Evaluating the persuasive influence of political microtargeting with large language models
Open accessAbstract
Recent advancements in large language models (LLMs) have raised the prospect of scalable, automated, and fine-grained political microtargeting on a scale previously unseen; however, the persuasive influence of microtargeting with LLMs remains unclear. Here, we build a custom web application capable of integrating self-reported demographic and political data into GPT-4 prompts in real-time, facilitating the live creation of unique messages tailored to persuade individual users on four political issues. We then deploy this application in a preregistered randomized control experiment ( n = 8,587) to investigate the extent to which access to individual-level data increases the persuasive influence of GPT-4. Our approach yields two key findings. First, messages generated by GPT-4 were broadly persuasive, in some cases increasing support for an issue stance by up to 12 percentage points. Second, in aggregate, the persuasive impact of microtargeted messages was not statistically different from that of non-microtargeted messages (4.83 vs. 6.20 percentage points, respectively, P = 0.226). These trends hold even when manipulating the type and number of attributes used to tailor the message. These findings suggest—contrary to widespread speculation—that the influence of current LLMs may reside not in their ability to tailor messages to individuals but rather in the persuasiveness of their generic, nontargeted messages. We release our experimental dataset, GPTarget2024 , as an empirical baseline for future research.
The claims and statistics of this paper haven’t been checked yet.
Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.
Integrity signals
Registry
Cites
Statistics
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
The findings support dissenting voices on the persuasive impact of either customization or elaboration.
“At the very least, they support dissenting voices on the persuasive impact of either customization or elaboration (11, 28, 74, 76, 77).”
From Testing theories of political persuasion using AI (2025) · cites this paper as [28]
The Results report no persuasive advantage for accurately microtargeted GPT-4 messages over nontargeted messages, supporting skepticism about customization in this setting. The Experimental Materials section says message length was kept around 200 words; the paper does not test whether greater elaboration changes persuasive impact. (LLM, conf=0.88)
Experimental — accuracy not yet measured
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Cites
Proceedings of the 25th international conference on Machine learning
Cohen, W. W., McCallum, A., & Roweis, S. T. (2008) · Cited by 5,691 · see citing works
Age of Surveillance Capitalism – The Fight for a Human Future at the New Frontier of Power.
Kendell, D. (2020) · Brock Education Journal · doi:10.26522/brocked.v29i2.849 · Cited by 2,159 · see citing works
Psychological targeting as an effective approach to digital mass persuasion
Matz, S. C., Kosinski, M., Nave, G., & Stillwell, D. J. (2017) · Proceedings of the National Academy of Sciences · doi:10.1073/pnas.1710966114 · Cited by 892 · cited 1× · see citing works
“This finding builds on prior work suggesting that political advertisements have increased persuasive power when tailored to an individual's position on the introversion–extroversion spectrum (29) or other psychological characteristics (30).”
BackgroundOut of One, Many: Using Language Models to Simulate Human Samples
Argyle, L. P., Busby, E. C., Fulda, N., Gubler, J. R., Rytting, C., & Wingate, D. (2023) · Political Analysis · doi:10.1017/pan.2023.2 · Cited by 783 · cited 4× · see citing works
“Participants were screened such that all were located in the United States, spoke English as their first language, and were over the age of 18.”
BackgroundInfluentialTaxonomy of Risks posed by Language Models
Weidinger, L., Uesato, J., Rauh, M., Griffin, C., Huang, P.-S., Mellor, J., Glaese, A., Cheng, M., Balle, B., Kasirzadeh, A., et al. (2022) · 2022 ACM Conference on Fairness Accountability and Transparency · doi:10.1145/3531146.3533088 · Cited by 764 · cited 1× · see citing works
“This raises a clear worry: When integrated with existing databases of personal data, LLMs could tailor their messages to appeal to the vulnerabilities and values of specific individuals, potentially reinforcing their existing beliefs, or persuading them to adopt new ones (13–16).”
BackgroundHow is ChatGPT's behavior changing over time?
Chen, L., Zaharia, M., & Zou, J. (2023) · arXiv · doi:10.48550/arxiv.2307.09009 · Cited by 507 · cited 1× · see citing works
“Finally, some work has suggested that OpenAI made changes to the GPT - 4 model which degraded it’s performance on some tasks during the time this experiment was being conducted (35).”
ResultAll the News That’s Fit to Fabricate: AI-Generated Text as a Tool of Media Misinformation
Kreps, S., McCain, R. M., & Brundage, M. (2022) · Journal of Experimental Political Science · doi:10.1017/xps.2020.37 · Cited by 360 · cited 1× · see citing works
“These rules were motivated by an attempt to capture the strongest possible persuasive effects: voters are broadly disapproving of microtargeting practices (40–42) and skeptical of AI - authored texts (43).”
BackgroundHuman heuristics for AI-generated language are flawed
Jakesch, M., Hancock, J. T., & Naaman, M. (2023) · Proceedings of the National Academy of Sciences · doi:10.1073/pnas.2208839120 · Cited by 342 · cited 1× · see citing works
“Across these domains, humans are no longer able to consistently distinguish human and LLM - generated texts (8, 10).”
BackgroundOnline Political Microtargeting: Promises and Threats for Democracy
Zuiderveen Borgesius, F. J., Möller, J., Kruikemeier, S., Ó Fathaigh, R., Irion, K., Dobber, T., Bodo, B., & De Vreese, C. (2018) · Utrecht Law Review · doi:10.18352/ulr.420 · Cited by 340 · see citing works
AI model GPT-3 (dis)informs us better than humans
Spitale, G., Biller-Andorno, N., & Germani, F. (2023) · Science Advances · doi:10.1126/sciadv.adh1850 · Cited by 241 · see citing works
Zarouali, B., Dobber, T., Pauw, G. D., & de Vreese, C. H. (2020) · Communication Research · doi:10.1177/0093650220961965 · Cited by 225 · see citing works
Persuasion for Good: Towards a Personalized Persuasive Dialogue System for Social Good
Wang, X., Shi, W., Kim, R., Oh, Y., Yang, S., Zhang, J., & Yu, Z. (2019) · Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics · doi:10.18653/v1/p19-1566 · Cited by 222 · cited 1× · see citing works
“This raises a clear worry: When integrated with existing databases of personal data, LLMs could tailor their messages to appeal to the vulnerabilities and values of specific individuals, potentially reinforcing their existing beliefs, or persuading them to adopt new ones (13–16).”
