19736"},{"key":"1038_CR37","unstructured":"Alex Hern. 2024. TechScape: How cheap, outsourced labour in Africa is shaping AI English. The Guardian. Retrieved August 25, 2025 from https:\/\/www.theguardian.com\/technology\/2024\/apr\/16\/techscape-ai-gadgest-humane-ai-pin-chatgpt"},{"key":"1038_CR38","unstructured":"Geert Hofstede. 2001. Culture\u2019s consequences: Comparing values, behaviors, institutions and organizations across nations. Sage publications."},{"key":"1038_CR39","unstructured":"David Hume. 1739. A Treatise on Human Nature: Being an Attempt to Introduce the Experimental Method of Reasoning Into Moral Subjects. Vol. I [-III]. John Noon."},{"key":"1038_CR40","doi-asserted-by":"publisher","unstructured":"Ben Hutchinson, Negar Rostamzadeh, Christina Greer, Katherine Heller, and Vinodkumar Prabhakaran. 2022. Evaluation gaps in machine learning practice. June 20, 2022. ACM Digital Library, 1859\u20131876. https:\/\/doi.org\/10.1145\/3531146.3533233","DOI":"10.1145\/3531146.3533233"},{"key":"1038_CR41","doi-asserted-by":"crossref","unstructured":"Ben Hutchinson, Andrew Smart, Alex Hanna, Remi Denton, Christina Greer, Oddur Kjartansson, Parker Barnes, and Margaret Mitchell. 2021. Towards accountability for machine learning datasets: Practices from software engineering and infrastructure. In Proceedings of the 2021 ACM conference on fairness, accountability, and transparency, 2021. 560\u2013575.","DOI":"10.1145\/3442188.3445918"},{"key":"1038_CR42","doi-asserted-by":"crossref","unstructured":"Jef Huysmans. 2000. The European Union and the securitization of migration. JCMS: Journal of common market studies 38, 5 (2000), 751\u2013777.","DOI":"10.1111\/1468-5965.00263"},{"key":"1038_CR43","doi-asserted-by":"crossref","unstructured":"Ahmed Izzidien. 2022. Word vector embeddings hold social ontological relations capable of reflecting meaningful fairness assessments. AI & SOCIETY 37, 1 (2022), 299\u2013318.","DOI":"10.1007\/s00146-021-01167-3"},{"key":"1038_CR44","doi-asserted-by":"publisher","unstructured":"Johnson, R. L., Pistilli, G., Men\u00e9dez-Gonz\u00e1lez, N., Dias Duran, L. D., Panai, E., Kalpokiene, J., Bertulfo, D. J. 2022. The Ghost in the Machine has an American accent: value conflict in GPT-3. https:\/\/doi.org\/10.48550\/arXiv.2203.07785","DOI":"10.48550\/arXiv.2203.07785"},{"key":"1038_CR45","unstructured":"Meghann Jones. 2019. #IWD2019: Perceptions of violence against women in France and the United States | Ipsos. IPSOS. Retrieved November 17, 2021 from https:\/\/www.ipsos.com\/en\/iwd2019-perceptions-violence-against-women-france-and-united-states"},{"key":"1038_CR46","doi-asserted-by":"crossref","unstructured":"Peter Jonkers. 2019. How to Respond to Conflicts Over Value Pluralism? Journal of Nationalism, Memory & Language Politics 13, 2 (2019), 183\u2013204.","DOI":"10.2478\/jnmlp-2019-0013"},{"key":"1038_CR47","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.2112.01716","author":"B Koch","year":"2021","unstructured":"Koch, B., Denton, E., Hanna, A., Foster, J.G.: Reduced, reused and recycled: the life of a dataset in machine learning research. arXiv (2021). https:\/\/doi.org\/10.48550\/arXiv.2112.01716","journal-title":"arXiv"},{"key":"1038_CR48","unstructured":"George Lakoff and Mark Johnson. 2008. Metaphors we live by. University of Chicago press."