Leveraging AI for democratic discourse: Chat interventions can improve online political conversations at scale
Open accessAbstract
Political discourse is the soul of democracy, but misunderstanding and conflict can fester in divisive conversations. The widespread shift to online discourse exacerbates many of these problems and corrodes the capacity of diverse societies to cooperate in solving social problems. Scholars and civil society groups promote interventions that make conversations less divisive or more productive, but scaling these efforts to online discourse is challenging. We conduct a large-scale experiment that demonstrates how online conversations about divisive topics can be improved with AI tools. Specifically, we employ a large language model to make real-time, evidence-based recommendations intended to improve participants’ perception of feeling understood. These interventions improve reported conversation quality, promote democratic reciprocity, and improve the tone, without systematically changing the content of the conversation or moving people’s policy attitudes.
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?
In prior research by the authors, LLMs improved the tone of politically fraught conversation, leading to improvement in participants' democratic reciprocity that was not dependent on a change in their policy attitudes.
“In prior research, our team demonstrated that LLMs can be used to improve the tone of politically fraught conversation, leading to improvement in participants’ democratic reciprocity that is not dependent on a change in their policy attitudes (7).”
From Testing theories of political persuasion using AI (2025) · cites this paper as [7]
LLMs have considerable potential for coding the topic of unstructured conversations between multiple people using a mobile chat platform.
“Argyle et al. also demonstrate that LLMs have considerable potential for coding the topic of unstructured conversations between multiple people using a mobile chat platform (53).”
From Can Generative AI improve social science? (2024) · cites this paper as [53]
Argyle et al. found that an intervention where GPT-3 rephrased messages made conversations about gun control more productive and less stressful for those whose partner used GPT-3 recommendations.
“The researchers found this intervention made conversations about gun control more productive and less stressful for those whose partner used recommendations from GPT-3.”
From Can Generative AI improve social science? (2024) · cites this paper as [53]
In Argyle et al., researchers did not force participants to accept GPT-3 rephrasings; participants could choose, edit, or reject them.
“Furthermore, the researchers did not force human participants to accept the rephrasings proposed by GPT-3; rather, they were allowed to choose from several of them, edit their original message, or reject all of them.”
From Can Generative AI improve social science? (2024) · cites this paper as [53]
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
Language Models are Few-Shot Learners
Brown, T. B., Mann, B., Ryder, N., Subbiah, M., Kaplan, J., Dhariwal, P., Neelakantan, A., Shyam, P., Sastry, G., Askell, A., et al. (2020) · arXiv · doi:10.48550/arxiv.2005.14165 · Cited by 23,372 · see citing works
Dissecting racial bias in an algorithm used to manage the health of populations
Obermeyer, Z., Powers, B., Vogeli, C., & Mullainathan, S. (2019) · Science · doi:10.1126/science.aax2342 · Cited by 7,137 · see citing works
On the Dangers of Stochastic Parrots: Can Language Models Be Too Big? 🦜
Bender, E. M., Gebru, T., McMillan-Major, A., & Shmitchell, S. (2021) · Proceedings of the 2021 ACM Conference on Fairness, Accountability, and Transparency · doi:10.1145/3442188.3445922 · Cited by 6,988 · see citing works
Semantics derived automatically from language corpora contain human-like biases
Caliskan, A., Bryson, J. J., & Narayanan, A. (2017) · Science · doi:10.1126/science.aal4230 · Cited by 2,887 · see citing works
Gutmann, A., & Thompson, D. F. (2004) · Princeton University Press eBooks · doi:10.1515/9781400826339 · Cited by 2,667 · see citing works
Exposure to opposing views on social media can increase political polarization
Bail, C. A., Argyle, L. P., Brown, T. W., Bumpus, J. P., Chen, H., Hunzaker, M. B. F., Lee, J., Mann, M., Merhout, F., & Volfovsky, A. (2018) · Proceedings of the National Academy of Sciences · doi:10.1073/pnas.1804840115 · Cited by 1,863 · see citing works