BackgroundTargeted Campaign Appeals and the Value of Ambiguity
Hersh, E. D., & Schaffner, B. F. (2013) · The Journal of Politics · doi:10.1017/s0022381613000182 · Cited by 203 · see citing works
Coppock, A., Hill, S. J., & Vavreck, L. (2020) · Science Advances · doi:10.1126/sciadv.abc4046 · Cited by 171 · see citing works
Karinshak, E., Liu, S. X., Park, J. S., & Hancock, J. T. (2023) · Proceedings of the ACM on Human-Computer Interaction · doi:10.1145/3579592 · Cited by 167 · cited 1× · see citing works
“Early research suggests that most capable LLMs can directly persuade humans on political issues (5), draft more persuasive public communications than actual government agencies and political communication experts (6, 7), and generate convincing disinfor-mation and fake news (1, 8, 9).”
BackgroundGoldstein, J. A., Sastry, G., Musser, M., DiResta, R., Gentzel, M., & Sedova, K. (2023) · arXiv · doi:10.48550/arxiv.2301.04246 · Cited by 152 · cited 3× · see citing works
“Crucially, these models are also highly scalable, allowing for essentially limitless production of political messages at an extremely low cost (2, 11).”
BackgroundThe Long-lasting Effects of Newspaper Op-Eds on Public Opinion
Coppock, A., Ekins, E. E., & Kirby, D. (2018) · Quarterly Journal of Political Science · doi:10.1561/100.00016112 · Cited by 140 · cited 1× · see citing works
“However, a body of research suggests that such divergent reactions are rare, and that persuasive political messages tend to sway people, regardless of their demographic traits, in broadly similar ways (23–25).”
BackgroundHow persuasive is AI-generated propaganda?
Goldstein, J. A., Chao, J., Grossman, S., Stamos, A., & Tomz, M. (2024) · PNAS Nexus · doi:10.1093/pnasnexus/pgae034 · Cited by 133 · see citing works
Moral Mimicry: Large Language Models Produce Moral Rationalizations Tailored to Political IdentityPreprint
Simmons, G. (2022) · Annual Meeting of the Association for Computational Linguistics · doi:10.48550/arxiv.2209.12106 · Cited by 122 · cited 1× · see citing works
“For instance, when prompted to adopt a liberal or conservative political identity, these models produce text reflecting the respective moral biases (32).”
BackgroundKirk, H. R., Vidgen, B., Röttger, P., & Hale, S. A. (2024) · Nature Machine Intelligence · doi:10.1038/s42256-024-00820-y · Cited by 122 · see citing works
Whose Opinions Do Language Models Reflect?
Santurkar, S., Durmus, E., Ladhak, F., Lee, C., Liang, P., & Hashimoto, T. (2023) · arXiv · doi:10.48550/arxiv.2303.17548 · Cited by 114 · cited 5× · see citing works
“Third, research has suggested that the dominant approach of aligning LLMs with RLHF can have the effect of pushing models to converge to the most common view of a given group, collapsing the diversity of opinions held by, for example, different Democrats, into a single modal response (31).”
BackgroundInfluentialLLM-generated messages can persuade humans on policy issues
Bai, H., Voelkel, J. G., Muldowney, S., Eichstaedt, J. C., & Willer, R. (2025) · Nature Communications · doi:10.1038/s41467-025-61345-5 · Cited by 110 · see citing works
Persuasion in Parallel: How Information Changes Minds about Politics
Coppock, A. (2022) · doi:10.7208/chicago/9780226821832.001.0001 · Cited by 91 · cited 1× · see citing works
“However, a body of research suggests that such divergent reactions are rare, and that persuasive political messages tend to sway people, regardless of their demographic traits, in broadly similar ways (23–25).”
BackgroundQuantifying the potential persuasive returns to political microtargeting
Tappin, B. M., Wittenberg, C., Hewitt, L. B., Berinsky, A. J., & Rand, D. G. (2023) · Proceedings of the National Academy of Sciences · doi:10.1073/pnas.2216261120 · Cited by 83 · cited 2× · see citing works
“Attributes were selected for targeting based on a combination of theoretical significance and real - world relevance: their use in previous microtargeting research (28) and also their existence in real - world voter databases a campaign might access (37–39).”
BackgroundTruth, Lies, and Automation: How Language Models Could Change Disinformation
Center for Security and Emerging Technology, Buchanan, B., Lohn, A., Musser, M., & Sedova, K. (2021) · doi:10.51593/2021ca003 · Cited by 74 · cited 2× · see citing works
“This confluence of events has caused great concern over the potential for AI technologies to disproportionately influence voters’ behavior and electoral outcomes (1, 2).”
BackgroundMoral Mimicry: Large Language Models Produce Moral Rationalizations Tailored to Political Identity
Simmons, G. (2023) · Proceedings of the conference - Association for Computational Linguistics. Meeting · doi:10.18653/v1/2023.acl-srw.40 · Cited by 60 · see citing works
Ali, M., Sapieżyński, P., Korolova, A., Mislove, A. E., & Rieke, A. (2021) · ACM International Conference on Web Search and Data Mining (WSDM) · doi:10.1145/3437963.3441801 · Cited by 52 · cited 2× · see citing works
“Further, many documented real - life microtarget-ing operations have attempted to remain covert (17).”
BackgroundTargeted Issue Messages and Voting Behavior
Endres, K. (2019) · American Politics Research · doi:10.1177/1532673x19875694 · Cited by 38 · cited 1× · see citing works
“Several studies on microtargeting found inconclusive or underwhelming evidence supporting its effects (26, 27).”
BackgroundArtificial Intelligence Can Persuade Humans on Political IssuesPreprint
Bai, H., Voelkel, J. G., Eichstaedt, J. C., & Willer, R. (2023) · doi:10.31219/osf.io/stakv · Cited by 38 · see citing works
Ad Delivery Algorithms: The Hidden Arbiters of Political Messaging
Ali, M., Sapiezynski, P., Korolova, A., Mislove, A., & Rieke, A. (2019) · arXiv · doi:10.48550/arxiv.1912.04255 · Cited by 30 · cited 2× · see citing works
“Further, many documented real - life microtarget-ing operations have attempted to remain covert (17).”