},{"key":"1038_CR49","doi-asserted-by":"publisher","DOI":"10.9783\/9780812209914","volume-title":"The Evolution of International Human Rights: Visions Seen","author":"PG Lauren","year":"2011","unstructured":"Lauren, P.G.: The Evolution of International Human Rights: Visions Seen. University of Pennsylvania Press (2011)"},{"key":"1038_CR50","doi-asserted-by":"publisher","unstructured":"Hector Levesque, Ernest Davis, and Leora Morgenstern. 2012. The winograd schema challenge. 2012. AAAI Press, Rome, Italy. Retrieved from https:\/\/dl.acm.org\/doi\/https:\/\/doi.org\/10.5555\/3031843.3031909","DOI":"10.5555\/3031843.3031909"},{"key":"1038_CR51","unstructured":"Percy Liang, Rishi Bommasani, Tony Lee, Dimitris Tsipras, Dilara Soylu, Michihiro Yasunaga, Yian Zhang, Deepak Narayanan, Yuhuai Wu, and Ananya Kumar. 2022. Holistic evaluation of language models. arXiv preprint arXiv:2211.09110 (2022)."},{"key":"1038_CR52","doi-asserted-by":"crossref","unstructured":"Lu, K., Mardziel, P., Wu, F., Amancharla, P., Datta, A., 2020. Gender bias in neural natural language processing. In\u00a0Logic, language, and security: essays dedicated to Andre Scedrov on the occasion of his 65th birthday\u00a0(pp. 189-202).","DOI":"10.1007\/978-3-030-62077-6_14"},{"key":"1038_CR53","doi-asserted-by":"crossref","unstructured":"Lucy, L., Bamman, D. Gender and representation bias in GPT-3 generated stories. In\u00a0Proceedings of the third workshop on narrative understanding\u00a0(pp. 48-55).","DOI":"10.18653\/v1\/2021.nuse-1.5"},{"issue":"11","key":"1038_CR54","doi-asserted-by":"publisher","first-page":"1363","DOI":"10.1177\/00187267025511005","volume":"55","author":"B McSweeney","year":"2002","unstructured":"McSweeney, B.: The essentials of scholarship: a reply to Geert Hofstede. Hum. Rel. 55(11), 1363\u20131372 (2002)","journal-title":"Hum. Rel."},{"issue":"6","key":"1038_CR55","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1145\/3457607","volume":"54","author":"N Mehrabi","year":"2021","unstructured":"Mehrabi, N., Morstatter, F., Saxena, N., Lerman, K., Galstyan, A.: A survey on bias and fairness in machine learning. ACM Comput. Surv. 54(6), 1\u201335 (2021). https:\/\/doi.org\/10.1145\/3457607","journal-title":"ACM Comput. Surv."},{"key":"1038_CR56","doi-asserted-by":"publisher","first-page":"222","DOI":"10.1016\/j.geoforum.2018.10.027","volume":"101","author":"L Mehta","year":"2019","unstructured":"Mehta, L., Huff, A., Allouche, J.: The new politics and geographies of scarcity. Geoforum 101, 222\u2013230 (2019). https:\/\/doi.org\/10.1016\/j.geoforum.2018.10.027","journal-title":"Geoforum"},{"key":"1038_CR57","unstructured":"Angela Merkel. 2015. Sommerpressekonferenz von Bundeskanzlerin Merkel. Thema: Aktuelle Themen der Innen-und Au\u00dfenpolitik 31, (2015), 2015."},{"key":"1038_CR58","doi-asserted-by":"publisher","unstructured":"Mitchell, M., Wu, S., Zaldivar, A., Barnes, P., Vasserman, L., Hutchinson, B., Spitzer, E., Raji, I.D., Gebru, T., January. Model cards for model reporting. In\u00a0Proceedings of the conference on fairness, accountability, and transparency, 2019. ACM Digital Library, Atlanta, USA, 220\u2013229. https:\/\/doi.org\/10.1145\/3287560.328759","DOI":"10.1145\/3287560.328759"},{"issue":"13","key":"1038_CR59","doi-asserted-by":"publisher","DOI":"10.1073\/pnas.2215907120","volume":"120","author":"M Mitchell","year":"2023","unstructured":"Mitchell, M., Krakauer, D.C.: The debate over understanding in AI\u2019s large language models. Proc. Natl. Acad. Sci. U. S. A. 120(13), e2215907120 (2023). https:\/\/doi.org\/10.1073\/pnas.2215907120","journal-title":"Proc. Natl. Acad. Sci. U. S. A."},{"key":"1038_CR60","doi-asserted-by":"crossref","unstructured":"Johannes Morsink. 