{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,4,25]],"date-time":"2026-04-25T06:47:56Z","timestamp":1777099676664,"version":"3.51.4"},"reference-count":31,"publisher":"Springer Science and Business Media LLC","issue":"1","license":[{"start":{"date-parts":[[2025,7,16]],"date-time":"2025-07-16T00:00:00Z","timestamp":1752624000000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"},{"start":{"date-parts":[[2025,7,16]],"date-time":"2025-07-16T00:00:00Z","timestamp":1752624000000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"}],"content-domain":{"domain":["link.springer.com"],"crossmark-restriction":false},"short-container-title":["AI &amp; Soc"],"published-print":{"date-parts":[[2026,1]]},"abstract":"<jats:title>Abstract<\/jats:title>\n                  <jats:p>\n                    Advances in large language models (LLMs) could significantly disrupt political communication. In a large-scale pre-registered experiment (\n                    <jats:italic>n<\/jats:italic>\n                    \u2009=\u20094955), we prompted GPT-4 to generate persuasive messages impersonating the language and beliefs of U.S. political parties\u2014a technique we term \u201cpartisan role-play\u201d\u2014and directly compared their persuasiveness to that of human persuasion experts. In aggregate, the persuasive impact of role-playing messages generated by GPT-4 was not significantly different from that of non-role-playing messages. However, the persuasive impact of GPT-4 rivaled, and on some issues exceeded, that of the human experts. Taken together, our findings suggest that\n                    <jats:italic>\u2014<\/jats:italic>\n                    contrary to popular concern\n                    <jats:italic>\u2014<\/jats:italic>\n                    instructing current LLMs to role-play as partisans offers limited persuasive advantage, but also that current LLMs can rival and even exceed the persuasiveness of human experts. These results potentially portend widespread adoption of AI tools by persuasion campaigns, with important implications for the role of AI in politics and democracy.\n                  <\/jats:p>","DOI":"10.1007\/s00146-025-02464-x","type":"journal-article","created":{"date-parts":[[2025,7,16]],"date-time":"2025-07-16T13:26:40Z","timestamp":1752672400000},"page":"351-361","update-policy":"https:\/\/doi.org\/10.1007\/springer_crossmark_policy","source":"Crossref","is-referenced-by-count":5,"title":["Comparing the persuasiveness of role-playing large language models and human experts on polarized U.S. political issues"],"prefix":"10.1007","volume":"41","author":[{"ORCID":"https:\/\/orcid.org\/0000-0003-2071-1726","authenticated-orcid":false,"given":"Kobi","family":"Hackenburg","sequence":"first","affiliation":[]},{"given":"Lujain","family":"Ibrahim","sequence":"additional","affiliation":[]},{"given":"Ben M.","family":"Tappin","sequence":"additional","affiliation":[]},{"given":"Manos","family":"Tsakiris","sequence":"additional","affiliation":[]}],"member":"297","published-online":{"date-parts":[[2025,7,16]]},"reference":[{"key":"2464_CR1","doi-asserted-by":"publisher","DOI":"10.1017\/PAN.2023.2","author":"LP Argyle","year":"2023","unstructured":"Argyle LP, Busby EC, Fulda N, Gubler JR, Rytting C, Wingate D (2023) Out of one many: using language models to simulate human samples. Polit Anal. https:\/\/doi.org\/10.1017\/PAN.2023.2","journal-title":"Polit Anal"},{"key":"2464_CR2","doi-asserted-by":"publisher","DOI":"10.31219\/OSF.IO\/STAKV","author":"H Bai","year":"2023","unstructured":"Bai H, Voelkel J, Eichstaedt G, Johannes C, Willer R (2023) Artificial intelligence can persuade humans on political issues. OSF Preprints. https:\/\/doi.org\/10.31219\/OSF.IO\/STAKV","journal-title":"OSF Preprints"},{"issue":"10","key":"2464_CR3","doi-asserted-by":"publisher","first-page":"814","DOI":"10.1111\/J.1467-9280.2005.01619.X","volume":"16","author":"JN Bailenson","year":"2005","unstructured":"Bailenson