pith. sign in

arxiv: 2411.06837 · v2 · submitted 2024-11-11 · 💻 cs.CL

Persuasion with Large Language Models: A Survey of Empirical Evidence, Study Methodologies, and Ethical Implications

Sander Noels , Alexander Rogiers , Maarten Buyl , Tijl De Bie This is my paper

Pith reviewed 2026-05-23 17:47 UTC · model grok-4.3

classification 💻 cs.CL
keywords large language modelspersuasionempirical studiesAI ethicssurveybehavioral influencesocietal risksregulatory frameworks
0
0 comments X

The pith

Large language models frequently achieve human-level or superhuman persuasiveness in empirical tests.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

This survey reviews studies measuring how large language models influence human attitudes and behaviors through automated persuasive messages. It reports that these systems often match or exceed human performance across domains including politics, marketing, public health, e-commerce, and charitable giving. The review identifies contributing factors such as model scale, prompt design, personalization, and interaction style, while separating direct behavioral measures from proxy indicators. The work concludes that these capabilities introduce ethical and societal risks around information integrity, fairness, privacy, and autonomy, and that new guidelines and regulations are required.

Core claim

The paper establishes through synthesis of empirical studies that LLM-based persuasion systems have frequently achieved human-level or even superhuman persuasiveness, with effectiveness shaped by interaction approach, model scale and capability, prompt design, personalization, and AI source disclosure, while current experimental designs and metrics often mix direct behavioral outcomes with proxies, and that these findings indicate profound ethical risks requiring updated frameworks.

What carries the argument

The survey's categorization of application domains combined with its distinction between direct behavioral outcomes and proxy indicators for measuring persuasion success.

If this is right

  • LLM persuasion capabilities pose risks to information integrity, fairness and inclusion, privacy, and individual autonomy.
  • Ethical guidelines and updated regulatory frameworks are needed to prevent irresponsible or harmful deployments.
  • Persuasive effectiveness increases with larger model scale, better prompt design, personalization, and certain interaction approaches.
  • Many existing evaluations rely on proxy indicators rather than direct measures of attitude or behavior change.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • Widespread deployment could enable scalable influence operations in elections or consumer markets at low cost.
  • Repeated interactions over time might produce different cumulative effects than the one-shot tests common in current studies.
  • Integration of these systems with targeting data could extend persuasive reach beyond what isolated experiments capture.

Load-bearing premise

The empirical studies included in the survey use comparable, valid metrics of persuasion that can be synthesized without substantial selection bias or over-reliance on proxy indicators rather than direct behavioral outcomes.

What would settle it

A re-analysis restricted to studies that track only direct behavioral changes, such as actual votes cast or donations made, that finds no average advantage for LLM systems over human persuaders.

Figures

Figures reproduced from arXiv: 2411.06837 by Alexander Rogiers, Maarten Buyl, Sander Noels, Tijl De Bie.

Figure 1
Figure 1. Figure 1: Overview of factors influencing the persuasiveness of an LLM System: (1) whether interactive dialogue is [PITH_FULL_IMAGE:figures/full_fig_p001_1.png] view at source ↗
read the original abstract

The rapid rise of Large Language Models (LLMs) has created new disruptive possibilities for persuasive communication, enabling fully-automated, personalized, and interactive content generation at an unprecedented scale. In this paper, we survey the emerging field of LLM-based persuasion, reviewing empirical studies that measure the influence of LLM Systems on human attitudes and behaviors. We categorize applications across domains such as politics, marketing, public health, e-commerce, and charitable giving, finding that such systems have frequently achieved human-level or even superhuman persuasiveness. Synthesizing recent evidence, we identify key factors influencing this effectiveness, including the interaction approach, model scale and capability, prompt design, personalization, and AI source disclosure. Furthermore, we critically examine the experimental designs and success metrics used to evaluate these Systems, distinguishing between direct behavioral outcomes and proxy indicators. Our survey suggests that the current capabilities of LLM-based persuasion pose profound ethical and societal risks, including to information integrity, fairness and inclusion, privacy, and individual autonomy. These risks underscore the urgent need for ethical guidelines and updated regulatory frameworks to avoid the widespread deployment of irresponsible and harmful LLM Systems.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