BackgroundAmericans Roundly Reject Tailored Political Advertising
Turow, J., Carpini, M. X. D., Draper, N. A., & Howard–Williams, R. (2012) · ScholarlyCommons (University of Pennsylvania) · Cited by 29 · see citing works
Hackenburg, K., Ibrahim, L., Tappin, B. M., & Manos Tsakiris, A. (2025) · AI & Society · doi:10.1007/s00146-025-02464-x · Cited by 25 · cited 1× · see citing works
“Early research suggests that most capable LLMs can directly persuade humans on political issues (5), draft more persuasive public communications than actual government agencies and political communication experts (6, 7), and generate convincing disinfor-mation and fake news (1, 8, 9).”
BackgroundSelf-citationHackenburg, K., Ibrahim, L., Tappin, B. M., et al. (2023) · doi:10.31219/osf.io/ey8db · Cited by 17 · see citing works
Self-citationPolitical microtargeting: Towards a pragmatic approach
Aagaard, P., & Marthedal, S. (2023) · Internet Policy Review · doi:10.14763/2023.1.1690 · Cited by 10 · cited 1× · see citing works
“Attributes were selected for targeting based on a combination of theoretical significance and real - world relevance: their use in previous microtargeting research (28) and also their existence in real - world voter databases a campaign might access (37–39).”
BackgroundCommunityLM: Probing Partisan Worldviews from Language ModelsPreprint
Jiang, H., Beeferman, D., Roy, B., & Roy, D. K. (2022) · arXiv (Cornell University) · doi:10.48550/arxiv.2209.07065 · Cited by 9 · cited 3× · see citing works
“Existing work has shown that fine - tuning can increase model alignment for specific political use cases (33) and it remains unexplored in the micro-targeting domain.”
BackgroundMethodKoto, F., Lau, J. H., & Baldwin, T. (2022) · Workshop on e-Commerce and NLP (ECNLP) · doi:10.18653/v1/2022.ecnlp-1.27 · Cited by 6 · see citing works
Tiwari, A., Saha, S., Sengupta, S., Maitra, A., Ramnani, R., & Bhattacharyya, P. (2022) · Conference of the Asia-Pacific Chapter of the Association for Computational Linguistics and the International Joint Conference on Natural Language Processing (AACL-IJCNLP) · doi:10.18653/v1/2022.aacl-main.76 · Cited by 6 · see citing works
Thousands of AI Authors on the Future of AI
Grace, K., Stewart, H., Sandkühler, J. F., et al. (2024) · arXiv · doi:10.48550/arxiv.2401.02843 · cited 1×
“However, even as researchers and industry leaders list model persuasiveness as a primary safety concern (3, 4), understandings of model capabilities in this domain remain nascent.”
Background
Flagged first, then by citations
Research works printed but not linked: 5
- M. Harrison Dupré, Sam Altman warns that AI is learning ‘Superhuman Persuasion’. Futurism (2023), https://futurism.com/sam-altman-ai-superhuman-persuasion. Accessed 10 February 2024.
- J. Goldstein, G. Sastry, The coming age of AI-powered propaganda. Foreign Affairs (2023).
- BBC, Vote Leave’s Targeted Brexit Ads Released by Facebook (BBC News, 2018). Accessed 13 January 2024.
- M. Scott, Cambridge Analytica helped ‘cheat’ Brexit vote and US election, claims whistleblower. POLITICO, 27 March 2018.
- Open Rights Group, Public are kept in the dark over data driven political campaigning, poll finds. Open Rights Group, 10 January 2020. https://www.openrightsgroup.org/press-releases/public-are-kept-in-the-dark-over-data-driven-political-campaigning-poll-finds/. Accessed 23 January 2024.
Cited by
Sorted by citations
Durably reducing conspiracy beliefs through dialogues with AI
Costello, T. H., Pennycook, G., & Rand, D. G. (2024) · Science · doi:10.1126/science.adq1814 · Cited by 354 · see citing works
On the conversational persuasiveness of GPT-4
Salvi, F., Ribeiro, M. H., Gallotti, R., & West, R. (2025) · Nature Human Behaviour · doi:10.1038/s41562-025-02194-6 · Cited by 326 · see citing works
“After the debate, in Other research has focused on personalization, observing conse-quential yet non-unanimous evidence about the impact of LLMs on microtargeting 47–49 .”
Using natural language processing to analyse text data in behavioural science
Feuerriegel, S., Maarouf, A., Bär, D., Geissler, D., Schweisthal, J., Pröllochs, N., Robertson, C. E., Rathje, S., Hartmann, J., Mohammad, S. M., et al. (2025) · Nature Reviews Psychology · doi:10.1038/s44159-024-00392-z · Cited by 140 · see citing works
“For example, studies published in the past 2 years have used NLP to generate persuasive messages tailored to specific personality profiles 139 and to build therapeutic chatbots 140 .”
Why human–AI relationships need socioaffective alignment
Kirk, H. R., Gabriel, I., Summerfield, C., Vidgen, B., & Hale, S. A. (2025) · Humanities and Social Sciences Communications · doi:10.1057/s41599-025-04532-5 · Cited by 146 · see citing works
Misinformation reloaded? Fears about the impact of generative AI on misinformation are overblown
Simon, F. M., Altay, S., & Mercier, H. (2023) · Harvard Kennedy School Misinformation Review · doi:10.37016/mr-2020-127 · Cited by 85 · see citing works
“Even with advances on these fronts, current evidence suggests that the persuasive effects of microtargeting — including with LLMs — are often limited and highly context-dependent (Hackenburg & Margetts, 2023; Tappin et al., 2023).”
BackgroundLLM-generated messages can persuade humans on policy issues
Bai, H., Voelkel, J. G., Muldowney, S., Eichstaedt, J. C., & Willer, R. (2025) · Nature Communications · doi:10.1038/s41467-025-61345-5 · Cited by 110 · see citing works
Ideation with Generative AI—in Consumer Research and Beyond
Freitas, J. D., Nave, G., & Puntoni, S. (2025) · Journal of Consumer Research · doi:10.1093/jcr/ucaf012 · Cited by 61 · see citing works
“Furthermore, recent studies found that LLMs were more persuasive than humans in domains where swaying opinions is challenging: politics (Hackenburg and Margetts 2024) and belief in conspiracy theories (Costello et al. 2024).”
