The DOI matches its Crossref record; 28 of the 28 references it cites were found in a registry or an index; 1 of the 3 sources its claims cite were checked for support.
How persuasive is AI-generated propaganda?
Open accessAbstract
Can large language models, a form of artificial intelligence (AI), generate persuasive propaganda? We conducted a preregistered survey experiment of US respondents to investigate the persuasiveness of news articles written by foreign propagandists compared to content generated by GPT-3 davinci (a large language model). We found that GPT-3 can create highly persuasive text as measured by participants’ agreement with propaganda theses. We further investigated whether a person fluent in English could improve propaganda persuasiveness. Editing the prompt fed to GPT-3 and/or curating GPT-3’s output made GPT-3 even more persuasive, and, under certain conditions, as persuasive as the original propaganda. Our findings suggest that propagandists could use AI to create convincing content with limited effort.
Overview
CoChat Summary
TL;DR
A preregistered survey experiment with 8,221 US respondents found that GPT-3 davinci could generate propaganda articles that were highly persuasive, though slightly less persuasive than real-world foreign covert propaganda on average. Human editing of prompts and curation of outputs made GPT-3-generated propaganda as persuasive as, or more persuasive than, the original propaganda.
Key findings
- Original propaganda was highly persuasive: agreement with thesis statements rose from 24.4% in the control to 47.4% among those who read original propaganda, a 23 percentage point increase.
- GPT-3-generated propaganda was also highly persuasive: agreement was 43.5% versus 24.4% in the control, a 19.1 percentage point increase, but 3.9 percentage points lower than original propaganda on average.
- Human curation helped: after removing two of 18 GPT-3 outputs that did not advance the intended thesis, agreement rose to 45.6%, and the difference from original propaganda was no longer statistically significant.
- Editing prompts and selecting the best of three outputs produced even stronger effects: agreement was 52.7% compared with 47.4% for original propaganda; edited-prompt-only articles reached 46.4%, not statistically different from 47.4%.
Methods
- Preregistered survey experiment in December 2021 with US adults recruited via Lucid using quota sampling; final sample was 8,221 after attention-check failures and completion times under 3 minutes were excluded.
- Six English-language foreign covert propaganda articles (151–308 words) from real-world campaigns originating from Iran or Russia were compared with GPT-3 davinci outputs on the same topics; GPT-3 generated three articles per topic, with outputs under 686 or over 1,936 characters discarded.
- Respondents rated agreement with thesis statements for four randomly selected topics without reading an article (control), then read articles on the remaining two topics; analyses regressed percent agreement and scaled agreement on issue and article indicators.
Limitations
- The paper states its estimates may represent a lower bound because large language models are rapidly improving and because the experiment tested exposure to a single article, whereas propagandists could expose citizens to many articles.
- The paper states future research could test AI-generated propaganda across a wider range of issues and with information from multiple sources.
- Evident from the text: the study uses US respondents, six topics, and GPT-3 davinci in December 2021, so findings may not generalize to other populations, topics, models, or later AI systems.
Model-written summary. Check it against the paper before relying on it.
Integrity signals
Registry
Cites
Statistics
Figures



Verification
Claims the paper makes
1 Valid · 1 Not valid · 2 Not checked · 29 Own results and interpretations
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“Two of the GPT-3 propaganda articles (out of 18 total) did not advance the intended claim.”
§ How persuasive is AI-generated propaganda? > Results > Human–machine teaming
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Rests on an assumption not shown here.
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“If a propagandist edited the input and selected the best of the three outputs on each topic, the GPT-3-generated propaganda would be even more persuasive than the original propaganda (52.7% compared to 47.4%).”
§ How persuasive is AI-generated propaganda? > Results > Human–machine teaming
hedgedstates a cause- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Rests on 2 assumptions not shown here.
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“For most issues, each GPT-3-generated article was about as persuasive as the original propaganda.”
§ How persuasive is AI-generated propaganda? > Results > Persuasiveness of GPT-3-generated propaganda
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Rests on 2 assumptions not shown here.