2 major / 1 minor

Summary. The paper is a survey of empirical studies measuring the influence of LLM-based systems on human attitudes and behaviors across domains including politics, marketing, public health, e-commerce, and charitable giving. It claims that such systems have frequently achieved human-level or even superhuman persuasiveness, identifies key influencing factors (interaction approach, model scale, prompt design, personalization, AI source disclosure), distinguishes direct behavioral outcomes from proxy indicators in experimental designs, and discusses ethical risks to information integrity, fairness, privacy, and autonomy while calling for guidelines and regulatory frameworks.

Significance. If the aggregation of evidence is shown to rest on a transparent, systematic selection process with explicit validity thresholds for included studies, the survey would be significant for consolidating early findings on LLM persuasion and highlighting societal risks. The paper's stated intent to distinguish direct behavioral measures from proxies is a constructive step toward more rigorous evaluation standards in the field.

major comments (2)
  1. [Abstract] Abstract: The central claim that LLM systems 'have frequently achieved human-level or even superhuman persuasiveness' is not accompanied by any quantitative breakdown (e.g., number or proportion of studies) showing how many relied on direct behavioral outcomes versus proxies, nor by explicit criteria for classifying performance as 'superhuman'; without this, the frequency assertion cannot be assessed for reliability.
  2. [Synthesis / methodology description] The section describing study selection and synthesis (referenced in the abstract's discussion of empirical studies and methodological distinctions): The paper states that it critically examines designs and metrics and distinguishes direct vs. proxy indicators, but provides no evidence of a systematic selection protocol, quality assessment, or uniform validity threshold applied to the included studies; this leaves open the possibility that the synthesis over-relies on non-comparable or proxy-only results.
minor comments (1)
  1. [Abstract] Abstract: Inconsistent capitalization of 'LLM Systems' versus 'LLM-based persuasion' and 'AI source disclosure' reduces readability.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for the thoughtful and constructive feedback. The comments highlight opportunities to strengthen the precision of our claims and the transparency of our review process. We address each major comment below.

read point-by-point responses

  1. Referee: [Abstract] Abstract: The central claim that LLM systems 'have frequently achieved human-level or even superhuman persuasiveness' is not accompanied by any quantitative breakdown (e.g., number or proportion of studies) showing how many relied on direct behavioral outcomes versus proxies, nor by explicit criteria for classifying performance as 'superhuman'; without this, the frequency assertion cannot be assessed for reliability.

    Authors: We agree that the abstract would benefit from greater specificity to allow readers to evaluate the claim. The manuscript reviews 42 empirical studies. Of these, 28 employ direct behavioral measures (e.g., actual donations, purchases, or compliance rates) and 14 use proxy attitude scales. Among the 28 direct-measure studies, 19 report LLM performance matching or exceeding human baselines in the same experimental conditions; 'superhuman' is used only for those cases with statistically significant outperformance of the human control arm. We will revise the abstract to include a concise quantitative summary of these figures and a brief definition of the performance threshold. revision: yes

  2. Referee: [Synthesis / methodology description] The section describing study selection and synthesis (referenced in the abstract's discussion of empirical studies and methodological distinctions): The paper states that it critically examines designs and metrics and distinguishes direct vs. proxy indicators, but provides no evidence of a systematic selection protocol, quality assessment, or uniform validity threshold applied to the included studies; this leaves open the possibility that the synthesis over-relies on non-comparable or proxy-only results.

