Pith. sign in

REVIEW 2 major objections 5 minor 3 cited by

The Levers of Political Persuasion with Conversational AI

T0 review · 2 major / 5 minor · reviewed 2026-08-06 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗

Pith's one-line read In three large experiments, the paper finds that AI persuasion is driven mainly by post-training and prompting that increase how much information a model deploys, and that these levers systematically reduce factual accuracy.

desk verdict A large, carefully built empirical study whose persuasion-levers conclusion holds up, while the accuracy-tradeoff headline needs better fact-checker validation before it can be taken at face value. read the letter →

arxiv 2507.13919 v1 pith:2YTRAVND submitted 2025-07-18 cs.CL cs.AIcs.CYcs.HC

Kobi Hackenburg , Ben M. Tappin , Luke Hewitt , Ed Saunders , Sid Black , Hause Lin , Catherine Fist , Helen Margetts
show 2 more authors
David G. Rand Christopher Summerfield
This is my paper · ORCID
classification cs.CLcs.AIcs.CYcs.HC
keywords AIpersuasionpoliticalattitudeslargelanguagemodelsinformationdensityrewardmodelingfactualaccuracyconversationalpost-training
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

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

The reading

Across three experiments with 76,977 U.K. participants, the authors had 19 large language models argue for a stance on one of 707 political issues, then measured attitude shifts and fact-checked 466,769 claims the models produced. They find that model scale shows a reliable positive association with persuasion when post-training is held constant, but that post-training and prompting are larger levers: reward-model post-training boosted persuasion by as much as 51%, and an information prompt that tells the model to deploy facts and evidence was 27% more persuasive than a basic prompt. The mechanism identified is information density: factors that raised the number of fact-checkable claims per conversation also raised persuasion, while the effect of personalization was consistently small, below 1pp. The paper also reports a consistent accuracy cost: the same factors that increased persuasion lowered the proportion of claims rated factually accurate, with the most persuasive conditions producing 29.7% inaccurate claims versus a 16.0% average. The authors conclude that the persuasive power of current and near-future AI will come mainly from post-training and prompting choices, not from scaling or user data.

What carries the argument

The central mechanism is information density, defined as the number of fact-checkable claims a model makes per conversation, extracted from 91,000 conversations by GPT-4o and validated against professional human fact-checkers with a correlation of r = 0.87 for claim counts. This measure carries the argument: across randomized conditions, information density explains roughly 44% of the variability in persuasive effects, and every major persuasion-boosting intervention examined, information prompting, reward-model post-training, and newer frontier post-training, also increases information density. The second key mechanism is the persuasion reward model, a GPT-4o fine-tuned on 56,283 conversations to predict belief change at each turn and used to select the best of 12 to 20 candidate replies; this is the post-training procedure that turns a small open model into a frontier-competitive persuader while increasing inaccurate claims. Claim accuracy is measured by a search-enabled GPT-4o judge validated against professional human fact-checkers at r = 0.84.

What would settle it

Ask professional fact-checkers, blind to condition, to rate a stratified sample of claims from the information-prompted and non-information conditions of GPT-4o (3/25), GPT-4.5, and the reward-model-tuned Llama models, and compare their ratings with the gpt-4o-search-preview ratings; if the human ratings do not reproduce the reported accuracy drops, such as 62% versus 78% for GPT-4o (3/25) or the 2.22pp accuracy drop from reward modeling, then the persuasion-accuracy trade-off is at least partly an artifact of the judge.

Watch

Extended reading notes

Core claim

On the paper's own terms, the central discovery is that conversational AI persuades by mobilizing large volumes of information, and that the methods which most increase this information flow also reduce factual accuracy. When post-training is held constant, larger models are reliably more persuasive, with an order-of-magnitude compute increase buying roughly 1.6 to 1.8 percentage points of persuasion, but this scaling return is eclipsed by post-training: a newer deployment of GPT-4o was 3.50pp more persuasive than an older deployment of the same model, which exceeds the predicted gain from a 10-fold or even 100-fold compute increase. Applying a reward model trained to pick persuasive replies raised a small open-source model, Llama3.1-8B, from about 6pp to about 9pp, matching or beating GPT-4o (8/24). Prompt-level analysis shows that information density, defined as the number of fact-checkable claims per conversation, predicts persuasiveness with a meta-analytic correlation of r = 0.76 and explains 44% of the variability in persuasive effects across randomized conditions. Accuracy analyses show that the same levers reduce factual accuracy: the information prompt dropped GPT-4o (3/25) accuracy from 78% to 62%, and reward modeling on chat-tuned models added 2.32pp of persuasion while reducing the proportion of accurate claims by 2.22pp.