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“Figures S7 and S8 show that the persuasive effects of the original propaganda and GPT-3 propaganda were fairly consistent across social groups. We did not find substantial heterogeneity in treatment effects when we split the sample by demographic variables, partisanship/ideology, news consumption, time spent on social media, and more.”
§ How persuasive is AI-generated propaganda? > Results > Persuasiveness of GPT-3-generated propaganda
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Rests on 2 assumptions not shown here.
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“The poor performance of these articles and a few others caused GPT-3 to perform slightly less well than the original propaganda on average, when, in Fig. 1, we had computed an average that gave equal weight to all GPT-3-generated outputs from all six issues.”
§ How persuasive is AI-generated propaganda? > Results > Persuasiveness of GPT-3-generated propaganda
plainstates a cause- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Rests on 2 assumptions not shown here.
11 more claims
Create a free account to see all 33 claims and every source verdict.
- It may state two separate claims as one. 11
- Probably not a substantial claim. 12
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 source backs part of the claim, not all of it — or backs it for a different population, condition or comparison.
In the citing paper
“But users increasingly cannot tell the difference between human writing and LLM writing when presented with text (30–32), and between a human interlocutor or a contemporary chatbot when in interaction (17, 26).”
From The benefits and dangers of anthropomorphic conversational agents (2025) · cites this paper as [32]
Experimental — accuracy not yet measuredThe cited source says what the claim attributes to it.
In the citing paper
“as well as persuasive propaganda (20).”
From The impact of generative artificial intelligence on socioeconomic inequalities and policy making (2024) · cites this paper as [20]
The check wasn’t confident either way, so it gives no verdict — never disputed.
In the citing paper
“These mechanisms illustrate how affect-driven content can override cognitive vigilance, aligning with recent work on emotional misinformation (Goldstein et al., 2024; Perach et al., 2023).”
From The illusion of reality: How AI-generated images (AIGIs) are fooling social media users (2026) · cites this paper as Goldstein et al., 2024
Experimental — accuracy not yet measuredThe check wasn’t confident either way, so it gives no verdict — never disputed.
In the citing paper
“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 [74]
Experimental — accuracy not yet measured
Citations
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
On the Opportunities and Risks of Foundation Models
Bommasani, R., Hudson, D. A., Adeli, E., Altman, R., Arora, S., von Arx, S., Bernstein, M. S., Bohg, J., Bosselut, A., Brunskill, E., et al. (2021) · arXiv · doi:10.48550/arxiv.2108.07258 · Cited by 2,676
Self-citationPrior exposure increases perceived accuracy of fake news
Pennycook, G., Cannon, T. D., & Rand, D. G. (2018) · Journal of Experimental Psychology General · doi:10.1037/xge0000465 · Cited by 1,268 · cited 1×
“SEs and 95% CIs are shown. exposed multiple times (18), that repeated exposure to content on a social media platform leads to a higher rate of sharing that content (19, 20), and that belief in false information continues to grow loga-rithmically with additional exposures (21).”
BackgroundFrequency and the conference of referential validity
Hasher, L., Goldstein, D., & Toppino, T. (1977) · Journal of Verbal Learning and Verbal Behavior · doi:10.1016/s0022-5371(77)80012-1 · Cited by 920
Taxonomy 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×
“Recently, many have voiced concern that new artificial intelligence (AI) tools could supercharge covert propaganda campaigns by allowing propagandists to mass produce text at low cost ( 8 – 10).”
BackgroundHolistic Evaluation of Language Models
Bommasani, R., Liang, P., & Lee, T. (2023) · Annals of the New York Academy of Sciences · doi:10.1111/nyas.15007 · Cited by 480
Köbis, N., & Mossink, L. D. (2021) · Computers in Human Behavior · doi:10.1016/j.chb.2020.106553 · Cited by 406
All 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×
“Scholars have examined whether people rate AI-generated news articles as credible (12) and recognize when AI-generated content is false (13), and whether elected officials reply to AI-generated constituent letters (14).”