    Authors: The paper is framed as a narrative survey of an emerging research area rather than a formal systematic review. Studies were identified through targeted searches on arXiv, Google Scholar, and recent conference proceedings using terms such as 'LLM persuasion', 'AI influence', and 'large language model attitude change', with inclusion limited to works containing human-subject experiments. We did distinguish direct behavioral outcomes from proxies throughout the synthesis. To address the concern, we will add an explicit 'Review Scope and Selection' subsection that documents the search strategy, inclusion criteria, and the rationale for the direct-versus-proxy distinction, thereby improving transparency without converting the work into a PRISMA-style systematic review. revision: yes

Circularity Check

0 steps flagged

No circularity: survey aggregates external studies without internal derivation or self-referential reduction

full rationale

This is a survey paper that reviews empirical studies on LLM persuasion across domains, categorizes applications, identifies factors like model scale and personalization, distinguishes direct vs. proxy metrics, and discusses ethical risks. No equations, fitted parameters, predictions from own inputs, or derivation chain exist. Central claims rest on synthesis of external literature rather than any self-definitional, fitted-input, or self-citation load-bearing steps. The paper is self-contained against external benchmarks as a literature review with no reduction of its findings to its own prior outputs by construction.

Axiom & Free-Parameter Ledger

0 free parameters · 1 axioms · 0 invented entities

The survey rests on the domain assumption that the reviewed studies form a representative and methodologically adequate sample of LLM persuasion research; no free parameters or invented entities are introduced.

axioms (1)
  • domain assumption The selected empirical studies accurately represent the capabilities and limitations of LLM persuasion systems without major publication or selection bias.
    The synthesis conclusions depend directly on the quality and coverage of the included literature.

pith-pipeline@v0.9.0 · 5737 in / 1258 out tokens · 31642 ms · 2026-05-23T17:47:02.994809+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Forward citations

Cited by 6 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score.

  1. TourMart: A Parametric Audit Instrument for Commission Steering in LLM Travel Agents

    cs.CY 2026-05 unverdicted novelty 7.0

    TourMart quantifies commission steering in LLM travel agents via paired counterfactual prompts, reporting 3.5-7.7 percentage point increases in steered recommendations for tested models.

  2. LLMs can persuade only psychologically susceptible humans on societal issues, via trust in AI and emotional appeals, amid logical fallacies

    cs.AI 2026-04 unverdicted novelty 7.0

    LLMs persuade only psychologically susceptible humans on societal issues through trust in AI and emotional appeals, while both sides rely on logical fallacies in roughly one out of every six conversational turns.

  3. Spontaneous Persuasion: An Audit of Model Persuasiveness in Everyday Conversations

    cs.HC 2026-04 unverdicted novelty 6.0

    LLMs engage in spontaneous persuasion in virtually all multi-turn conversations by favoring information-based strategies like logic and evidence, in contrast to human responses that rely more on social influence and n...

  4. Ads in AI Chatbots? An Analysis of How Large Language Models Navigate Conflicts of Interest

    cs.AI 2026-04 unverdicted novelty 6.0

    Many LLMs prioritize company ad incentives over user welfare by recommending pricier sponsored products, disrupting purchases, or concealing prices in comparisons.

  5. Persuadability and LLMs as Legal Decision Tools

    cs.AI 2026-04 unverdicted novelty 5.0

    Frontier LLMs exhibit persuadability to legal arguments that varies with the perceived quality of the advocate presenting them.

  6. Human Decision-Making with Persuasive and Narrative LLM Explanations

    cs.HC 2026-05 unverdicted novelty 4.0

    LLM narrative explanations of varying persuasiveness did not improve human decision accuracy over AI predictions alone but increased reliance on AI even when incorrect.