JN, Yee N (2005) Digital chameleons: Automatic assimilation of nonverbal gestures in immersive virtual environments. Psychol Sci 16(10):814\u2013819. https:\/\/doi.org\/10.1111\/J.1467-9280.2005.01619.X","journal-title":"Psychol Sci"},{"key":"2464_CR4","doi-asserted-by":"publisher","unstructured":"Buchanan, B., Lohn, A., Musser, M., & Sedova, K. (2021). Truth, Lies, and Automation How Language Models Could Change Disinformation. https:\/\/doi.org\/10.51593\/2021CA003","DOI":"10.51593\/2021CA003"},{"issue":"1","key":"2464_CR5","doi-asserted-by":"publisher","first-page":"35","DOI":"10.1177\/0146167203258838","volume":"30","author":"JM Burger","year":"2004","unstructured":"Burger JM, Messian N, Patel S, Del Prado A, Anderson C (2004) What a coincidence! the effects of incidental similarity on compliance. Personal Soc Psychol Bull 30(1):35\u201343. https:\/\/doi.org\/10.1177\/0146167203258838","journal-title":"Personal Soc Psychol Bull"},{"key":"2464_CR6","unstructured":"Cialdini, R. B. (2009). Influence: Science and Practice (5th Edition). 272. https:\/\/books.google.com\/books\/about\/Influence.html?id=d91vPwAACAAJ"},{"issue":"1","key":"2464_CR8","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1038\/s41467-022-35576-9","volume":"14","author":"G Eady","year":"2023","unstructured":"Eady G, Paskhalis T, Zilinsky J, Bonneau R, Nagler J, Tucker JA (2023) Exposure to the Russian internet research agency foreign influence campaign on twitter in the 2016 us election and its relationship to attitudes and voting behavior. Nat Commun 14(1):1\u201311. https:\/\/doi.org\/10.1038\/s41467-022-35576-9","journal-title":"Nat Commun"},{"issue":"12","key":"2464_CR9","doi-asserted-by":"publisher","first-page":"1665","DOI":"10.1177\/0146167215607842","volume":"41","author":"M Feinberg","year":"2015","unstructured":"Feinberg M, Willer R (2015) From gulf to bridge: when do moral arguments facilitate political influence? Pers Soc Psychol Bull 41(12):1665\u20131681. https:\/\/doi.org\/10.1177\/0146167215607842","journal-title":"Pers Soc Psychol Bull"},{"key":"2464_CR10","doi-asserted-by":"publisher","DOI":"10.1111\/spc3.12501","author":"M Feinberg","year":"2019","unstructured":"Feinberg M, Willer R (2019) Moral reframing: A technique for effective and persuasive communication across political divides. Soc Personal Psychol Compass. https:\/\/doi.org\/10.1111\/spc3.12501","journal-title":"Soc Personal Psychol Compass"},{"issue":"3","key":"2464_CR11","doi-asserted-by":"publisher","first-page":"560","DOI":"10.1177\/0894439320914853","volume":"40","author":"D Freelon","year":"2022","unstructured":"Freelon D, Bossetta M, Wells C, Lukito J, Xia Y, Adams K (2022) Black trolls matter: racial and ideological asymmetries in social media disinformation. Soc Sci Comput Rev 40(3):560\u2013578. https:\/\/doi.org\/10.1177\/0894439320914853","journal-title":"Soc Sci Comput Rev"},{"issue":"2","key":"2464_CR12","doi-asserted-by":"publisher","first-page":"177","DOI":"10.1017\/S0047404500000701","volume":"2","author":"H Giles","year":"1973","unstructured":"Giles H, Taylor DM, Bourhis R (1973) Towards a theory of interpersonal accommodation through language: some Canadian data. Lang Soc 2(2):177\u2013192. https:\/\/doi.org\/10.1017\/S0047404500000701","journal-title":"Lang Soc"},{"key":"2464_CR13","unstructured":"Goldstein JA., Sastry G., Musser M., DiResta R., Gentzel M., & Sedova, K (2023). Generative Language Models and Automated Influence Operations: Emerging Threats and Potential Mitigations. ArXiv. http:\/\/arxiv.org\/abs\/2301.04246"},{"issue":"2","key":"2464_CR15","doi-asserted-by":"publisher","first-page":"999","DOI":"10.1016\/J.CHB.2005.08.006","volume":"23","author":"RE Guadagno","year":"2007","unstructured":"Guadagno RE, Cialdini RB (2007) Persuade him by email, but see her in person: online persuasion revisited. Comput Hum Behav 23(2):999\u20131015. https:\/\/doi.org\/10.1016\/J.CHB.2005.08.006","journal-title":"Comput