BackgroundMisinformation During a Pandemic
Bursztyn, L., Rao, A., Roth, C., & Yanagizawa-Drott, D. (2020) · National Bureau of Economic Research · doi:10.3386/w27417 · Cited by 265
Holistic Evaluation of Language Models
Liang, P., Bommasani, R., Lee, T., Tsipras, D., Soylu, D., Yasunaga, M., Zhang, Y., Narayanan, D., Wu, Y., Kumar, A., et al. (2022) · arXiv · doi:10.48550/arxiv.2211.09110 · Cited by 255
Cited by
The impact of generative artificial intelligence on socioeconomic inequalities and policy making
Capraro, V., Lentsch, A., Acemoglu, D., Akgun, S., Akhmedova, A., Bilancini, E., Bonnefon, J.-F., Brañas-Garza, P., Butera, L., Douglas, K. M., et al. (2024) · PNAS Nexus · doi:10.1093/pnasnexus/pgae191 · Cited by 371
On the Conversational Persuasiveness of Large Language Models: A Randomized Controlled Trial
Salvi, F., Ribeiro, M. H., Gallotti, R., & West, R. (2024) · doi:10.21203/rs.3.rs-4429707/v1 · Cited by 285
On the conversational persuasiveness of GPT-4
Salvi, F., Horta Ribeiro, M., Gallotti, R., & West, R. (2025) · Nature Human Behaviour · doi:10.1038/s41562-025-02194-6 · Cited by 264
“Our study suggests that concerns around personalization and AI persuasion are warranted, reinforcing previous results 42,44,45,48 by show-casing how LLMs can outpersuade humans in online conversations through microtargeting.”
Evaluating the persuasive influence of political microtargeting with large language models
Hackenburg, K., & Margetts, H. (2024) · Proceedings of the National Academy of Sciences · doi:10.1073/pnas.2403116121 · Cited by 114
LLM-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
The benefits and dangers of anthropomorphic conversational agents
Peter, S., Riemer, K., & West, J. D. (2025) · Proceedings of the National Academy of Sciences · doi:10.1073/pnas.2415898122 · Cited by 110
Fact-checking information from large language models can decrease headline discernment
DeVerna, M., Yan, H. Y., Yang, K.-C., & Menczer, F. (2024) · Proceedings of the National Academy of Sciences · doi:10.1073/pnas.2322823121 · Cited by 145
AI language model rivals expert ethicist in perceived moral expertise
Dillion, D., Mondal, D., Tandon, N., & Gray, K. J. (2025) · Scientific Reports · doi:10.1038/s41598-025-86510-0 · Cited by 107
Quinn, M., & Gutt, D. (2025) · Journal of Management Information Systems · doi:10.1080/07421222.2025.2487313 · Cited by 61
“For example, GenAI can be used to compose propaganda texts that are as convincing as human-composed texts [28].”
Danry, 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
“LLMs like ChatGPT have demonstrated remarkable capabilities in generating human-like text, raising concerns about their potential for spreading misinformation at scale [6, 36, 37] and changing people’s attitudes [47, 53, 92].”
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 183
Davidson, T., & Karell, D. (2025) · Sociological Methods & Research · doi:10.1177/00491241251339184 · Cited by 56
Comparing the Willingness to Share for Human-generated vs. AI-generated Fake News
Bashardoust, A., Feuerriegel, S., & Shrestha, Y. R. (2024) · Proceedings of the ACM on Human-Computer Interaction · doi:10.1145/3687028 · Cited by 73
Deception and manipulation in generative AI
Tarsney, C. J. (2025) · Philosophical Studies · doi:10.1007/s11098-024-02259-8 · Cited by 62
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 97
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 53
How spammers and scammers leverage AI-generated images on Facebook for audience growth
DiResta, R., & Goldstein, J. A. (2024) · Harvard Kennedy School Misinformation Review · doi:10.37016/mr-2020-151 · Cited by 35
Self-citationHow malicious AI swarms can threaten democracy
Schroeder, D. T., Cha, M., Baronchelli, A., Bostrom, N., Christakis, N. A., Garcia, D., Goldenberg, A., Kyrychenko, Y., Leyton-Brown, K., Lutz, N., et al. (2026) · Science · doi:10.1126/science.adz1697 · Cited by 53
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 66
“Experimental research has detailed the potential for generative-AI to aid in the production of credible text (8, 9) and in the persuasion of survey respondents (5, 28).”