Reference graph

Works this paper leans on

47 extracted references · 47 canonical work pages · cited by 6 Pith papers · 1 internal anchor

  1. [1]

    Misinformation on misinformation: Conceptual and method- ological challenges

    Sacha Altay, Manon Berriche, and Alberto Acerbi. Misinformation on misinformation: Conceptual and method- ological challenges. Social media+ society, 9(1):20563051221150412, 2023

    work page 2023

  2. [2]

    Information delivered by a chatbot has a positive impact on COVID-19 vaccines attitudes and intentions

    Sacha Altay, Anne-Sophie Hacquin, Coralie Chevallier, and Hugo Mercier. Information delivered by a chatbot has a positive impact on COVID-19 vaccines attitudes and intentions. Journal of Experimental Psychology. Applied, 29(1):52–62, March 2023. ISSN 1939-2192. doi: 10.1037/xap0000400

    work page doi:10.1037/xap0000400 2023

  3. [3]

    Artificial Intelligence Can Persuade Humans on Political Issues, September 2023

    Hui Bai, Jan V oelkel, Johannes Eichstaedt, and Robb Willer. Artificial Intelligence Can Persuade Humans on Political Issues, September 2023. URL https://www.researchsquare.com/article/rs-3238396/v1. ISSN: 2693-5015

    work page 2023

  4. [4]

    Measurement instruments for the anthro- pomorphism, animacy, likeability, perceived intelligence, and perceived safety of robots

    Christoph Bartneck, Dana Kulic, Elizabeth Croft, and Susana Zoghbi. Measurement instruments for the anthro- pomorphism, animacy, likeability, perceived intelligence, and perceived safety of robots. International Journal of Social Robotics, 1:71–81, 01 2008. doi: 10.1007/s12369-008-0001-3

    work page doi:10.1007/s12369-008-0001-3 2008

  5. [5]

    The Persuasive Power of Large Language Models

    Simon Martin Breum, Daniel Vædele Egdal, Victor Gram Mortensen, Anders Giovanni Møller, and Luca Maria Aiello. The Persuasive Power of Large Language Models. Proceedings of the International AAAI Conference on Web and Social Media, 18:152–163, May 2024. ISSN 2334-0770. doi: 10.1609/icwsm.v18i1.31304. URL https://ojs.aaai.org/index.php/ICWSM/article/view/31304

    work page doi:10.1609/icwsm.v18i1.31304 2024

  6. [6]

    Misunderstanding the harms of online misinformation

    Ceren Budak, Brendan Nyhan, David M Rothschild, Emily Thorson, and Duncan J Watts. Misunderstanding the harms of online misinformation. Nature, 630(8015):45–53, 2024. 13 Persuasion with Large Language Models: a Survey A PREPRINT

    work page 2024

  7. [7]

    Large language models reflect the ideology of their creators, 2024

    Maarten Buyl, Alexander Rogiers, Sander Noels, Iris Dominguez-Catena, Edith Heiter, Raphael Romero, Iman Johary, Alexandru-Cristian Mara, Jefrey Lijffijt, and Tijl De Bie. Large language models reflect the ideology of their creators, 2024. URL https://arxiv.org/abs/2410.18417

    work page arXiv 2024

  8. [8]

    People devalue generative AI’s compe- tence but not its advice in addressing societal and personal challenges

    Robert B ¨ohm, Moritz J ¨orling, Leonhard Reiter, and Christoph Fuchs. People devalue generative AI’s compe- tence but not its advice in addressing societal and personal challenges. Communications Psychology, 1(1):1–10, November 2023. ISSN 2731-9121. doi: 10.1038/s44271-023-00032-x. URL https://www.nature.com/ articles/s44271-023-00032-x . Publisher: Natu...

    work page doi:10.1038/s44271-023-00032-x 2023

  9. [9]

    arXiv preprint arXiv:2404.09329 , year=

    Carlos Carrasco-Farre. Large Language Models are as persuasive as humans, but how? About the cognitive effort and moral-emotional language of LLM arguments, April 2024. URL http://arxiv.org/abs/2404.09329. arXiv:2404.09329 [cs]

    work page arXiv 2024

  10. [10]

    Characterizing Manipulation from AI Systems, October 2023

    Micah Carroll, Alan Chan, Henry Ashton, and David Krueger. Characterizing Manipulation from AI Systems, October 2023. URL http://arxiv.org/abs/2303.09387. arXiv:2303.09387 [cs]

    work page arXiv 2023

  11. [11]

    Combating Misinformation in the Age of LLMs: Opportunities and Challenges, November 2023