Hum Behav"},{"issue":"24","key":"2464_CR16","doi-asserted-by":"publisher","DOI":"10.1073\/PNAS.2403116121","volume":"121","author":"K Hackenburg","year":"2024","unstructured":"Hackenburg K, Margetts H (2024) Evaluating the persuasive influence of political microtargeting with large language models. Proc Natl Acad Sci 121(24):e2403116121. https:\/\/doi.org\/10.1073\/PNAS.2403116121","journal-title":"Proc Natl Acad Sci"},{"key":"2464_CR18","doi-asserted-by":"publisher","DOI":"10.1007\/s10639-023-11834-1","author":"J Jeon","year":"2023","unstructured":"Jeon J, Lee S (2023) Large language models in education: a focus on the complementary relationship between human teachers and ChatGPT. Educ Inf Technol. https:\/\/doi.org\/10.1007\/s10639-023-11834-1","journal-title":"Educ Inf Technol"},{"key":"2464_CR20","doi-asserted-by":"publisher","unstructured":"Karinshak, E., Hancock, J. T., Liu, S. X., & Park, J. S. (2023). Working with AI to persuade: Examining a large language model\u2019s ability to generate pro-vaccination messages. https:\/\/doi.org\/10.1145\/3579592","DOI":"10.1145\/3579592"},{"issue":"2","key":"2464_CR21","doi-asserted-by":"publisher","first-page":"256","DOI":"10.1080\/10584609.2019.1661888\/SUPPL_FILE\/UPCP_A_1661888_SM2582.PDF","volume":"37","author":"FB Keller","year":"2020","unstructured":"Keller FB, Schoch D, Stier S, Yang JH (2020) Political astroturfing on twitter: how to coordinate a disinformation campaign. Polit Commun 37(2):256\u2013280. https:\/\/doi.org\/10.1080\/10584609.2019.1661888\/SUPPL_FILE\/UPCP_A_1661888_SM2582.PDF","journal-title":"Polit Commun"},{"issue":"1","key":"2464_CR23","doi-asserted-by":"publisher","first-page":"104","DOI":"10.1017\/XPS.2020.37","volume":"9","author":"S Kreps","year":"2022","unstructured":"Kreps S, McCain RM, Brundage M (2022) All the news that\u2019s fit to fabricate: ai-generated text as a tool of media misinformation. J Exp Political Sci 9(1):104\u2013117. https:\/\/doi.org\/10.1017\/XPS.2020.37","journal-title":"J Exp Political Sci"},{"key":"2464_CR24","doi-asserted-by":"crossref","unstructured":"Moore Wang, Z., Peng, Z., Que, H., Liu, J., Zhou, W., Wu, Y., Guo, H., Gan, R., Ni, Z., Zhang, M., Zhang, Z., Ouyang, W., Xu, K., Chen, W., Fu, J., & Peng, J. (2023). Rolellm: benchmarking, eliciting, and enhancing role-playing abilities of large language models. https:\/\/chat.openai.com\/","DOI":"10.18653\/v1\/2024.findings-acl.878"},{"key":"2464_CR25","doi-asserted-by":"publisher","DOI":"10.1145\/3411763.3451760","author":"L Reynolds","year":"2021","unstructured":"Reynolds L, Ai M, Ai K, Mcdonell K (2021) Prompt programming for large language models: beyond the few-shot paradigm. Conf Hum Fact Comput Syst - Proceed. https:\/\/doi.org\/10.1145\/3411763.3451760","journal-title":"Conf Hum Fact Comput Syst - Proceed"},{"key":"2464_CR27","unstructured":"Santurkar, S., Durmus, E., Ladhak, F., Lee, C., Liang, P., & Hashimoto, T. (2023). Whose Opinions Do Language Models Reflect? Proceedings of the 40th International Conference on Machine Learning."},{"key":"2464_CR28","doi-asserted-by":"crossref","unstructured":"Shanahan, M., McDonell, K., & Reynolds, L. (2023). Role-Play with Large Language Models. http:\/\/arxiv.org\/abs\/2305.16367","DOI":"10.1038\/s41586-023-06647-8"},{"key":"2464_CR30","unstructured":"Simmons, G. (2022). Moral Mimicry: Large Language Models Produce Moral Rationalizations Tailored to Political Identity. https:\/\/arxiv.org\/abs\/2209.12106v2"},{"key":"2464_CR31","doi-asserted-by":"publisher","DOI":"10.1145\/3359229","author":"K Starbird","year":"2019","unstructured":"Starbird K, Arif A, Wilson T (2019) Disinformation as collaborative work: surfacing the participatory nature of strategic information operations. PACM Hum-Comput Interact. https:\/\/doi.org\/10.1145\/3359229","journal-title":"PACM Hum-Comput Interact"},{"issue":"4","key":"2464_CR32","doi-asserted-by":"publisher","first-page":"862","DOI":"10.1093\/POQ\/NFAC052","volume":"86","author":"BM