The Persuasive Power of Large Language Models
Breum, S. M., Egdal, D. V., Gram Mortensen, V., Møller, A. G., & Aiello, L. M. (2024) · Proceedings of the International AAAI Conference on Web and Social Media · doi:10.1609/icwsm.v18i1.31304 · Cited by 57 · see citing works
The levers of political persuasion with conversational artificial intelligence
Hackenburg, K., Tappin, B. M., Hewitt, L., Saunders, E., Black, S., Lin, H., Fist, C., Margetts, H. Z., Rand, D. G., & Summerfield, C. (2025) · Science · doi:10.1126/science.aea3884 · Cited by 59 · see citing works
Self-citationDanry, V., Pataranutaporn, P., Groh, M., & Epstein, Z. (2025) · Proceedings of the 2025 CHI Conference on Human Factors in Computing Systems · doi:10.1145/3706598.3713408 · Cited by 35 · see citing works
How developments in natural language processing help us in understanding human behaviour
Mihalcea, R. F., Biester, L., Boyd, R. L., Jin, Z., Pérez‐Rosas, V., Wilson, S. R., & Pennebaker, J. W. (2024) · Nature Human Behaviour · doi:10.1038/s41562-024-01938-0 · Cited by 34 · see citing works
“One example is political microtarget-ing with LLMs to generate personalized persuasive text 163,180 .”
Scaling language model size yields diminishing returns for single-message political persuasion
Hackenburg, K., Tappin, B. M., Röttger, P., Hale, S. A., Bright, J., & Margetts, H. Z. (2025) · Proceedings of the National Academy of Sciences · doi:10.1073/pnas.2413443122 · Cited by 41 · see citing works
Self-citationBiased AI writing assistants shift users’ attitudes on societal issues
Williams‐Ceci, S., Jakesch, M., Bhat, A., Kadoma, K., Zalmanson, L., & Naaman, M. (2026) · Science Advances · doi:10.1126/sciadv.adw5578 · Cited by 27 · see citing works
Generative propaganda: Evidence of AI’s impact from a state-backed disinformation campaign
Wack, M., Ehrett, C., Linvill, D. L., & Warren, P. L. (2025) · PNAS Nexus · doi:10.1093/pnasnexus/pgaf083 · Cited by 36 · see citing works
Testing theories of political persuasion using AI
Argyle, L. P., Busby, E. C., Gubler, J. R., Lyman, A., Olcott, J., Pond, J., & Wingate, D. (2025) · Proceedings of the National Academy of Sciences · doi:10.1073/pnas.2412815122 · Cited by 25 · see citing works
Hackenburg, K., Ibrahim, L., Tappin, B. M., & Manos Tsakiris, A. (2025) · AI & Society · doi:10.1007/s00146-025-02464-x · Cited by 25 · see citing works
“Notably, in a supplementary analysis we also find that a role-playing LLM was highly effective at persuading Democrats on issues they would normally oppose (see Supplementary Materials Section 6).”
Self-citationThe Use of AI by Election Campaigns
Foos, F. (2024) · LSE Public Policy Review · doi:10.31389/lseppr.112 · Cited by 23 · see citing works
Frontier AI developers need an internal audit function
Schuett, J. (2024) · Risk Analysis · doi:10.1111/risa.17665 · Cited by 32 · see citing works
“For example, models might become able to deceive, persuade, and manipulate people (Carlsmith, 2023; El-Sayed et al., 2024; Hackenburg & Margetts, 2024; Hagendorff, 2024; Hubinger et al., 2024; Park et al., 2023), find and exploit cyber vulnerabilities (Fang et al., 2024; Lohn & Jackson, 2022; Mirsky…”
How do people react to political bias in generative artificial intelligence (AI)?
Messer, U. (2024) · Computers in Human Behavior Artificial Humans · doi:10.1016/j.chbah.2024.100108 · Cited by 24 · see citing works
Understanding model power in social AI
Brandtzæg, P. B., Skjuve, M. B., & Følstad, A. (2024) · AI & Society · doi:10.1007/s00146-024-02053-4 · Cited by 20 · see citing works
“…when Social AI influences users’ decision-making, their understanding of the world, and political or ideological stances (Bai et al. 2023; Goldstein et al. 2024; Hackenburg and Margetts 2023; Palmer et al. Palmer and Spirling 2023), as well as moral and ethical judgments (Krugel et al. 2023).”
Yan, H. Y., Morrow, G., Yang, K.-C., & Wihbey, J. (2025) · Harvard Kennedy School Misinformation Review · doi:10.37016/mr-2020-171 · Cited by 18 · see citing works
“However, some research suggests the effectiveness of AI-generated political messages may not always rely on personalization (Hackenburg & Margetts, 2023), and the impact can vary depending on the measures used (Simchon et al., 2024).”
Williams, A. R., Burke-Moore, L., Chan, R. S.-Y., Enock, F. E., Nanni, F., Sippy, T., Chung, Y.-L., Gabašová, E., Hackenburg, K., & Bright, J. (2025) · PLoS ONE · doi:10.1371/journal.pone.0317421 · Cited by 17 · see citing works
“To understand how persuasive LLMs are when microtargeted to individuals on political issues, Hackenburg et al. [35] integrated user data into GPT-4 prompts, and found that, although persuasive, microtargeted messages were not statistically more persuasive than non-targeted messages.”
Self-citationThe impact of advanced AI systems on democracy
Summerfield, C., Argyle, L. P., Bakker, M. A., Collins, T., Durmus, E., Eloundou, T., Gabriel, I., Ganguli, D., Hackenburg, K., Hadfield, G. K., et al. (2025) · Nature Human Behaviour · doi:10.1038/s41562-025-02309-z · Cited by 41 · see citing works
Self-citationSong, T.-Q., Tan, Y., Zhu, Z.-C., Feng, Y.-B., & Lee, Y.-C. (2024) · Proc. ACM Hum. Comput. Interact. · doi:10.1145/3757633 · Cited by 36 · see citing works
Labeling messages as AI-generated does not reduce their persuasive effects
Gallegos, I. O., Shani, C., Shi, W., Bianchi, F., Gainsburg, I., Jurafsky, D., & Willer, R. (2026) · PNAS Nexus · doi:10.1093/pnasnexus/pgag008 · Cited by 16 · see citing works
“Even though 92.0% of participants assigned to the AI and human label conditions believed the authorship labels, this filtered analysis ensured that we retained only those participants who were successfully influenced by the manipulation ( N = 1,515).”