1999. The Universal Declaration of Human Rights: origins, drafting, and intent. University of Pennsylvania Press.","DOI":"10.9783\/9780812200416"},{"key":"1038_CR61","unstructured":"Nadeem, M., Bethke, A., Reddy, S. StereoSet: measuring stereotypical bias in pretrained language models. arXiv preprint arXiv:2004.09456 (2020)."},{"key":"1038_CR62","unstructured":"Nagel, T. The fragmentation of value. In Mortal Questions. Cambridge University Press, New York, 1979."},{"key":"1038_CR63","unstructured":"Gitanas Naus\u0117da. 2021. 80 years after the start of the terrible deportations, we are still grieving. We are fighting for historical justice. Retrieved June 13, 2025 from https:\/\/www.lrs.lt\/sip\/portal.show?p_r=35403&p_k=1&p_t=277060"},{"key":"1038_CR64","doi-asserted-by":"crossref","unstructured":"Nawyn, S. J. Refugees in the United States and the politics of crisis. The Oxford handbook of migration crisis (2019), 163\u2013180","DOI":"10.1093\/oxfordhb\/9780190856908.013.23"},{"issue":"17","key":"1038_CR65","doi-asserted-by":"publisher","first-page":"1581","DOI":"10.1056\/NEJMp2102574","volume":"384","author":"J Negin","year":"2021","unstructured":"Negin, J., Alpers, P., Nassar, N., Hemenway, D.: Australian firearm regulation at 25\u2014successes, ongoing challenges, and lessons for the world. New England J. Med. 384(17), 1581\u20131583 (2021)","journal-title":"New England J. Med."},{"issue":"1","key":"1038_CR66","doi-asserted-by":"publisher","first-page":"40","DOI":"10.1080\/13876988.2015.1029334","volume":"19","author":"J Newman","year":"2017","unstructured":"Newman, J., Head, B.: The national context of wicked problems: comparing policies on gun violence in the US, Canada, and Australia. J. Comp. Policy Anal Res. Pract. 19(1), 40\u201353 (2017)","journal-title":"J. Comp. Policy Anal Res. Pract."},{"key":"1038_CR67","unstructured":"Osborne, J. W., Overbay, A. The power of outliers (and why researchers should always check for them). Pract. Assess. Res. Eval. 9, 1 (2004)."},{"issue":"1","key":"1038_CR68","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1075\/eww.25.1.02pee","volume":"25","author":"B Peeters","year":"2004","unstructured":"Peeters, B.: Tall poppies and egalitarianism in Australian discourse: from key word to cultural value. English world-wide 25(1), 1\u201325 (2004)","journal-title":"English world-wide"},{"issue":"2","key":"1038_CR69","doi-asserted-by":"publisher","first-page":"133","DOI":"10.1075\/ijolc.2.2.001pe","volume":"2","author":"B Peeters","year":"2015","unstructured":"Peeters, B.: Language and cultural values. Int. J. Lang. Cult. 2(2), 133\u2013141 (2015)","journal-title":"Int. J. Lang. Cult."},{"key":"1038_CR70","unstructured":"Pew Research Center. 2019. In a Politically Polarized Era, Sharp Divides in Both Partisan Coalitions. Retrieved from https:\/\/www.pewresearch.org"},{"key":"1038_CR71","unstructured":"Inioluwa Deborah Raji, Emily M Bender, Amandalynne Paullada, Emily Denton, and Alex Hanna. 2021. AI and the everything in the whole wide world benchmark. arXiv preprint arXiv:2111.15366 (2021)."