Self-citationHearing the Other Side: Deliberative versus Participatory Democracy
Mutz, D. C. (2006) · doi:10.1017/cbo9780511617201 · Cited by 1,687 · see citing works
Field Experiments: Design, Analysis, and Interpretation
GREEN, D. P. (2024) · doi:10.61700/sgeirzx0v0mjh1252 · Cited by 1,381 · see citing works
Cross-cutting Social Networks: Testing Democratic Theory in Practice
MUTZ, D. C. (2002) · American Political Science Review · doi:10.1017/s0003055402004264 · Cited by 1,069 · see citing works
Human Decisions and Machine Predictions
Kleinberg, J., Lakkaraju, H., Leskovec, J., Ludwig, J., & Mullainathan, S. (2018) · The Quarterly Journal of Economics · doi:10.1093/qje/qjx032 · Cited by 1,040 · see citing works
Political Sectarianism in America
Finkel, E. J., Bail, C. A., Cikara, M., Ditto, P. H., Iyengar, S., Klar, S., Mason, L., McGrath, M. C., Nyhan, B., Rand, D. G., et al. (2020) · Science · doi:10.1126/science.abe1715 · Cited by 931 · see citing works
Self-citationDurably reducing transphobia: A field experiment on door-to-door canvassing
Broockman, D., & Kalla, J. (2016) · Science · doi:10.1126/science.aad9713 · Cited by 798 · see citing works
Out 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 · see citing works
Self-citationMeasuring Political Deliberation: A Discourse Quality Index
Steenbergen, M. R., Bächtiger, A., Spörndli, M., & Steiner, J. (2003) · Comparative European Politics · doi:10.1057/palgrave.cep.6110002 · Cited by 724 · see citing works
The Public and Its Problems: An Essay in Political Inquiry
Dewey, J., & Rogers, M. L. (2016) · University of Michigan Library Repository · Cited by 717 · see citing works
The contact hypothesis re-evaluated
Paluck, E., Green, D. P., & Green, S. (2018) · Behavioural Public Policy · doi:10.1017/bpp.2018.25 · Cited by 711 · see citing works
Prejudice Reduction: Progress and Challenges
Paluck, E. L., Porat, R., Clark, C. S., & Green, D. P. (2021) · Annual Review of Psychology · doi:10.1146/annurev-psych-071620-030619 · Cited by 686 · see citing works
Human Decisions and Machine Predictions
Kleinberg, J., Lakkaraju, H., Leskovec, J., Ludwig, J., & Mullainathan, S. (2017) · doi:10.3386/w23180 · Cited by 545 · see citing works
Artificial intelligence and algorithmic bias: implications for health systems
Panch, T., Mattie, H., & Atun, R. (2019) · Journal of Global Health · doi:10.7189/jogh.09.020318 · Cited by 533 · see citing works
Sharma, A., Lin, I. W., Miner, A. S., Atkins, D. C., & Althoff, T. (2023) · Nature Machine Intelligence · doi:10.1038/s42256-022-00593-2 · Cited by 500 · see citing works
Harlow, S. (2011) · New Media & Society · doi:10.1177/1461444811410408 · Cited by 440 · see citing works
Scaling Social Movements Through Social Media: The Case of Black Lives Matter
Mundt, M. D., Ross, K., & Burnett, C. M. (2018) · Social Media + Society · doi:10.1177/2056305118807911 · Cited by 403 · see citing works
Tweetment Effects on the Tweeted: Experimentally Reducing Racist Harassment
Munger, K. (2017) · Political Behavior · doi:10.1007/s11109-016-9373-5 · Cited by 395 · see citing works
Frenemies: How Social Media Polarizes America
Settle, J. E. (2018) · doi:10.1017/9781108560573 · Cited by 377 · see citing works
Kalmoe, N. P., & Mason, L. (2022) · doi:10.7208/chicago/9780226820279.001.0001 · Cited by 373 · see citing works
Does affective polarization undermine democratic norms or accountability? Maybe not
Broockman, D. E., Kalla, J. L., & Westwood, S. J. (2023) · American Journal of Political Science · doi:10.1111/ajps.12719 · Cited by 297 · see citing works
Jacobs, L. R., Cook, F. L., & Carpini, M. X. D. (2009) · University of Chicago Press eBooks · doi:10.7208/chicago/9780226389899.001.0001 · Cited by 290 · see citing works
Dobson, A. (2014) · Oxford University Press eBooks · doi:10.1093/acprof:oso/9780199682447.001.0001 · Cited by 256 · see citing works
Wojcieszak, M., & Warner, B. R. (2020) · Political Communication · doi:10.1080/10584609.2020.1760406 · Cited by 232 · see citing works
Toward understanding understanding : The importance of feeling understood in relationships
Reis, H. T., Lemay, E. P., & Finkenauer, C. (2017) · Social and Personality Psychology Compass · doi:10.1111/spc3.12308 · Cited by 211 · see citing works
Social Media, News Media and the Stock Market
Jiao, P., Veiga, A., & Walther, A. (2016) · Journal of Economic Behavior & Organization · doi:10.1016/j.jebo.2020.03.002 · Cited by 192 · see citing works
Responsiveness Beyond Policy Satisfaction: Does It Matter to Citizens?