InfluentialArtificial Intelligence in Election Campaigns: Perceptions, Penalties, and Implications
Jungherr, A. M., Rauchfleisch, A., & Wuttke, A. (2026) · Political Communication · doi:10.1080/10584609.2025.2611913 · Cited by 15 · see citing works
“AI is also used to enhance voter outreach by identifying likely supporters, testing and optimizing message content, and delivering personalized appeals across digital platforms (e. g., Burke & Sunderman, 2024; Chatterjee, 2024; Hackenburg & Margetts, 2024).”
Balancing Large Language Model Alignment and Algorithmic Fidelity in Social Science Research
Lyman, A., Hepner, B., Argyle, L. P., Busby, E. C., Gubler, J. R., & Wingate, D. (2025) · Sociological Methods & Research · doi:10.1177/00491241251342008 · Cited by 19 · see citing works
“Many researchers currently use LLMs to present respondents with a persuasive message (Argyle et al. 2024; Hackenburg and Margetts 2024; Palmer and Spirling 2023); in these circumstances, they might want that message to follow a particular template or to hold speci fi c characteristics (e.g., length…”
Vasanthan, R., Babu, M. R. N., Dzuvichu, K., Chong, S. T., Rana, V. S., & Singh, A. (2025) · Enterprise Development and Microfinance · doi:10.3362/1755-1986.25-00009 · Cited by 14 · see citing works
We need to understand “when” not “if” generative AI can enhance personalized persuasion
Teeny, J. D., & Matz, S. (2024) · Proceedings of the National Academy of Sciences · doi:10.1073/pnas.2418005121 · Cited by 12 · see citing works
“In a recent article, Hackenburg and Margetts ( 1 ) empirically inves - tigated this question by using ChatGPT - 4 to dynamically per - sonalize political messages to one or more sociodemographic characteristics, concluding that personalization is an ineffec - tive strategy and that gen - AI is…”
BackgroundRelational AI: Facilitating Intergroup Cooperation with Socially Aware Conversational Support
Claggett, E. L., Kraut, R. E., & Shirado, H. (2025) · CHI Conference on Human Factors in Computing Systems · doi:10.1145/3706598.3713757 · Cited by 11 · see citing works
“While LLMs offer great flexibility in generating messages tailored to individuals, this personalization can risk reinforcing group differences and in-group biases [63], as observed in online social media [56, 121, 129].”
Karell, D., Sachs, J. D., & Barrett, R. (2024) · Poetics · doi:10.1016/j.poetic.2024.101966 · Cited by 10 · see citing works
Triantafyllopoulos, A., Spiesberger, A. A., Tsangko, I., Jing, X., Distler, V., Dietz, F., Alt, F., & Schuller, B. W. (2025) · Computer Speech & Language · doi:10.1016/j.csl.2025.101802 · Cited by 14 · see citing works
How AI sources can increase openness to opposing views
Lu, L., Tormala, Z. L., & Duhachek, A. (2025) · Scientific Reports · doi:10.1038/s41598-025-00791-z · Cited by 9 · see citing works
Generative AI outperforms humans in social media engagement: Evidence from GPT-4 and the FIIT model
Huang, J., & Zhou, A. Y. (2025) · Public Relations Review · doi:10.1016/j.pubrev.2025.102643 · Cited by 11 · see citing works
Nezhad, M. M., Kisomi, M. A., & Gholinezhad, F. (2025) · International Conference on Web Research (ICWR) · doi:10.1109/icwr65219.2025.11006192 · Cited by 7 · see citing works
“In contrast, other research suggests that their performance is comparable to human-level persuasion [16-18].”
People defer to AI moral advice, but not blindly
Landes, E., Francis, K., & Everett, J. A. C. (2026) · Cognition · doi:10.1016/j.cognition.2026.106504 · Cited by 7 · see citing works
Contextualized Counterspeech: Strategies for Adaptation, Personalization, and Evaluation
Cima, L., Miaschi, A., Trujillo, A., Avvenuti, M., Dell’Orletta⋄, F., & Cresci, S. (2025) · International World Wide Web Conference (WWW) · doi:10.1145/3696410.3714507 · Cited by 6 · see citing works
Exploring Laws of Robotics: A Synthesis of Constitutional AI and Constitutional Economics
Küsters, A., & Wörsdörfer, M. (2025) · Digital Society · doi:10.1007/s44206-025-00204-8 · Cited by 6 · see citing works
“…for LLMs to impact the integrity of information and shape democratic discourse, e.g., by fine-tuning the micro-targeting of voters via social media (Hackenburg & Margetts, 2024) or through automated influence operations (Goldstein et al., 2023 ), it is vital that legal and ethical guidelines for…”
Meyer, M., Enders, A., Klofstad, C. A., Stoler, J., & Uscinski, J. E. (2024) · Harvard Kennedy School Misinformation Review · doi:10.37016/mr-2020-164 · Cited by 14 · see citing works
Zhu, J., Kempermann, M., Cannanure, V. K., Hartland, A., Navarrete, R. M., Carteny, G., Braun, D., & Weber, I. G. (2025) · ACM Conference on Conversational User Interfaces (CUI) · doi:10.1145/3719160.3736611 · Cited by 6 · see citing works
“Some survey responses suggested adding more GUI elements to streamline the responses, such as a button to indicate high interest in a particular topic.”
Arti-‘fickle’ intelligence: using LLMs as a tool for inference in the political and social sciences
Argyle, L. P., Busby, E. C., Gubler, J. R., Hepner, B., Lyman, A., & Wingate, D. (2025) · Nature Computational Science · doi:10.1038/s43588-025-00843-4 · Cited by 13 · see citing works
“As a further example, another set of studies have substantial overlap in content and design, but feature LLMs used to create and administer some kind of intervention or treatment to human participants 11–14 .”