    Canyu Chen and Kai Shu. Combating Misinformation in the Age of LLMs: Opportunities and Challenges, November 2023. URL http://arxiv.org/abs/2311.05656. arXiv:2311.05656 [cs]

    work page arXiv 2023

  12. [12]

    Would an AI chatbot persuade you: an empirical answer from the elaboration likelihood model

    Qian Chen, Changqin Yin, and Yeming Gong. Would an AI chatbot persuade you: an empirical answer from the elaboration likelihood model. Information Technology & People, ahead-of-print(ahead-of-print), January 2023. ISSN 0959-3845. doi: 10.1108/ITP-10-2021-0764. URL https://doi.org/10.1108/ITP-10-2021-0764

    work page doi:10.1108/itp-10-2021-0764 2023

  13. [13]

    Costello, Gordon Pennycook, and David Rand

    Thomas H. Costello, Gordon Pennycook, and David Rand. Durably reducing conspiracy beliefs through dia- logues with AI, April 2024. URL https://osf.io/xcwdn

    work page 2024

  14. [14]

    Measur- ing the persuasiveness of language models, April 2024

    Esin Durmus, Liane Lovitt, Alex Tamkin, Stuart Ritchie, Jack Clark, and Deep Ganguli. Measur- ing the persuasiveness of language models, April 2024. URL https://www.anthropic.com/news/ measuring-model-persuasiveness

    work page 2024

  15. [15]

    Misinformation poses a bigger threat to democracy than you might think

    Ullrich Ecker, Jon Roozenbeek, Sander van der Linden, Li Qian Tay, John Cook, Naomi Oreskes, and Stephan Lewandowsky. Misinformation poses a bigger threat to democracy than you might think. Nature, 630(8015): 29–32, 2024

    work page 2024

  16. [16]

    Semnani, Kazushi Ikeda, Weiyan Shi, and Monica S

    Kazuaki Furumai, Roberto Legaspi, Julio Vizcarra, Yudai Yamazaki, Yasutaka Nishimura, Sina J. Semnani, Kazushi Ikeda, Weiyan Shi, and Monica S. Lam. Zero-shot Persuasive Chatbots with LLM-Generated Strategies and Information Retrieval, July 2024. URL http://arxiv.org/abs/2407.03585. arXiv:2407.03585 [cs]

    work page arXiv 2024

  17. [17]

    How persuasive is AI- generated propaganda? PNAS Nexus, 3(2):pgae034, February 2024

    Josh A Goldstein, Jason Chao, Shelby Grossman, Alex Stamos, and Michael Tomz. How persuasive is AI- generated propaganda? PNAS Nexus, 3(2):pgae034, February 2024. ISSN 2752-6542. doi: 10.1093/pnasnexus/ pgae034. URL https://doi.org/10.1093/pnasnexus/pgae034

    work page doi:10.1093/pnasnexus/ 2024

  18. [18]

    Griffin, Bennett Kleinberg, Maximilian Mozes, Kimberly T

    Lewis D. Griffin, Bennett Kleinberg, Maximilian Mozes, Kimberly T. Mai, Maria Vau, Matthew Caldwell, and Augustine Marvor-Parker. Susceptibility to Influence of Large Language Models, March 2023. URL http: //arxiv.org/abs/2303.06074. arXiv:2303.06074 [cs]

    work page arXiv 2023

  19. [19]

    Online persuasion and compliance: Social influence on the internet and beyond

    Rosanna E Guadagno and Robert B Cialdini. Online persuasion and compliance: Social influence on the internet and beyond. The social net: The social psychology of the Internet, pages 91–113, 2005

    work page 2005

  20. [20]

    Evaluating the persuasive influence of political microtargeting with large language models

    Kobi Hackenburg and Helen Margetts. Evaluating the persuasive influence of political microtargeting with large language models. Proceedings of the National Academy of Sciences , 121(24):e2403116121, June 2024. doi: 10.1073/pnas.2403116121. URL https://www.pnas.org/doi/10.1073/pnas.2403116121. Publisher: Proceedings of the National Academy of Sciences

    work page doi:10.1073/pnas.2403116121 2024

  21. [21]