Esaiasson, P., Gilljam, M., & Persson, M. (2016) · Comparative Political Studies · doi:10.1177/0010414015626445 · Cited by 187 · see citing works
Breaking the social media prism: how to make our platforms less polarizing
Sufi, M. (2023) · Information Communication & Society · doi:10.1080/1369118x.2023.2258608 · Cited by 187 · see citing works
Krupnikov, Y., & Ryan, J. B. (2022) · doi:10.1017/9781108923323 · Cited by 175 · see citing works
Interventions to reduce partisan animosity
Hartman, R., Blakey, W., Womick, J., Bail, C., Finkel, E. J., Han, H., Sarrouf, J., Schroeder, J., Sheeran, P., Van Bavel, J. J., et al. (2022) · Nature Human Behaviour · doi:10.1038/s41562-022-01442-3 · Cited by 171 · see citing works
Self-citationPolitical Decision Making, Deliberation and Participation
Carpini, M. X. D. (2002) · Cited by 152 · see citing works
A review and provocation: On polarization and platforms
Kreiss, D., & McGregor, S. C. (2024) · New Media & Society · doi:10.1177/14614448231161880 · Cited by 151 · see citing works
Conversational receptiveness: Improving engagement with opposing views
Yeomans, M., Minson, J. A., Collins, H. K., Chen, F. S., & Francesca Gino, B. (2020) · Organizational Behavior and Human Decision Processes · doi:10.1016/j.obhdp.2020.03.011 · Cited by 138 · see citing works
Gordon, A. M., & Chen, S. (2016) · Journal of Personality and Social Psychology · doi:10.1037/pspi0000039 · Cited by 137 · see citing works
Racism, Gun Ownership and Gun Control: Biased Attitudes in US Whites May Influence Policy Decisions
O’Brien, K., Forrest, W., Lynott, D., et al. (2013) · PLoS ONE · doi:10.1371/journal.pone.0077552 · Cited by 131 · see citing works
Santoro, E., & Broockman, D. E. (2022) · Science Advances · doi:10.1126/sciadv.abn5515 · Cited by 126 · see citing works
Personal and Public Levels of Political Incivility
Muddiman, A. (2017) · International journal of communication · Cited by 123 · see citing works
How to foster perceived partner responsiveness: High‐quality listening is key
Itzchakov, G., Reis, H. T., & Weinstein, N. (2021) · Social and Personality Psychology Compass · doi:10.1111/spc3.12648 · Cited by 120 · see citing works
Scudder, M. F. (2020) · doi:10.1093/oso/9780197535455.001.0001 · Cited by 108 · see citing works
Shah, D. V. (2016) · Communication and the Public · doi:10.1177/2057047316628310 · Cited by 105 · see citing works
What Goes Without Saying: Navigating Political Discussion in America
Carlson, T. N., & Settle, J. E. (2022) · doi:10.1017/9781108912495 · Cited by 99 · see citing works
Pollmann, M. M. H., & Finkenauer, C. (2009) · Personality and Social Psychology Bulletin · doi:10.1177/0146167209342754 · Cited by 95 · see citing works
Sydnor, E. (2019) · Columbia University Press eBooks · doi:10.7312/sydn18924 · Cited by 90 · see citing works
Can I tell you how I feel? Perceived partner responsiveness encourages emotional expression.
Ruan, Y., Reis, H. T., Clark, M. S., Hirsch, J. L., & Bink, B. D. (2019) · Emotion · doi:10.1037/emo0000650 · Cited by 88 · see citing works
Receptiveness to Opposing Views: Conceptualization and Integrative Review
Minson, J. A., & Chen, F. S. (2021) · Personality and Social Psychology Review · doi:10.1177/10888683211061037 · Cited by 86 · see citing works
Kriplean, T., Toomim, M., Morgan, J. T., Borning, A. H., & Ko, A. J. (2012) · CHI Conference on Human Factors in Computing Systems · doi:10.1145/2207676.2208621 · Cited by 84 · see citing works
Gun Ownership as a Social Identity: Estimating Behavioral and Attitudinal Relationships
Lacombe, M. J., Howat, A., & Rothschild, J. E. (2019) · Social Science Quarterly · doi:10.1111/ssqu.12710 · Cited by 83 · see citing works
Quantifying 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 · see citing works
Does Talking to the Other Side Reduce Inter-party Hostility? Evidence from Three Studies
Amsalem, E., Merkley, E., & Loewen, P. J. (2021) · Political Communication · doi:10.1080/10584609.2021.1955056 · Cited by 77 · see citing works
“They just don't understand us”: The role of felt understanding in intergroup relations.