},{"key":"1038_CR72","doi-asserted-by":"crossref","unstructured":"Ramezani, A., Xu, Y. Knowledge of cultural moral norms in large language models. arXiv preprint arXiv:2306.01857 (2023).","DOI":"10.18653\/v1\/2023.acl-long.26"},{"key":"1038_CR73","doi-asserted-by":"publisher","DOI":"10.1093\/acprof:oso\/9780198268345.001.0001","volume-title":"Practical Reason and Norms","author":"J Raz","year":"1999","unstructured":"Raz, J.: Practical Reason and Norms. OUP Oxford (1999)"},{"key":"1038_CR74","volume-title":"Understanding human values","author":"M Rokeach","year":"2008","unstructured":"Rokeach, M.: Understanding human values. Simon and Schuster (2008)"},{"key":"1038_CR75","first-page":"11","volume":"1949","author":"G Ryle","year":"1949","unstructured":"Ryle, G.: Descartes\u2019 myth. Concept Mind 1949, 11\u201324 (1949)","journal-title":"Concept Mind"},{"key":"1038_CR76","doi-asserted-by":"crossref","unstructured":"Saguy, A. C. 2012. French and U.S. Legal approaches to sexual harassment: The Pre and post dsk scandal. Travail, genre et soci\u00e9t\u00e9s 28, 2 89\u2013106.","DOI":"10.3917\/tgs.028.0089"},{"key":"1038_CR77","doi-asserted-by":"crossref","unstructured":"Sap, M., Card, D., Gabriel, S., Choi, Y., Smith, N.A. 2019 The risk of racial bias in hate speech detection. In Proceedings of the 57th annual meeting of the association for computational linguistics, 1668\u20131678.","DOI":"10.18653\/v1\/P19-1163"},{"key":"1038_CR78","doi-asserted-by":"publisher","unstructured":"Schlangen, D. 2020. Targeting the benchmark: on methodology in current natural language processing research. https:\/\/doi.org\/10.48550\/arXiv.2007.04792","DOI":"10.48550\/arXiv.2007.04792"},{"issue":"3","key":"1038_CR79","doi-asserted-by":"publisher","first-page":"268","DOI":"10.1177\/0022022101032003002","volume":"32","author":"SH Schwartz","year":"2001","unstructured":"Schwartz, S.H., Bardi, A.: Value hierarchies across cultures: taking a similarities perspective. J. Cross Cult. Psychol. 32(3), 268\u2013290 (2001)","journal-title":"J. Cross Cult. Psychol."},{"key":"1038_CR80","doi-asserted-by":"crossref","unstructured":"John R Searle. 1969. How to derive \u2018ought\u2019from \u2018is.\u2019 In The is-ought question: a collection of papers on the central problem in moral philosophy. Springer, 120\u2013134.","DOI":"10.1007\/978-1-349-15336-7_13"},{"key":"1038_CR81","unstructured":"Small Arms Survey. 2021. Global Firearms Holdings. Retrieved November 17, 2021 from https:\/\/www.smallarmssurvey.org\/database\/global-firearms-holdings"},{"issue":"2021","key":"1038_CR82","first-page":"5861","volume":"34","author":"I Solaiman","year":"2021","unstructured":"Solaiman, I., Dennison, C.: Process for adapting language models to society (palms) with values-targeted datasets. Adv. Neural. Inf. Process. Syst. 34(2021), 5861\u20135873 (2021)","journal-title":"Adv. Neural. Inf. Process. Syst."},{"key":"1038_CR83","unstructured":"Statista. 2023. France: opinion on the 1905 law on secularism in 2023. Retrieved June 13, 2025 from https:\/\/www.statista.com\/statistics\/1237030\/people-consider-secularism-danger-france\/"},{"key":"1038_CR84","unstructured":"Statista. 2025. Global internet penetration rate from 2009 to 2024, by region. Retrieved May 26, 2025 from https:\/\/www.statista.com\/statistics\/265149\/internet-penetration-rate-by-region\/"},{"issue":"32","key":"1038_CR85","doi-asserted-by":"publisher","first-page":"14425","DOI":"10.1073\/pnas.1008662107","volume":"107","author":"GJ Stephens","year":"2010","unstructured":"Stephens, G.J., Silbert, L.J., Hasson, U.: Speaker\u2013listener neural coupling underlies successful communication. Proc. Natl. Acad. Sci. 107(32), 14425\u2013144304 (2010)","journal-title":"Proc. Natl. Acad. Sci."},{"key":"1038_CR86","unstructured":"Robert Leon Stern. 1967. Technology and World Trade: Proceedings. US Department of Commerce, National Bureau of Standards."