Kozlowski, A. C., & Evans, J. A. (2025) · Sociological Methods & Research · doi:10.1177/00491241251337316 · Cited by 70
The impact of generative artificial intelligence on socioeconomic inequalities and policy makingPreprint
Capraro, V., Lentsch, A., Acemoğlu, D., Akgün, S., Akhmedova, A., Bilancini, E., Bonnefon, J., Brañas‐Garza, P., Butera, L., Douglas, K. M., et al. (2023) · PsyArXiv (OSF Preprints) · doi:10.31234/osf.io/6fd2y · Cited by 48
Nygren, T., Samuelsson, M., Hansson, P.-O., Efimova, E., & Bachelder, S. (2025) · International Journal of Artificial Intelligence in Education · doi:10.1007/s40593-025-00484-8 · Cited by 29
“The risks of persuasion also need to be noted since generative AI can be used for propaganda and manipulation for evil purposes (Goldstein et al., 2024).”
Biased 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 42
Decoding persuasion: a survey on ML and NLP methods for the study of online persuasion
Bassi, D., Fomsgaard, S. K., & Pereira-Fariña, M. (2024) · Frontiers in Communication · doi:10.3389/fcomm.2024.1457433 · Cited by 23
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 26
“…concerning 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.…”
GenAI and misinformation in education: a systematic scoping review of opportunities and challenges
Fulsher, A., Pagkratidou, M., & Kendeou, P. (2025) · AI & Society · doi:10.1007/s00146-025-02536-y · Cited by 38
Łabuz, M., & Nehring, C. (2024) · Politics & Policy · doi:10.1111/polp.12617 · Cited by 29
Yan, H. Y., Morrow, G., Yang, K.-C., & Wihbey, J. (2025) · Harvard Kennedy School Misinformation Review · doi:10.37016/mr-2020-171 · Cited by 50
“Studies have shown that personalized and targeted political messages produced by advanced generative AI tools can be highly persuasive (Bai et al., 2023; Costello et al., 2024; Goldstein et al., 2024; Karinshak et al., 2023), sometimes more so than human-generated messages (Huang & Wang, 2023).”
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 56
“Similarly, Goldstein et al. [32] found that GPT-3 could write persuasive text with limited effort.”
Farrell, H. J. (2025) · Annual Review of Political Science · doi:10.1146/annurev-polisci-040723-013245 · Cited by 30
Pataranutaporn, P., Archiwaranguprok, C., Chan, S., Loftus, E. F., & Maes, P. (2025) · CHI Conference on Human Factors in Computing Systems · doi:10.1145/3706598.3713697 · Cited by 49
“In two conditions, the static images were converted into 5-second videos using a generative AI tool.”
When cutting edge meets silver tongue: Understanding the word-of-machine effect on travel decisions
Sun, D., Wong, I. A., Xiong, X., & Li, S. (2025) · Tourism Management · doi:10.1016/j.tourman.2025.105271 · Cited by 23
The 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 84
A meta-analysis of the persuasive power of large language models
Hölbling, L., Maier, S., & Feuerriegel, S. (2025) · Scientific Reports · doi:10.1038/s41598-025-30783-y · Cited by 46
“Across the reviewed studies, human-generated messages were typically more emotionally vivid and personally engaging, whereas LLM-generated texts relied more on analytical reasoning and informational coherence [1, 2, 17, 22, 31].”