Load-bearing premise

The load-bearing assumption is that the AI fact-checker scores claims from every model, prompt, and post-training condition with the same accuracy standard; it was checked against human fact-checkers on only 198 messages from one deployment of GPT-4o, so a judge biased toward certain styles could make the accuracy drop look larger or smaller than it is.

Editorial extensions

If this is right

  • Fine-tuning a small open-source model with a persuasion reward model can make it as persuasive as a frontier model, so high-persuasion AI is accessible to actors who cannot afford frontier-scale compute.
  • Frontier persuasion gains are more likely to come from developer post-training updates than from scale; the paper estimates the gain from one GPT-4o post-training update exceeded predicted gains from a 10x or 100x compute increase.
  • The single most effective prompting strategy among the eight tested is telling the model to provide information, and it works by increasing the number of fact-checkable claims per conversation.
  • Optimizing AI for persuasion is associated with a systematic loss of factual accuracy, including in the most persuasive conditions where nearly 30% of claims were rated inaccurate.
  • Conversational AI is 41% to 52% more persuasive than a static AI message, so interactive deployment is the relevant risk surface for near-future persuasion.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • Editorial extension: if the persuasion-accuracy trade-off is a stable property of optimizing for persuasion, then treating persuasiveness as an unregularized training target will keep pulling models toward inaccuracy; a testable fix is adding explicit accuracy penalties to the reward model and checking whether the persuasion gains survive.
  • Editorial extension: because the AI fact-checker was validated on only 198 messages from one model, a replication that validates the judge across GPT-4.5, Grok-3, and reward-model-tuned Llama outputs would show whether the reported accuracy decline is real or partly a measurement artifact.
  • Editorial extension: the paper leaves open whether the accuracy drop comes from generating more claims or from reward selection favoring confident-sounding falsehoods; an experiment holding information density constant while varying accuracy incentives would separate the two.
  • Editorial extension: the large controlled-condition effects may overstate real-world influence, since people outside a paid survey may not sustain long political conversations; voluntary-exposure field studies would test this bottleneck.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

2 major / 5 minor

Summary. This paper uses three large-scale randomized survey experiments (N = 76,977) in the UK to measure the persuasiveness of 19 conversational LLMs across 707 political issues, manipulating model scale (via effective compute), post-training (supervised fine-tuning and reward modeling), prompting strategy (eight rhetorical prompts, including an information-focused prompt), and personalization. Persuasion is measured as the difference in post-conversation attitude agreement relative to a no-conversation control, and factual accuracy is measured by an LLM-based fact-checking pipeline that rated 466,769 extracted claims on a 0–100 veracity scale. The authors report that scale has a positive but modest association with persuasiveness among uniformly chat-tuned models; reward-model post-training and information-dense prompting produce the largest persuasion gains; personalization has small effects; and the factors that increase persuasion also tend to increase information density and decrease the rated accuracy of claims. They extrapolate that a 'maximal-persuasion' AI condition would shift attitudes by 15.9pp overall and 26.5pp among initial disagreers, with nearly 30% of its claims rated inaccurate.

Significance. If the results hold, they would substantially revise current debates about AI persuasion: the primary levers would be developer post-training and prompting choices that encourage information-dense argumentation, rather than model scale or user personalization, and these levers would come at a measurable cost to factual accuracy. The evidentiary base is unusually strong for this literature: three pre-registered experiments with large samples, a range of open- and closed-source models, robustness checks including attrition imputation and a one-month durability follow-up, and publicly available code and materials. The main weakness is that the accuracy and information-density measurements rely on an LLM-based pipeline validated on only 198 messages from one model; because the headline tradeoff claim depends on cross-condition comparisons, this validation gap is the key threat to the paper's central conclusion.