State media control influences large language models
Waight, H., Yang, E., Yuan, Y., Messing, S., Roberts, M. E., Stewart, B. M., & Tucker, J. A. (2026) · Nature · doi:10.1038/s41586-026-10506-7 · Cited by 9 · see citing works
Piao, J., Lu, Z., Gao, C., & Li, Y. (2025) · International World Wide Web Conference (WWW) · doi:10.1145/3696410.3714537 · Cited by 4 · see citing works
Comparison of Large Language Model with Aphasia
Watanabe, T., Inoue, K., Kuniyoshi, Y., Nakajima, K., & Aihara, K. (2025) · Advanced Science · doi:10.1002/advs.202414016 · Cited by 5 · see citing works
Impact of Agent-Generated Rationales on Online Social Conformity
de Jong, S., Jacobsen, R. M., Wester, J., Wijenayake, S., Gonçalves, J., & van Berkel, N. (2025) · ACM Conference on Fairness, Accountability, and Transparency (FAccT) · doi:10.1145/3715275.3732217 · Cited by 5 · see citing works
Rethinking news framing with large language models
Tohidi, A., Haider, S., & Watts, D. J. (2025) · Scientific Reports · doi:10.1038/s41598-025-29519-9 · Cited by 4 · see citing works
Sehgal, N. K. R., Rai, S., Tonneau, M., Agarwal, A. K., Cappella, J. N., Kornides, M. L., Ungar, L., Buttenheim, A. M., & Guntuku, S. C. (2026) · JAMA Network Open · doi:10.1001/jamanetworkopen.2026.16822 · Cited by 4 · see citing works
Jones, C., & Bergen, B. K. (2026) · Artificial Intelligence Review · doi:10.1007/s10462-026-11517-6 · Cited by 17 · see citing works
Intelligent Systems, Vulnerable Minds: A Framework for Radicalization to Violence in the Age of AI
Kunst, J., Obaidi, M., Gollwitzer, A., Brandtzæg, P., Hinrichs, Y., Saini, N., & Schroeder, D. (2026) · Personality and Social Psychology Review · doi:10.1177/10888683261430089 · Cited by 5 · see citing works
“Regarding generative AI, the per-suasive content capabilities of LLMs are well-established (Hackenburg & Margetts, 2024; Hackenburg et al., 2025; Williams et al., 2025).”
A polycrisis threat model for AI
Bales, A. (2025) · AI & Society · doi:10.1007/s00146-025-02371-1 · Cited by 4 · see citing works
“12 More work is needed to reach conclusions about the size of any effect here (Tappin et al. 2023; Simchon et al. 2024; Hackenburg and Margetts 2024; Salvi et al. 2024).”
The persuasive potential of AI-paraphrased information at scale
Dash, S., Xu, Y., Jalbert, M. C., & Spiro, E. S. (2025) · PNAS Nexus · doi:10.1093/pnasnexus/pgaf207 · Cited by 3 · see citing works
Matthes, J., & Vranken, S. (2025) · Communication and Change · doi:10.1007/s44382-025-00019-8 · Cited by 3 · see citing works
“Various scholars have used GAI tools to create stimuli, including text-based micro-targeting (Hackenburg & Margetts, 2024), text-based health messages (Karinshak et al., 2023; Schmälzle & Wilcox, 2022), visual ads (van Berlo et al., 2024) or even deepfakes (Eberl et al., 2022).”
Political persuasion by artificial intelligence
Argyle, L. P. (2025) · Science · doi:10.1126/science.aec9293 · Cited by 3 · see citing works
Karanja, J. M., & Musili, M. M. (2026) · Social Science Computer Review · doi:10.1177/08944393261416781 · Cited by 6 · see citing works
Bozdag, N. B., Mehri, S., Tür, G., & Hakkani-Tur, D. Z. (2026) · ACM Conference on AI and Agentic Systems (CAIS) · doi:10.1145/3786335.3813181 · Cited by 9 · see citing works
“LLMs have been applied for social good, promoting public health and prosocial behaviors while adhering to ethical, non-coercive principles (Wang et al., 2019; Furumai et al., 2024; Karinshak et al., 2023; Kampik et al., 2018).”
AI Realtor: Towards Grounded Persuasive Language Generation for Automated Copywriting
Wu, J., Yang, C., Wu, Y., Mahns, S., Wang, C., Zhu, H., Fang, F., & Xu, H. (2026) · ACM Conference on AI and Agentic Systems (CAIS) · doi:10.1145/3786335.3813172 · Cited by 3 · see citing works
“All human-subject experiments were conducted in compliance with ethical research standards.”
Party Equalization or Normalization Through Visual Generative AI in the 2025 German Federal Election
Kruschinski, S., & Votta, F. A. (2026) · Media and Communication · doi:10.17645/mac.11859 · Cited by 3 · see citing works
“By adjusting style, symbolism, and narrative cues, campaigns can create visuals that maximize resonance and relevance for different demographics (Hackenburg & Margetts, 2024).”
Ghodke, P., Katoch, S., Tarwal, S., Bedekar, P. P., Murthy, D. N., & Kaiwade, A. (2025) · Enterprise Development and Microfinance · doi:10.3362/1755-1986.25-00006 · Cited by 3 · see citing works
Wu, X., Whittington, D., Chen, Y. J., & Zuo, R. (2024) · Journal of Asian Public Policy · doi:10.1080/17516234.2024.2425874 · Cited by 3 · see citing works
“…AI has been explored in the context of political microtargeting, with findings suggesting that the influence of current LLMs may reside more in the persuasiveness of their generic, non-targeted messages rather than their ability to tailor messages to individuals (Hackenburg & Margetts, 2024).”