},{"key":"1038_CR87","doi-asserted-by":"publisher","DOI":"10.1017\/9781108661669","volume-title":"An introduction to Japanese society","author":"Y Sugimoto","year":"2020","unstructured":"Sugimoto, Y.: An introduction to Japanese society. Cambridge University Press (2020)"},{"key":"1038_CR88","doi-asserted-by":"publisher","unstructured":"Sukiennik, N., Gao, C., Xu, F., Li, Y. 2025. An evaluation of cultural value alignment in LLM. (2025). https:\/\/doi.org\/10.48550\/ARXIV.2504.08863","DOI":"10.48550\/ARXIV.2504.08863"},{"key":"1038_CR89","unstructured":"Taleb, N. 2007. The black swan: why don\u2019t we learn that we don\u2019t learn. NY: Random House 1145, (first 2007)."},{"key":"1038_CR90","doi-asserted-by":"publisher","unstructured":"Tamkin, A., Brundage, M., Clark, J., Ganguli, D. 2021. Understanding the capabilities, limitations, and societal impact of large language models. https:\/\/doi.org\/10.48550\/arXiv.2102.02503","DOI":"10.48550\/arXiv.2102.02503"},{"issue":"9","key":"1038_CR91","doi-asserted-by":"publisher","DOI":"10.1093\/pnasnexus\/pgae346","volume":"3","author":"Y Tao","year":"2024","unstructured":"Tao, Y., Viberg, O., Baker, R.S., Kizilcec, R.F.: Cultural bias and cultural alignment of large language models. PNAS Nexus 3(9), pgae346 (2024). https:\/\/doi.org\/10.1093\/pnasnexus\/pgae346","journal-title":"PNAS Nexus"},{"key":"1038_CR92","doi-asserted-by":"crossref","unstructured":"Tausch, A. 2015. Hofstede, Inglehart and beyond. New directions in empirical global value research. University Library of Munich, Germany.","DOI":"10.2139\/ssrn.2606243"},{"key":"1038_CR93","unstructured":"The United Nations. 1979. Convention on the elimination of all forms of discrimination against women. Retrieved November 17, 2022 from https:\/\/www.ohchr.org\/en\/instruments-mechanisms\/instruments\/convention-elimination-all-forms-discrimination-against-women"},{"key":"1038_CR94","unstructured":"Topal, M. O., Bas, A., & van Heerden, I. Exploring transformers in natural language generation: Gpt, bert, and xlnet.\u00a0arXiv preprint arXiv:2102.08036. (2021)"},{"key":"1038_CR95","unstructured":"John Wilder Tukey. 1977. Exploratory data analysis. Springer."},{"key":"1038_CR96","unstructured":"UNESCO. 2021. Draft text of the Recommendation on the Ethics of Artificial Intelligence. In Intergovernmental Meeting of Experts (Category II) related to a Draft Recommendation on the Ethics of Artificial Intelligence, 2021. UNESCO Digital Library, Online. Retrieved from https:\/\/unesdoc.unesco.org\/ark:\/48223\/pf0000377897"},{"issue":"4","key":"1038_CR97","doi-asserted-by":"publisher","first-page":"315","DOI":"10.1080\/01972240309487","volume":"19","author":"J Van Dijk","year":"2003","unstructured":"Van Dijk, J., Hacker, K.: The digital divide as a complex and dynamic phenomenon. Inform. Soc. 19(4), 315\u2013326 (2003)","journal-title":"Inform. Soc."},{"key":"1038_CR98","unstructured":"Vaswani, A., Shazeer, N., Parmar, N., Uszkoreit, J., Jones, L., Gomez, A.N., Kaiser, \u0141. and Polosukhin, I., 2017. Attention is all you need.\u00a0Adv. Neural Inform. Process. Syst.,\u00a030."},{"issue":"2020","key":"1038_CR99","first-page":"12388","volume":"33","author":"J Vig","year":"2020","unstructured":"Vig, J., Gehrmann, S., Belinkov, Y., Qian, S., Nevo, D., Singer, Y., Shieber, S.: Investigating gender bias in language models using causal mediation analysis. Adv. Neural. Inf. Process. Syst. 33(2020), 12388\u201312401 (2020)","journal-title":"Adv. Neural. Inf. Process. Syst."},{"key":"1038_CR100","unstructured":"Luz Helena Orozco y Villa and Natalia Menendez. 