The Impact of Generative Artificial Intelligence on Socioeconomic Inequalities and Policy MakingPreprint
Capraro, V., Lentsch, A., Acemoğlu, D., Akgün, S., Akhmedova, A., Bilancini, E., Bonnefon, J., Brañas‐Garza, P., Butera, L., Douglas, K. M., et al. (2024) · SSRN Electronic Journal · doi:10.2139/ssrn.4666103 · Cited by 18
The Psychology of Misinformation Across the Lifespan
Edelson, S. M., Reyna, V. F., Singh, A., & Roue, J. E. (2024) · Annual Review of Developmental Psychology · doi:10.1146/annurev-devpsych-010923-093547 · Cited by 21
Generative AI, Work and Risks in Cultural and Creative IndustriesPreprint
Walkowiak, E., & Potts, J. (2024) · SSRN Electronic Journal · doi:10.2139/ssrn.4830265 · Cited by 21
Azgın, B., & Kıralp, Ş. (2024) · Social Sciences · doi:10.3390/socsci13100510 · Cited by 20
Dillion, D., Mondal, D., Tandon, N., & Gray, K. J. (2024) · PsyArXiv (OSF Preprints) · doi:10.31234/osf.io/w7236 · Cited by 28
Ganaie, N. A. (2026) · Frontiers in Political Science · doi:10.3389/fpos.2025.1718396 · Cited by 23
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 56
Clicks and tricks: The dark art of online persuasion
Fagan, P. (2024) · Current Opinion in Psychology · doi:10.1016/j.copsyc.2024.101844 · Cited by 20
“…across three experiments and a sample of 4836 participants, messages created by ChatGPT were persuasive across a range of political issues including gun rights and carbon taxes e in fact, AI-generated messages were as persuasive as those written by humans [28], a finding echoed elsewhere [29].”
BackgroundAI-driven report-generation tools in mental healthcare: A review of commercial tools
Bouguettaya, A., Victoria Team, Stuart, E. M., & Aboujaoude, E. N. (2025) · General Hospital Psychiatry · doi:10.1016/j.genhosppsych.2025.02.018 · Cited by 13
A Lifecycle-Oriented Survey of Emerging Threats and Vulnerabilities in Large Language Models
de Maio, C., Di Gisi, M., Fenza, G., Gallo, M., & Loia, V. (2025) · IEEE Access · doi:10.1109/access.2025.3619764 · Cited by 10
“Authors in [34] tested the persuasive-ness of GPT-3-generated propaganda compared to real foreign disinformation campaigns (from Russia and Iran) in a survey of over 8 , 000 U.S. adults.”
Dierickx, L., & Lindén, C. (2024) · Media and Communication · doi:10.17645/mac.8668 · Cited by 17
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 24
Local AI Governance: Addressing Model Safety and Policy Challenges Posed by Decentralized AI
Sokhansanj, B. A. (2025) · AI · doi:10.3390/ai6070159 · Cited by 16
Karell, D., Sachs, J. D., & Barrett, R. (2024) · Poetics · doi:10.1016/j.poetic.2024.101966 · Cited by 14
Hu, B., Sheng, Q., Cao, J., Li, Y., & Wang, D. (2025) · International ACM SIGIR Conference on Research and Development in Information Retrieval (SIGIR) · doi:10.1145/3726302.3730027 · Cited by 33
“Then we provide further analysis about the occurrence of truth decay from a familiarity perspective.”
Zhao, S., Zhou, J., & Wang, T. (2025) · Journalism · doi:10.1177/14648849251381218 · Cited by 19
“This technological evolution has progressed from handling structured, routine tasks to generating sophisticated content that increasingly matches human quality and style (Clerwall, 2014; Goldstein et al., 2024; Spitale et al., 2023).”
Genç, H. U., Chandrasegaran, S. K., Dingler, T., & Verma, H. (2025) · CHI Conference on Human Factors in Computing Systems · doi:10.1145/3706598.3713579 · Cited by 15
“Studies have observed a rise in AI-created disinformation eforts [50, 72] and the elements that hinder people’s capacity to diferentiate between accurate and misleading information [49, 81].”
Nezhad, M. M., Kisomi, M. A., & Gholinezhad, F. (2025) · International Conference on Web Research (ICWR) · doi:10.1109/icwr65219.2025.11006192 · Cited by 7
“In contrast, other research suggests that their performance is comparable to human-level persuasion [16-18].”
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 8
“Some survey responses suggested adding more GUI elements to streamline the responses, such as a button to indicate high interest in a particular topic.”
Must Read: A Comprehensive Survey of Computational Persuasion
Bozdag, N. B., Mehri, S., Yang, X., Ha, H., Cheng, Z., Durmus, E., You, J., Ji, H., Tür, G., & Hakkani-Tur, D. Z. (2026) · ACM Computing Surveys · doi:10.1145/3800687 · Cited by 49
“Focusing more specifically on political communication, Goldstein et al. [64] found that AI-generated propaganda is already as persuasive as real-world examples.”