Beyond partisan leaning: a comparative analysis of political bias in large language models
Peng, T., Yang, K., Lee, S., Li, H., Chu, Y., Lin, Y., & Liu, H. (2026) · Journal of Information Technology & Politics · doi:10.1080/19331681.2026.2646990 · Cited by 10 · see citing works
Li, H., Wang, Q., & Wu, Y. (2025) · Mathematics · doi:10.3390/math13193062 · Cited by 4 · see citing works
Xu, S., Zhou, Z., & Zhao, N. (2026) · Frontiers in Psychiatry · doi:10.3389/fpsyt.2026.1756792 · Cited by 2 · see citing works
The seven roles of generative AI: Potential & pitfalls in combatting misinformation
Nygren, T., Spearing, E. R., Fay, N., Vega, D., Hardwick, I. I., Roozenbeek, J., & Ecker, U. K. H. (2026) · Behavioral Science & Policy · doi:10.1177/23794607261417815 · Cited by 6 · see citing works
Inteligencia artificial y elecciones: un experimento en el balotaje ecuatoriano 2025
Curvale, C. (2026) · Revista Mexicana de Derecho Electoral · doi:10.22201/iij.24487910e.2026.24.20952 · Cited by 2 · see citing works
The “synthetic validity” problem in AI-generated science
Teeny, J. D., Luttrell, A., & Lee, D. (2026) · Proceedings of the National Academy of Sciences · doi:10.1073/pnas.2603289123 · Cited by 2 · see citing works
Estimating divergent moral and diversity preferences between AI builders and AI users
Purcell, Z. A., Charbit, L., Borst, G., & Nußberger, A.-M. (2025) · Cognition · doi:10.1016/j.cognition.2025.106198 · Cited by 2 · see citing works
“…research has shown that AI systems such as LLMs do reflect the ideologies of their builders (Buyl et al., 2024) and also that these systems can shape beliefs, behaviors, and political views (Costello, Pennycook, & Rand, 2024; Hackenburg & Margetts, 2024; Karinshak, Liu, Park, & Hancock, 2023).”
BUILDING GENAI-DRIVEN WEB APPLICATIONS FOR MARKETING RESEARCH: A METHODOLOGICAL GUIDEPreprint
Jörling, M. (2025) · SSRN Electronic Journal · doi:10.2139/ssrn.5256590 · Cited by 2 · see citing works
Sarhan, H., Shahrezaye, M., & Hegelich, S. (2025) · AI and Ethics · doi:10.1007/s43681-025-00773-x · Cited by 1 · see citing works
Efficacy of Personality-Labeled Prompts in Generating Trait-Specific Persuasive Messages
Xu, S., & Zhao, N. (2025) · Lecture notes in electrical engineering · doi:10.1007/978-981-95-2113-5_15 · Cited by 1 · see citing works
When is political advocacy effective? Understanding the goals and actors of political advocacy
Hussein, M. A., & Catapano, R. (2025) · Consumer Psychology Review · doi:10.1002/arcp.70002
Characterizing Persuasion Patterns in Climate Discourse on Brazilian Portuguese YouTube Videos
Dong, W., Locatelli, M. S., Almeida, V., & Cha, M. (2025) · International Conference on Information Technology for Social Good (GoodIT) · doi:10.1145/3748699.3749799 · Cited by 1 · see citing works
How latent and prompting biases in AI-generated historical narratives influence opinions
Shu, M., Karell, D., Okura, K., & Davidson, T. (2026) · PNAS Nexus · doi:10.1093/pnasnexus/pgag022 · Cited by 1 · see citing works
The opportunities and limits of microtargeting cross-pressured voters
Moniz, P., Endres, K., & Panagopoulos, C. (2026) · Research & Politics · doi:10.1177/20531680261432765 · Cited by 1 · see citing works
“Third, research fi nds scant support for certain forms of heterogeneous treatment effects of cam-paign messages (Coppock et al. 2020; Hackenburg and Margetts 2024).”
Díaz, F., Cerna, N., Liza, R., & Motta, B. (2026) · Information · doi:10.3390/info17030292 · Cited by 1 · see citing works
Measuring Political Stance and Consistency in Large Language Models
Alali, S. F., Maasfeh, M. N., Kutlu, M., & Kardaş, Ş. (2026) · Lecture notes in computer science · doi:10.1007/978-3-032-21324-2_34 · Cited by 1 · see citing works
AI meets politics: Examining the effects of different targeting strategies across 15 countries
Kruikemeier, S., Schäfer, S., Hamilton, A., Guldemond, P., Vrielink, J., Dymanus, C., Remoortere, A. V., Tamboer, S. L., Vliegenthart, R., Vermeer, S. A. M., et al. (2026) · New Media & Society · doi:10.1177/14614448261445063 · Cited by 1 · see citing works
“This is in line with others, who argue that LLM-targeting could be persuasive, but not more persuasive than non-LLM-targeting (Hackenburg and Margetts, 2024).”
Lynch, C. J., Jensen, E., Gore, R. J., Zamponi, V., O’Brien, K., Feldhaus, B., Smith, K., Martinez, J. P., Munro, M. H., Ozkose, T. E., et al. (2025) · IEEE Access · doi:10.1109/access.2025.3600146 · Cited by 2 · see citing works
“Recent work reveals that LLM-generated messages can influence human opinions on complex policy issues [15] and can be leveraged for persuasive political messaging at scale [71].”
Reply to Teeny and Matz: Toward the robust measurement of personalized persuasion with generative AI
Hackenburg, K., & Margetts, H. Z. (2024) · Proceedings of the National Academy of Sciences · doi:10.1073/pnas.2418817121 · Cited by 1 · see citing works
Self-citationSelective Facts Can Be As Persuasive As FalsehoodsPreprint
Allen, J. N. L., Tohidi, A., Haider, S., Rothschild, D. M., & Watts, D. (2025) · SSRN Electronic Journal · doi:10.2139/ssrn.5400466 · Cited by 1 · see citing works
Linguistic traces of stochastic empathy in language models
Kleinberg, B., Zegers, J., Festor, J., Vida, S., Präsent, J., Loconte, R., & Peereboom, S. (2024) · arXiv · doi:10.48550/arxiv.2410.01675 · Cited by 1 · see citing works
“…help of large language models, and recent evidence indicates that LLM-generated content is highly effective in influencing opinion – both in an encouraging form shown to dampen conspiratorial thinking in humans [4] and more concerningly, in the form of persuasive messaging on political issues [5].”