Livingstone, A., Rodríguez, L. F., & Rothers, A. (2019) · Journal of Personality and Social Psychology · doi:10.1037/pspi0000221 · Cited by 66 · see citing works
Toward Conflict or Compromise? How Violent Metaphors Polarize Partisan Issue Attitudes
Kalmoe, N. P., Gubler, J. R., & Wood, D. A. (2017) · Political Communication · doi:10.1080/10584609.2017.1341965 · Cited by 65 · see citing works
Self-citationAggarwal, M., Allen, J. N. L., Coppock, A., Frankowski, D., Messing, S., Zhang, K., Barnes, J., Beasley, A., Hantman, H., & Zheng, S. (2023) · Nature Human Behaviour · doi:10.1038/s41562-022-01487-4 · Cited by 63 · see citing works
The politeness Package: Detecting Politeness in Natural Language
Yeomans, M., Kantor, A., & Tingley, D. H. (2019) · The R Journal · doi:10.32614/rj-2018-079 · Cited by 57 · see citing works
Firepower: How the NRA Turned Gun Owners into a Political Force
Lacombe, M. J. (2021) · Project Muse (Johns Hopkins University) · Cited by 53 · see citing works
Lee, A. H. (2020) · Political Communication · doi:10.1080/10584609.2020.1799124 · Cited by 49 · see citing works
Filindra, A., Kaplan, N. J., & Büyüker, B. (2020) · Sociological Inquiry · doi:10.1111/soin.12388 · Cited by 48 · see citing works
Sun, Q., Wojcieszak, M., & Davidson, S. R. (2021) · Frontiers in Political Science · doi:10.3389/fpos.2021.741605 · Cited by 47 · see citing works
Incivility and Standing Firm: A Second Layer of Partisan Division
Wolf, M. R., Strachan, J. C., & Shea, D. M. (2012) · PS Political Science & Politics · doi:10.1017/s1049096512000509 · Cited by 44 · see citing works
Kim, H., Kim, H., Jo, K. J., & Kim, J. (2021) · Proceedings of the ACM on Human-Computer Interaction · doi:10.1145/3449140 · Cited by 36 · see citing works
Knobloch, K. R., & Gastil, J. (2021) · Journal of Deliberative Democracy · doi:10.16997/jdd.942 · Cited by 34 · see citing works
Miller, P. R., & Conover, P. J. (2015) · Politics Groups and Identities · doi:10.1080/21565503.2014.992795 · Cited by 15 · see citing works
Flagged first, then by citations
Research works printed but not linked: 4
- [23]T. N. Carlson, J. E. Settle, What Goes Without Saying (Cambridge University Press, 2022).
- [32]M. S. Levendusky, D. A. Stecula, We Need to Talk (Cambridge University Press, 2022).
- [54]S. Kim, J. Eun, J. Serring, J. Lee, Moderator chatbot for deliberative discussion: Effects of discussion structure and discussant facilitation. Proc. ACM Hum.-Comput. Interact. 5, 1– 26 (2021).36644216
- [56]J. Fishkin, “ Deliberative democracy with the online deliberation platform” in The 7th AAAI Conference on Human Computation and Crowdsourcing (HCOMP 2019) (2019).
Cited by
Sorted by citations
Can Generative AI improve social science?
Bail, C. A. (2024) · Proceedings of the National Academy of Sciences · doi:10.1073/pnas.2314021121 · Cited by 369 · see citing works
Self-citationAI can help people feel heard, but an AI label diminishes this impact
Yin, Y., Jia, N., & Wakslak, C. J. (2024) · Proceedings of the National Academy of Sciences · doi:10.1073/pnas.2319112121 · Cited by 247 · see citing works
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
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
Self-citationExamining Human-AI Collaboration for Co-Writing Constructive Comments Online
Shahid, F., Dittgen, M., Naaman, M., & Vashistha, A. (2025) · Proceedings of the ACM on Human-Computer Interaction · doi:10.1145/3757591 · Cited by 3 · see citing works
- Authors
- Argyle, L. P., Bail, C. A., Busby, E. C., Gubler, J. R., Howe, T., Rytting, C., Sorensen, T., & Wingate, D.
- Affiliations
- Brigham Young University (US)
- Duke University (US)
- University of Washington (US)
- Corresponding author
- Argyle, Lisa P.
- Bail, Christopher A.
- ORCID
- Venue
- Proceedings of the National Academy of Sciences
- Publisher
- National Academy of Sciences
- Published
- 2023-10-03
- Volume
- 120
- Issue
- 41
- Pages
- e2311627120-e2311627120
- ISSN
- 0027-8424 (ISSN-L)
- 1091-6490
- PMC
- PMC10576030
- OpenAlex
- W4387299971
- Cited by
- 144
- References
- 66
- Citations by year
- 2026: 47 · 2025: 61 · 2024: 33 · 2023: 3
- FWCI
- 68.00
- 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)
- Funders
- National Science Foundation (2141680)
- Brigham Young University (n/a)
- Duke University (n/a)
- Topics
- Social Sciences › Communication › Social Media and Politics
- Social Sciences › Sociology and Political Science › Misinformation and Its Impacts
- Social Sciences › Safety Research › Ethics and Social Impacts of AI
- Keywords
- online political discourse, large language models, feeling understood, conversation quality, policy attitudes, evidence-based recommendations
- MeSH terms
- Humans, Language, Policy
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