Public trust in AI: A dynamic social media view
Thaiduong, N. (2026) · Social Sciences & Humanities Open · doi:10.1016/j.ssaho.2026.102457 · Cited by 8
Looking Under the Hood: How LLMs Attempt Political Persuasion and Microtargeting
Lyman, A., Busby, E. C., Argyle, L. P., Gubler, J. R., Hepner, B., & Wingate, D. (2026) · Chinese Political Science Review · doi:10.1007/s41111-026-00336-3 · Cited by 6
State media control influences large language models
Waight, H. C., 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 30
Krak, I. V., Дідур, В., Molchanova, M. O., Mazurets, O. V., Собко, О., ЗАЛУЦЬКА, О., & Barmak, O. (2024) · International Scientific and Practical Programming Conference (UkrPROG) · doi:10.55056/ceur-ws.org/vol-3806/s_36_krak.pdf · Cited by 5
“…to the intellectualization of propaganda detection processes, which allows avoiding a number of technological problems in monitoring media sources [5], and the problem of separating manifestations of propaganda techniques from other manipulative influences [6, 7] are especially relevant at the…”
A Detailed Analysis of Article 50 of the EU's Artificial Intelligence ActPreprint
Gils, T. (2024) · SSRN Electronic Journal · doi:10.2139/ssrn.4865427 · Cited by 11
The illusion of reality: How AI-generated images (AIGIs) are fooling social media users
Miskolczi, M. (2026) · Computers in Human Behavior · doi:10.1016/j.chb.2025.108876 · Cited by 5
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
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 16
Generative AI Disproportionately Harms Long Tail Users
Maung, B. M., McBride, K., Lucas, J. S., Tabar, M., & Lee, D. (2024) · Computer · doi:10.1109/mc.2024.3408594 · Cited by 5
“With the democratization of GenAI, disinformation powered by persuasive and personalized machine-generated [5] texts and synthetic media can erode this trust.”
BackgroundThe Role of Generative AI Use in 2024 Elections Worldwide
(IPIE), I. P. O. T. I. E., Trauthig, I. K., Valenzuela, S., & Howard, P. N. (2025) · doi:10.61452/hzue9853 · Cited by 21
Aimen, T. (2025) · IFIP advances in information and communication technology · doi:10.1007/978-3-031-91054-8_3 · Cited by 3
The Risks of AI-Generated, Hyper-Personalized Digital Advertisements
LeBrun, A. (2025) · Philosophy & Technology · doi:10.1007/s13347-025-00935-z · Cited by 9
“(Indeed, there is some recent evidence that AI-generated political information itself is more persuasive than human generated (Goldstein et al., 2024; Karinshak et al., 2023; Bai et al., 2023).”
The persuasive potential of AI-paraphrased information at scale
Dash, S., Xu, Y.-W., Jalbert, M., & Spiro, E. S. (2025) · PNAS Nexus · doi:10.1093/pnasnexus/pgaf207 · Cited by 8
Herckis, L., & Tse, E. (2025) · Journal of Nursing Regulation · doi:10.1016/j.jnr.2025.08.008 · Cited by 4
Social media, AI, and the rise of extremism during intergroup conflict
Lavie-Driver, N. I., & van der Linden, S. (2025) · Frontiers in Social Psychology · doi:10.3389/frsps.2025.1711791 · Cited by 10
“Lastly, LLMs can now create persuasive propaganda (Wack et al., 2025a), often as or more persuasive than human-written propaganda (Goldstein et al., 2024) which can then be micro-targeted at users.”
Żywiołek, J. (2025) · Management Systems in Production Engineering · doi:10.2478/mspe-2025-0057 · Cited by 3
Jones, C., & Bergen, B. K. (2026) · Artificial Intelligence Review · doi:10.1007/s10462-026-11517-6 · Cited by 56
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 4
“Additionally, the variance in the USA metric is high and increases with more provided shots, underscoring the challenges of personality simulation, as highlighted in (Wang et al., 2024).”