Ko, H., & Williams, N. W. (2025) · SSRN Electronic Journal · doi:10.2139/ssrn.5118151
Exploring Laws of Robotics: A Synthesis of Constitutional AI and Constitutional EconomicsPreprint
Wörsdörfer, M., & Küsters, A. (2025) · SSRN Electronic Journal · doi:10.2139/ssrn.5248374
Gollust, S. E., Kim, Y., Kissell, P., & Fowler, E. F. (2025) · Journal of Health Politics Policy and Law · doi:10.1215/03616878-11995192
Utilizing Generative AI to Boost Public Support for Carbon Pricing
Molnár, A., & Mermelstein, I. (2026) · Hawaii International Conference on System Sciences (HICSS) · doi:10.24251/hicss.2026.090
Utilizing Generative AI to Boost Public Support for Carbon PricingPreprint
Molnár, A., & Mermelstein, I. (2025) · SSRN Electronic Journal · doi:10.2139/ssrn.5325959
Munibi, A. Z. (2025) · Jurnal Multidisiplin West Science · doi:10.58812/jmws.v4i07.2439
Cebrián, M. (2025) · Mètode: Revista de difusió de la investigació de la Universitat de Valencia · doi:10.7203/metode.15.30429
Generating the Past: How Artificial Intelligence Summaries of Historical Events Affect Knowledge
Karell, D., Shu, M., Davidson, T., & Okura, K. (2025) · Social Science Computer Review · doi:10.1177/08944393251409744 · Cited by 1 · see citing works
Bring Wisdom into AI Education (Before it’s Too Late)
Williams‐Ceci, S., & Williams, W. M. (2026) · Roeper Review · doi:10.1080/02783193.2025.2593014
“While GPT successfully shifted participants’ opinions on these issues, the treatment effect was not significantly different depending on whether the LLM knew the demographics and attitudes of the recipients of its messaging (Hackenburg & Margetts, 2024).”
Potencjał sztucznej inteligencji (AI) w budowaniu skutecznych narracji politycznych
Darmoń, K., & Marmola, M. (2025) · Studia Politicae Universitatis Silesiensis · doi:10.31261/spus.22851
Thakur, A., Murray, T., Rodriguez, P., & Hilbert, M. (2026) · SSRN Electronic Journal · doi:10.2139/ssrn.6083909
Cognitive effects of political microtargeting: an eye-tracking study
Noetzel, S., Binder, A., & Matthes, J. (2025) · Media Psychology · doi:10.1080/15213269.2025.2527612
“In other words, similar to other research in the field, we compare accurate to inaccurate microtargeting (e.g., Decker & Krämer, 2023; Hackenburg & Margetts, 2024; Hirsch et al., 2023; Simchon et al., 2024; Zarouali et al., 2020).”
Generative AI for climate governance and acceptability-constrained policy design
Manivannan, A., Spaiser, V., Cann, T. J. B., Evans, J., Everall, J., Falkenberg, M., García, D., Guo, W., Herzog, R., Otto, I. M., et al. (2026) · npj Climate Action · doi:10.1038/s44168-026-00362-6 · Cited by 1 · see citing works
Large language models as a conduit for value shifts in contemporary China
Cao, L., & Dai, L. (2026) · Chinese Sociological Review · doi:10.1080/21620555.2026.2656193
Synthetic political actors: an exploration into emerging AI-driven candidates and parties
Malý, M., Sandri, G., & Barberá, Ó. (2026) · Open Research Europe · doi:10.12688/openreseurope.23686.1
“Most studies examine its role in electoral campaigning (Foos 2024, Núñez et al. 2025), political communication (Hackenburg and Margetts 2024), and voter persuasion (Lin et al. 2025), while campaign-focused scholarship has also analysed the spread of AI-generated synthetic content, including…”
Opportunities and limitations of digitalization for the development of political consulting
Fedorchenko, L. V. (2026) · Journal of Political Research · doi:10.12737/2587-6295-2026-10-1-228-237
Selective Persuasion: Structural Stance Bias of Generative AI in Controversial Issues
黎, 樟., & 塔, 娜. (2026) · Global Journal of Media Studies · doi:10.26599/gjms.2026.9330009 · Cited by 1 · see citing works
Chen, Z., Kalla, J., Le, Q., Nakamura-Sakai, S., Sekhon, J., & Wang, R. (2026) · Journal of Experimental Political Science · doi:10.1017/xps.2026.10032 · Cited by 2 · see citing works
Hedging and Non-Affirmation: Quantifying LLM Alignment on Questions of Human Rights
Javed, R., Parent, C., Kay, J., Yanni, D., Zaini, A., Sheikh, A., Rauh, M., Gerych, W., Comanescu, R., Gabriel, I., et al. (2026) · ACM Conference on Fairness, Accountability, and Transparency (FAccT) · doi:10.1145/3805689.3806734
Zhang, Z. W., Morgan, A. J., Armstrong, G., & Reavley, N. J. (2026) · Human Behavior and Emerging Technologies · doi:10.1155/hbe2/9069016
- Authors
- Hackenburg, K., & Margetts, H.
- Affiliations
- University of Oxford (GB)
- Corresponding author
- Hackenburg, Kobi
- Venue
- Proceedings of the National Academy of Sciences
- Publisher
- National Academy of Sciences
- Published
- 2024-06-07
- Volume
- 121
- Issue
- 24
- Pages
- e2403116121-e2403116121
- ISSN
- 0027-8424 (ISSN-L)
- 1091-6490
- PMC
- PMC11181035
- OpenAlex
- W4399426804
- Cited by
- 114
- References
- 33
- Citations by year
- 2026: 47 · 2025: 54 · 2024: 12 · 2023: 1
- FWCI
- 94.38
- Type
- journal article
- Language
- en
- Open access
- hybrid
- doi.org
- Licence
- Locations
- Proceedings of the National Academy of Sciences (publishedVersion, cc-by)
- PubMed Central (publishedVersion, cc-by)
- PubMed (publishedVersion)
- OSF Preprints (OSF Preprints) (submittedVersion)
- OSF Preprints (OSF Preprints) (submittedVersion, cc-by)
- Topics
- Social Sciences › Communication › Social Media and Politics
- Social Sciences › Sociology and Political Science › Media Influence and Politics
- Social Sciences › Political Science and International Relations › Electoral Systems and Political Participation
- Keywords
- political microtargeting, large language models, GPT-4, persuasion, microdata, message tailoring, political communication
- MeSH terms
- Humans, Language, Persuasive Communication, Politics
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