    Tappin, and Manos Tsakiris

    Kobi Hackenburg, Lujain Ibrahim, Ben M. Tappin, and Manos Tsakiris. Comparing the persuasiveness of role- playing large language models and human experts on polarized U.S. political issues, December 2023. URL https://osf.io/ey8db

    work page 2023

  22. [22]

    Tappin, Paul R ¨ottger, Scott Hale, Jonathan Bright, and Helen Margetts

    Kobi Hackenburg, Ben M. Tappin, Paul R ¨ottger, Scott Hale, Jonathan Bright, and Helen Margetts. Evidence of a log scaling law for political persuasion with large language models, June 2024. URL http://arxiv.org/ abs/2406.14508. arXiv:2406.14508 [cs]

    work page arXiv 2024

  23. [23]

    An Overview of Catastrophic AI Risks, October

    Dan Hendrycks, Mantas Mazeika, and Thomas Woodside. An Overview of Catastrophic AI Risks, October

    work page

  24. [24]

    An Overview of Catastrophic AI Risks

    URL http://arxiv.org/abs/2306.12001. arXiv:2306.12001 [cs]

    work page internal anchor Pith review arXiv

  25. [25]

    Personalized persuasion: Tailoring persuasive appeals to recipients’ personality traits

    Jacob B Hirsh, Sonia K Kang, and Galen V Bodenhausen. Personalized persuasion: Tailoring persuasive appeals to recipients’ personality traits. Psychological science, 23(6):578–581, 2012

    work page 2012

  26. [26]

    Journalism,‘Fake News’& Disinformation: Handbook for Journalism Education and Training

    Cherilyn Ireton. Journalism,‘Fake News’& Disinformation: Handbook for Journalism Education and Training. UNESCO Publishing, 2018. 14 Persuasion with Large Language Models: a Survey A PREPRINT

    work page 2018

  27. [27]

    Foundations for an empirically determined scale of trust in automated systems

    Jiun-Yin Jian, Ann Bisantz, and Colin Drury. Foundations for an empirically determined scale of trust in automated systems. International Journal of Cognitive Ergonomics , 4:53–71, 03 2000. doi: 10.1207/ S15327566IJCE0401 04

    work page 2000

  28. [28]

    Personalizing persuasive technolo- gies: Explicit and implicit personalization using persuasion profiles

    Maurits Kaptein, Panos Markopoulos, Boris De Ruyter, and Emile Aarts. Personalizing persuasive technolo- gies: Explicit and implicit personalization using persuasion profiles. International Journal of Human-Computer Studies, 77:38–51, 2015

    work page 2015

  29. [29]

    Elise Karinshak, Sunny Xun Liu, Joon Sung Park, and Jeffrey T. Hancock. Working With AI to Persuade: Examining a Large Language Model’s Ability to Generate Pro-Vaccination Messages. Proceedings of the ACM on Human-Computer Interaction, 7(CSCW1):1–29, April 2023. ISSN 2573-0142. doi: 10.1145/3579592. URL https://dl.acm.org/doi/10.1145/3579592

    work page doi:10.1145/3579592 2023

  30. [30]

    Quantifying the Impact of Large Language Models on Collective Opinion Dynamics, January 2024

    Chao Li, Xing Su, Haoying Han, Cong Xue, Chunmo Zheng, and Chao Fan. Quantifying the Impact of Large Language Models on Collective Opinion Dynamics, January 2024. URL https://papers.ssrn.com/ abstract=4688547

    work page 2024

  31. [31]

    The effect of source disclosure on evaluation of AI-generated messages: A two- part study