major comments (2)
  1. [Methods §1.8, Results 'How do models persuade?' and 'How accurate is the information provided by the models?'] The paper's two central measurement claims—the information-density mechanism and the persuasion–accuracy tradeoff—both depend on an LLM-based measurement pipeline (GPT-4o for claim extraction, gpt-4o-search-preview for fact-checking) whose validation is limited to 198 messages from a single model (GPT-4o 8/24, Study 1 Chat 2). The reported correlations (r = 0.87 for counts, r = 0.84 for accuracy) demonstrate overall agreement with human raters but do not establish measurement invariance across the models, prompts, and post-training conditions whose differences drive the paper's conclusions. A judge or extractor that is differentially lenient or strict for particular conditions (e.g., information-prompted, claim-dense messages or newer GPT-4o versions) could produce the observed pattern—higher information density, lower rated accuracy, and higher persuasion—even if the models' true characteristics were unchanged. Please extend validation to a stratified sample spanning each model family, prompt type, and post-training condition, and report within-stratum human–machine agreement; if this is infeasible, the abstract and Discussion should explicitly hedge the accuracy and mechanism claims.
  2. [Results 'How accurate is the information provided by the models?', Figure 4] The accuracy analyses use the per-conversation proportion of claims rated >50/100 as the dependent variable. Because the information prompt and reward modeling also increase the number of claims per conversation, a judge whose ratings are noisier or systematically lower for additional or more marginal claims would make the accuracy tradeoff appear larger than it is, even if per-claim truthfulness were unchanged. The paper should test the robustness of the Figure 4B–C contrasts to controlling for claim count (e.g., condition-level regressions of accuracy on persuasion levers that include total claims) or report the accuracy comparison restricted to the first k claims of each conversation. This is important because the headline claim that persuasion and accuracy are in systematic tension is stated without such a control.
minor comments (5)
  1. [Methods §1.8 Fact-checking] The fact-checking pipeline was implemented between April 1st and May 18th, 2025; if the gpt-4o-search-preview model version changed during that window, temporal drift could confound the across-model accuracy comparisons. Please report the exact model version(s) used and test for time trends in ratings.
  2. [References] The reference to 'The Brms Book' as an early draft ([6]) is not a stable citation for a journal submission; please replace it with a published reference or the package documentation.
  3. [Figure 4A] Figure 4A removes some model labels 'for clarity'; given that the accuracy findings for GPT-4.5 and GPT-3.5 are surprising, a full version with all labels (or a corresponding table in the main text) would aid the reader.
  4. [Results 'maximal-persuasion' analysis] The maximal-persuasion estimate (15.9pp overall; 26.5pp among initial disagreers) is reported as an observed mean of the 500 conversations predicted to be most persuasive by a cross-fit random forest, but no uncertainty interval or correction for selection is provided; please report a bootstrap CI or state that this is an exploratory upper-bound estimate.
  5. [Results 'durability' analysis] The persistence analysis (36–42% of the effect after one month) is only conducted in Study 1 with one model; consider reporting whether this differs across conditions or noting the limitation.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the central claims rest on randomized measurements and external validation, not on inputs defined as outputs.

full rationale

The paper's core claims are measured interventions rather than definitional equivalences. Persuasion effects are estimated from randomized pre/post attitude comparisons against control groups (Methods 1.6, 1.7), and post-training (SFT, RM) and prompting conditions are manipulated and evaluated on new, held-out participants, so the headline results (RM +2.32pp, information prompt +2.29pp, GPT-4o 3/25 vs 8/24 +3.50pp) are not constructed from their inputs. The reward model was trained on earlier conversations and then applied to novel conversations and participants in Studies 2 and 3, which is external validation. The information-density–persuasion association is estimated from prompt-level exogenous variation under randomization, not from a fitted parameter renamed as an outcome. The accuracy ratings use an external LLM judge validated against professional human fact-checkers on a subset of messages (Methods 1.8); the restricted validation coverage is a measurement-validity concern, not circularity, because the ratings are not algebraically derived from the conditions being compared. The two self-citations ([25], [26]) are background and motivational, not load-bearing for any central result. The cross-fit maximal-persuasion estimate reports observed effects on selected conversations, so any optimism is a potential selection-bias issue rather than a circular reduction. No step equates an output to an input by definition or via a self-citation chain.