Tappin","year":"2023","unstructured":"Tappin BM (2023) Estimating the between-issue variation in party elite cue effects. Public Opin Q 86(4):862\u2013885. https:\/\/doi.org\/10.1093\/POQ\/NFAC052","journal-title":"Public Opin Q"},{"key":"2464_CR33","doi-asserted-by":"publisher","DOI":"10.1016\/J.ISCI.2022.105441","author":"R Tucciarelli","year":"2022","unstructured":"Tucciarelli R, Vehar N, Chandaria S, Tsakiris M (2022) On the realness of people who do not exist: the social processing of artificial faces. Iscience. https:\/\/doi.org\/10.1016\/J.ISCI.2022.105441","journal-title":"Iscience"},{"issue":"8","key":"2464_CR34","doi-asserted-by":"publisher","first-page":"917","DOI":"10.1177\/1948550617729408","volume":"9","author":"JG Voelkel","year":"2018","unstructured":"Voelkel JG, Feinberg M (2018) Morally reframed arguments can affect support for political candidates. Social Psychol and Personal Sci 9(8):917\u2013924. https:\/\/doi.org\/10.1177\/1948550617729408","journal-title":"Social Psychol and Personal Sci"},{"key":"2464_CR36","unstructured":"Wei, J., Wang, X., Schuurmans, D., Bosma, M., Ichter, B., Xia, F., Chi Quoc, E. H., Le, V., & Zhou, D. (2022). Chain-of-Thought Prompting Elicits Reasoning in Large Language Models. https:\/\/arxiv.org\/abs\/2201.11903v6"},{"key":"2464_CR37","unstructured":"Weidinger, L., Rauh, M., Marchal, N., Manzini, A., Hendricks, L. A., Mateos-Garcia, J., Bergman, S., Kay, J., Griffin, C., Bariach, B., Gabriel, I., Rieser, V., & Isaac, W. (2023). Sociotechnical Safety Evaluation of Generative AI Systems. https:\/\/arxiv.org\/abs\/2310.11986v2"},{"key":"2464_CR38","doi-asserted-by":"crossref","unstructured":"Wu, N., Gong, M., Shou, L., Liang, S., & Jiang, D. (2023). Large Language Models are Diverse Role-Players for Summarization Evaluation. http:\/\/arxiv.org\/abs\/2303.15078","DOI":"10.1007\/978-3-031-44693-1_54"},{"key":"2464_CR39","unstructured":"Zhou, D., Sch\u00e4rli, N., Hou, L., Wei, J., Scales, N., Wang, X., Schuurmans, D., Cui, C., Bousquet, O., Le, Q., & Chi, E. (2022). Least-to-Most Prompting Enables Complex Reasoning in Large Language Models. https:\/\/arxiv.org\/abs\/2205.10625v3"}],"container-title":["AI &amp; SOCIETY"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/link.springer.com\/content\/pdf\/10.1007\/s00146-025-02464-x.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/link.springer.com\/article\/10.1007\/s00146-025-02464-x","content-type":"text\/html","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/link.springer.com\/content\/pdf\/10.1007\/s00146-025-02464-x.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,2,3]],"date-time":"2026-02-03T02:59:48Z","timestamp":1770087588000},"score":1,"resource":{"primary":{"URL":"https:\/\/link.springer.com\/10.1007\/s00146-025-02464-x"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,7,16]]},"references-count":31,"journal-issue":{"issue":"1","published-print":{"date-parts":[[2026,1]]}},"alternative-id":["2464"],"URL":"https:\/\/doi.org\/10.1007\/s00146-025-02464-x","relation":{},"ISSN":["0951-5666","1435-5655"],"issn-type":[{"value":"0951-5666","type":"print"},{"value":"1435-5655","type":"electronic"}],"subject":[],"published":{"date-parts":[[2025,7,16]]},"assertion":[{"value":"3 September 2024","order":1,"name":"received","label":"Received","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"23 June 2025","order":2,"name":"accepted","label":"Accepted","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"16 July 2025","order":3,"name":"first_online","label":"First Online","group":{"name":"ArticleHistory","label":"Article History"}},{"order":1,"name":"Ethics","group":{"name":"EthicsHeading","label":"Declarations"}},{"value":"\u00a0B.T. is co-founder of an organization that conducts public opinion research. The remaining authors declare no competing interests.","order":2,"name":"Ethics","group":{"name":"EthicsHeading","label":"Conflict of interest"}}]}}