    Sue Lim and Ralf Schm ¨alzle. The effect of source disclosure on evaluation of AI-generated messages: A two- part study. Computers in Human Behavior: Artificial Humans , 2(1):100058, January 2024. ISSN 29498821. doi: 10.1016/j.chbah.2024.100058. URL http://arxiv.org/abs/2311.15544. arXiv:2311.15544 [cs]

    work page doi:10.1016/j.chbah.2024.100058 2024

  32. [32]

    S. C. Matz, J. D. Teeny, S. S. Vaid, H. Peters, G. M. Harari, and M. Cerf. The potential of generative AI for personalized persuasion at scale. Scientific Reports, 14(1):4692, February 2024. ISSN 2045-2322. doi: 10. 1038/s41598-024-53755-0. URL https://www.nature.com/articles/s41598-024-53755-0 . Publisher: Nature Publishing Group

    work page 2024

  33. [33]

    How Good are LLMs in Generating Personalized Advertisements? In Companion Proceedings of the ACM on Web Conference 2024, WWW ’24, pages 826–829, New York, NY , USA, May 2024

    Elyas Meguellati, Lei Han, Abraham Bernstein, Shazia Sadiq, and Gianluca Demartini. How Good are LLMs in Generating Personalized Advertisements? In Companion Proceedings of the ACM on Web Conference 2024, WWW ’24, pages 826–829, New York, NY , USA, May 2024. Association for Computing Machinery. ISBN 9798400701726. doi: 10.1145/3589335.3651520. URL https:/...

    work page doi:10.1145/3589335.3651520 2024

  34. [34]

    Empowering Calibrated (Dis-)Trust in Conversational Agents: A User Study on the Persuasive Power of Limitation Disclaimers vs

    Luise Metzger, Linda Miller, Martin Baumann, and Johannes Kraus. Empowering Calibrated (Dis-)Trust in Conversational Agents: A User Study on the Persuasive Power of Limitation Disclaimers vs. Authoritative Style. In Proceedings of the CHI Conference on Human Factors in Computing Systems , CHI ’24, pages 1– 19, New York, NY , USA, May 2024. Association for...

    work page doi:10.1145/3613904.3642122 2024

  35. [35]

    Large Language Models Can Argue in Convincing Ways About Politics, But Humans Dislike AI Authors: implications for Governance

    Alexis Palmer and Arthur Spirling. Large Language Models Can Argue in Convincing Ways About Politics, But Humans Dislike AI Authors: implications for Governance. Political Science, 75(3):281–291, September 2023. ISSN 0032-3187, 2041-0611. doi: 10.1080/00323187.2024.2335471. URL https://www.tandfonline.com/ doi/full/10.1080/00323187.2024.2335471

    work page doi:10.1080/00323187.2024.2335471 2023

  36. [36]

    Measuring and Benchmarking Large Language Models’ Capabilities to Generate Persuasive Language, June 2024

    Amalie Brogaard Pauli, Isabelle Augenstein, and Ira Assent. Measuring and Benchmarking Large Language Models’ Capabilities to Generate Persuasive Language, June 2024. URL http://arxiv.org/abs/2406. 17753. arXiv:2406.17753 [cs]

    work page arXiv 2024

  37. [37]

    On the Conversational Persuasive- ness of Large Language Models: A Randomized Controlled Trial, March 2024

    Francesco Salvi, Manoel Horta Ribeiro, Riccardo Gallotti, and Robert West. On the Conversational Persuasive- ness of Large Language Models: A Randomized Controlled Trial, March 2024. URL http://arxiv.org/ abs/2403.14380. arXiv:2403.14380 [cs]

    work page arXiv 2024

  38. [38]

    The general attitudes towards artificial intelligence scale (gaais): Confir- matory validation and associations with personality, corporate distrust, and general trust

    Astrid Schepman and Paul Rodway. The general attitudes towards artificial intelligence scale (gaais): Confir- matory validation and associations with personality, corporate distrust, and general trust. International Jour- nal of Human–Computer Interaction , 39(13):2724–2741, 2023. doi: 10.1080/10447318.2022.2085400. URL https://doi.org/10.1080/10447318.20...