Epstein, R., Newland, A., & Tang, L. (2025) · PLoS ONE · doi:10.1371/journal.pone.0322900 · Cited by 6
Stalnaker, T., Wintersgill, N., Chaparro, O., Heymann, L. A., Tufo, M., Germán, D. M., & Poshyvanyk, D. (2025) · SSRN Electronic Journal · doi:10.2139/ssrn.5353252 · Cited by 4
Digital Literacy Interventions Can Boost Humans in Discerning DeepfakesPreprint
Geissler, D., Robertson, C. E., & Feuerriegel, S. (2025) · SSRN Electronic Journal · doi:10.2139/ssrn.5368742 · Cited by 5
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 9
Dancing in uniform: Thirst trap propaganda in military image wars on TikTok
Bösch, M., & Divon, T. (2026) · Platforms & Society · doi:10.1177/29768624251403864 · Cited by 2
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 7
Intelligent Systems, Vulnerable Minds: A Framework for Radicalization to Violence in the Age of AI
Kunst, J. R., Obaidi, M., Gollwitzer, A., Brandtzæg, P. B., Hinrichs, Y., Saini, N., & Schroeder, D. T. (2026) · Personality and Social Psychology Review · doi:10.1177/10888683261430089 · Cited by 14
“…can use LLMs to generate convincing articles, social media posts, and AI-generated deepfake videos, such as a government official appearing to admit to a conspiracy or a religious leader endorsing violence, to validate a worldview (Goldstein et al., 2024; Obaidi, Kunst, et al., 2021; Verma, 2024).”
Poredi, N., Sudarsan, M., Solomon, E., Nagothu, D., & Chen, Y. (2024) · Disruptive Technologies in Information Sciences · doi:10.1117/12.3013240 · Cited by 6
In Moderation: Automation in the Digital Public Sphere
Acosta-Navas, D. (2025) · Journal of Business Ethics · doi:10.1007/s10551-024-05912-8 · Cited by 9
“…on social media platforms may range from misleading posts by individual users to massive campaigns that employ frontier technologies to conduct coordinated influence operations (Bradshaw et al., 2021; DiResta, 2020, 2021; Goldstein et al., 2024; Mozilla Foundation, 2022; Starbird et al., 2023).”
Ho, T. M., & Hoang, T.-D. (2025) · AI & Society · doi:10.1007/s00146-025-02244-7 · Cited by 2
Instrumental Bot Identities and the Politics of Online Discourse
Letourneau, A. M. (2025) · Social Media + Society · doi:10.1177/20563051251378233 · Cited by 5
“Recent developments have allowed large language models to generate text that is almost as persuasive to US audiences as genuine foreign covert propaganda (Goldstein et al., 2024), a reality that Western democracies are ill-prepared to deal with.”
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 · Cited by 1
Interactive perceptions and persuasion: how AI-generated content influences users
Sun, Y., Fan, Z., Sheng, D., Zhang, X., & Xu, F. (2026) · The Electronic Library · doi:10.1108/el-05-2025-0172 · Cited by 1
Sosyal Medyada Politik Aktör Olarak Yapay Zekâ: Aktör-Ağ Teorisi Bağlamında Grok Yapay Zekâ Asistanı
Bulur, N. (2025) · Galatasaray Üniversitesi İleti-ş-im Dergisi · doi:10.16878/gsuilet.1756445 · Cited by 3
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 2
Framing AI in public innovation: the engagement-manipulation nexus
Adebajo, A., Okpara, N., & Kolawole, A. (2026) · Public Management Review · doi:10.1080/14719037.2026.2659262 · Cited by 2
Badade, A. B., & Dhanaraj, R. K. (2026) · International Conference on Emerging Smart Computing and Informatics (ESCI) · doi:10.1109/esci68015.2026.11493366 · Cited by 1
“…coming onto the scene has made the detection of astroturfing even harder, for the reason that huge language models are capable of creating disinformation that is very fitting to the context and emotionally touching with no human intervention at all (Goldstein et al., 2024; Yang & Menczer, 2024).”