2025. On \u2018Constitutional\u2019 AI \u2014 The Digital Constitutionalist. Retrieved August 25, 2025 from https:\/\/digi-con.org\/on-constitutional-ai\/"},{"key":"1038_CR101","unstructured":"Weidinger, L., Mellor, J., Rauh, M., Griffin, C., Uesato, J., Huang, P. S., & Gabriel, I. (2021). Ethical and social risks of harm from language models. arXiv preprint arXiv:2112.04359 (2021)."},{"issue":"1","key":"1038_CR102","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1016\/0010-0285(72)90002-3","volume":"3","author":"T Winograd","year":"1972","unstructured":"Winograd, T.: Understanding natural language. Cogn. Psychol. 3(1), 1\u2013191 (1972)","journal-title":"Cogn. Psychol."},{"key":"1038_CR103","unstructured":"World Values Survey Organisation. 2024. The World Values Survey. The World Values Survey. Retrieved January 18, 2024 from https:\/\/www.worldvaluessurvey.org\/wvs.jsp"},{"key":"1038_CR104","unstructured":"Malcolm X. 1964. The ballot or the bullet. Detroit, Michigan."},{"key":"1038_CR105","doi-asserted-by":"publisher","unstructured":"Zhao, W., Mondal, D., Tandon, N., Dillion, D., Gray, K., & Gu, Y. (2024, May). Worldvaluesbench: A large-scale benchmark dataset for multi-cultural value awareness of language models. 2024. arXiv. https:\/\/doi.org\/10.48550\/ARXIV.2404.16308","DOI":"10.48550\/ARXIV.2404.16308"}],"container-title":["AI and Ethics"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/link.springer.com\/content\/pdf\/10.1007\/s43681-026-01038-x.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/link.springer.com\/article\/10.1007\/s43681-026-01038-x","content-type":"text\/html","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/link.springer.com\/content\/pdf\/10.1007\/s43681-026-01038-x.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,6,16]],"date-time":"2026-06-16T20:43:32Z","timestamp":1781642612000},"score":1,"resource":{"primary":{"URL":"https:\/\/link.springer.com\/10.1007\/s43681-026-01038-x"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,3,23]]},"references-count":105,"journal-issue":{"issue":"2","published-print":{"date-parts":[[2026,4]]}},"alternative-id":["1038"],"URL":"https:\/\/doi.org\/10.1007\/s43681-026-01038-x","relation":{"has-preprint":[{"id-type":"doi","id":"10.21203\/rs.3.rs-7503184\/v1","asserted-by":"object"}]},"ISSN":["2730-5953","2730-5961"],"issn-type":[{"value":"2730-5953","type":"print"},{"value":"2730-5961","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,3,23]]},"assertion":[{"value":"1 September 2025","order":1,"name":"received","label":"Received","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"5 February 2026","order":2,"name":"accepted","label":"Accepted","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"23 March 2026","order":3,"name":"first_online","label":"First Online","group":{"name":"ArticleHistory","label":"Article History"}},{"order":1,"name":"Ethics","group":{"name":"EthicsHeading","label":"Declarations"}},{"value":"The authors declare no competing interests.","order":2,"name":"Ethics","group":{"name":"EthicsHeading","label":"Competing interests"}},{"value":"Not applicable. This study did not involve human participants, animals, or sensitive personal data.","order":3,"name":"Ethics","group":{"name":"EthicsHeading","label":"Ethical approval"}},{"value":"Not applicable.","order":4,"name":"Ethics","group":{"name":"EthicsHeading","label":"Consent to participate"}},{"value":"Not applicable.","order":5,"name":"Ethics","group":{"name":"EthicsHeading","label":"Consent for publication"}}],"article-number":"212"}}