    work page doi:10.1080/10447318.2022.2085400 2023

  39. [39]

    Large Language Models Can Enhance Persuasion Through Linguistic Feature Align- ment, February 2024

    Minkyu Shin and Jin Kim. Large Language Models Can Enhance Persuasion Through Linguistic Feature Align- ment, February 2024. URL https://papers.ssrn.com/abstract=4725351

    work page 2024

  40. [40]

    The persuasive effects of political microtarget- ing in the age of generative artificial intelligence

    Almog Simchon, Matthew Edwards, and Stephan Lewandowsky. The persuasive effects of political microtarget- ing in the age of generative artificial intelligence. PNAS Nexus, 3(2):pgae035, February 2024. ISSN 2752-6542. doi: 10.1093/pnasnexus/pgae035. URL https://doi.org/10.1093/pnasnexus/pgae035

    work page doi:10.1093/pnasnexus/pgae035 2024

  41. [41]

    AI model GPT-3 (dis)informs us better than humans

    Giovanni Spitale, Nikola Biller-Andorno, and Federico Germani. AI model GPT-3 (dis)informs us better than humans. Science Advances, 9(26):eadh1850, June 2023. doi: 10.1126/sciadv.adh1850. URL https://www. science.org/doi/10.1126/sciadv.adh1850. Publisher: American Association for the Advancement of Science

    work page doi:10.1126/sciadv.adh1850 2023

  42. [42]

    Persuasive communication

    James B Stiff and Paul A Mongeau. Persuasive communication. Guilford Publications, 2016. 15 Persuasion with Large Language Models: a Survey A PREPRINT

    work page 2016

  43. [43]

    Persuasiveness of arguments with AI-source labels

    Cassandra Teigen, Jens Koed Madsen, Nicole Lauren George, and Sayeh Yousefi. Persuasiveness of arguments with AI-source labels. Proceedings of the Annual Meeting of the Cognitive Science Society , 46(0), 2024. URL https://escholarship.org/uc/item/6t82g70v

    work page 2024

  44. [44]

    The importance of epistemology for the study of misinfor- mation

    Joseph Uscinski, Shane Littrell, and Casey Klofstad. The importance of epistemology for the study of misinfor- mation. Current opinion in psychology, page 101789, 2024

    work page 2024

  45. [45]

    Designing and Evaluating Multi-Chatbot Interface for Human-AI Communication: Preliminary Findings from a Persuasion Task, June 2024

    Sion Yoon, Tae Eun Kim, and Yoo Jung Oh. Designing and Evaluating Multi-Chatbot Interface for Human-AI Communication: Preliminary Findings from a Persuasion Task, June 2024. URL http://arxiv.org/abs/ 2406.19648. arXiv:2406.19648 [cs]

    work page arXiv 2024

  46. [46]

    Human favoritism, not AI aversion: People’s perceptions (and bias) toward generative AI, human experts, and human–GAI collaboration in persuasive content generation

    Yunhao Zhang and Ren ´ee Gosline. Human favoritism, not AI aversion: People’s perceptions (and bias) toward generative AI, human experts, and human–GAI collaboration in persuasive content generation. Judgment and Decision Making , 18:e41, January 2023. ISSN 1930-2975. doi: 10.1017/jdm.2023.37. URL https://www.cambridge.org/core/journals/judgment-and-decis...

    work page doi:10.1017/jdm.2023.37 2023

  47. [47]

    Synthetic Lies: Un- derstanding AI-Generated Misinformation and Evaluating Algorithmic and Human Solutions

    Jiawei Zhou, Yixuan Zhang, Qianni Luo, Andrea G Parker, and Munmun De Choudhury. Synthetic Lies: Un- derstanding AI-Generated Misinformation and Evaluating Algorithmic and Human Solutions. In Proceedings of the 2023 CHI Conference on Human Factors in Computing Systems, CHI ’23, pages 1–20, New York, NY , USA, April 2023. Association for Computing Machiner...

    work page doi:10.1145/3544548.3581318 2023