Li, J., Wu, Y., Li, H., & Xu, L. (2026) · International Journal of Human-Computer Interaction · doi:10.1080/10447318.2026.2674290 · Cited by 1
Geise, S., & Xu, Y. (2026) · Digital Journalism · doi:10.1080/21670811.2026.2700594 · Cited by 2
“Its ability to produce realistic synthetic images blurs the line between reality and fiction, raising concerns about disinformation and propaganda (Geise 2025; Goldstein et al. 2024).”
Peykani, P., Charles, V., Emrouznejad, A., Ghanidel, S., & Sabour, D. (2026) · Archives of Computational Methods in Engineering · doi:10.1007/s11831-026-10720-6 · Cited by 1
Punishment Trumps Warning in Prebunking Misinformation: Evidence From Survey Experiments in Taiwan
Sheen, G. C.-H. (2024) · Journal of Asian and African Studies · doi:10.1177/00219096241287710 · Cited by 1
“Scholars have recently found that the large language model can create highly persuasive propaganda messages (Goldstein et al., 2024).”
Heterogeneous Effects of Generative AI on Knowledge Seeking in Online CommunitiesPreprint
Quinn, M., & Gutt, D. (2023) · SSRN Electronic Journal · doi:10.2139/ssrn.4522386 · Cited by 1
Punishment trumps warning in prebunking misinformation: Evidence from survey experiments in TaiwanPreprint
Sheen, G. C.-H. (2024) · SSRN Electronic Journal · doi:10.2139/ssrn.4897558
Acceptable Use Policies for Foundation Models
Klyman, K. (2024) · SuperIntelligence - Robotics - Safety & Alignment · doi:10.70777/si.v1i1.10917
Propaganda, Democracy, and Emerging Technology
Martin, Z., Trauthig, I. K., & Woolley, S. (2025) · Springer Reference Sozialwissenschaften · doi:10.1007/978-3-658-08460-8_105-1
Hristoski, I. S. (2025) · IGI Global eBooks · doi:10.4018/979-8-3693-9601-8.ch012
HuPer: Human Factors Integrated Persuasive Dialogue Models
Kong, L., Jin, C., Song, R., Wang, X., & Chen, Y. (2025) · Lecture notes in computer science · doi:10.1007/978-981-96-8197-6_22
The Twin Disruption of Generative AI: Unbundling Capability and Responsibility in Creative WorkPreprint
Walkowiak, E., & Potts, J. (2025) · SSRN Electronic Journal · doi:10.2139/ssrn.5362872
Metadata
- Authors
- Goldstein, J. A., Chao, J., Grossman, S., Stamos, A., & Tomz, M.
- Affiliations
- Georgetown University (US)
- Stanford University (US)
- Stanford Institute for Economic Policy Research (US)
- Corresponding author
- Grossman, Shelby
- Venue
- PNAS Nexus
- Publisher
- Oxford University Press (OUP)
- Published
- 2024-02-01
- Volume
- 3
- Issue
- 2
- Pages
- pgae034-pgae034
- ISSN
- 2752-6542 (ISSN-L)
- PMC
- PMC10878360
- OpenAlex
- W4391985880
- Cited by
- 133
- References
- 18
- Citations by year
- 2026: 45 · 2025: 64 · 2024: 22
- FWCI
- 93.55
- Type
- journal article
- Language
- en
- Open access
- gold
- doi.org
- Licence
- Locations
- PNAS Nexus (publishedVersion, cc-by)
- PubMed Central (publishedVersion, cc-by)
- PubMed (publishedVersion)
- Funders
- Stanford University (#216301)
- Topics
- Social Sciences › Sociology and Political Science › Misinformation and Its Impacts
- Social Sciences › Sociology and Political Science › Media Influence and Politics
- Physical Sciences › Computer Science › Artificial Intelligence › Ethics and Social Impacts of AI
- Keywords
- large language models, GPT-3, persuasion, automated content generation, disinformation
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.
The claim verdicts this report counts are experimental: their accuracy is not yet measured.
A language model drew the claims from the paper on 2026-10-10, under version 3 of the claim definition (model deepseek-v4.1-flash, prompt extract-v3).
A second model labelled each claim on 2026-10-10 (model jev-1.13-20260917, definition revision gold-guideline-v0.2).
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
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.