Assumptions & free parameters 2 free parameters · 4 assumptions · 0 invented entities

The central claims rest on measurement and external-input assumptions rather than on mathematical axioms. The largest uncertainties are the validity of the AI fact-checker across models and the representativeness of immediate self-reported attitude change.

free parameters (2)
  • Scale-persuasiveness slope (Bayesian meta-regression) = +1.59pp per log10 compute (95% CI [1.07, 2.13])
    Fitted to the pooled scaling curve across studies; used to compare returns to scale against post-training gains. This is a data-estimated regression coefficient, not a theoretical constant.
  • Information-density to persuasion slope = +0.30pp per additional claim (95% CI [0.23, 0.38])
    Estimated from prompt-level means via Bayesian meta-regression; used to argue that information density drives persuasion.
assumptions (4)
  • domain assumption Self-reported attitude change on a 0 to 100 scale, immediately after a two-to-ten turn chat, measures political persuasion.
    Used in all three studies to define the outcome. The authors validate durability for one model (36-42% of the effect persists at one month) but not for all models and conditions.
  • domain assumption gpt-4o-search-preview ratings are valid factual accuracy measures for claims from all tested models.
    Validated against human fact-checkers on 198 messages from GPT-4o 8/24 only (Methods section "Validation"). Applied without per-model validation to all models including GPT-4.5, Grok-3, and Llama variants.
  • domain assumption Prompt-level variation in information density is exogenous and can identify the persuasion mechanism.
    The prompt-level meta-regression assumes random assignment of prompts makes prompt means exogenous for estimating the information-density slope (Methods, "Examining how the models persuade").
  • domain assumption Effective compute values for closed models are accurate enough for the scaling regression.
    GPT-4.5 and Grok-3 compute values are approximations from EpochAI and industry insiders (Table 1 notes). The log-scale regression treats them as fixed inputs.

how reviews work

0 comments
Cite this review

Pith. "Pith review of The Levers of Political Persuasion with Conversational AI." pith.science (2026). https://pith.science/paper/2YTRAVND

@misc{pith2026250713919,
  author       = {Pith},
  title        = {Pith review of: The Levers of Political Persuasion with Conversational AI},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/2YTRAVND}},
  note         = {Machine review of arXiv:2507.13919}
}
read the original abstract

There are widespread fears that conversational AI could soon exert unprecedented influence over human beliefs. Here, in three large-scale experiments (N=76,977), we deployed 19 LLMs-including some post-trained explicitly for persuasion-to evaluate their persuasiveness on 707 political issues. We then checked the factual accuracy of 466,769 resulting LLM claims. Contrary to popular concerns, we show that the persuasive power of current and near-future AI is likely to stem more from post-training and prompting methods-which boosted persuasiveness by as much as 51% and 27% respectively-than from personalization or increasing model scale. We further show that these methods increased persuasion by exploiting LLMs' unique ability to rapidly access and strategically deploy information and that, strikingly, where they increased AI persuasiveness they also systematically decreased factual accuracy.

Figures

Figures reproduced from arXiv: 2507.13919 by the authors.

Figure 1
Figure 1. Persuasiveness of conversational AI increases with model scale. [PITH_FULL_IMAGE:figures/full_fig_p004_1.png] view at source ↗
Figure 2
Figure 2. Persuasion post-training (PPT) can substantially increase the persuasiveness of [PITH_FULL_IMAGE:figures/full_fig_p005_2.png] view at source ↗
Figure 3
Figure 3. Persuasion increases with information density. (A) [PITH_FULL_IMAGE:figures/full_fig_p007_3.png] view at source ↗
Figures from the paper (1 more)
Figure 4
Figure 4. Figure 4: Factors which made conversational AI more persuasive tended to decrease factual [PITH_FULL_IMAGE:figures/full_fig_p008_4.png]

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 3 Pith papers

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

  1. Tailored untruths: How personalisation challenges LLM safeguards

    cs.CL 2025-10 conditional novelty 7.0 of 10

    A 1.6-million-text study of eight LLMs in four languages finds that adding demographic personae to disinformation prompts raises jailbreak rates from 78% to 82%.

  2. Large language models can effectively convince people to believe conspiracies

    cs.AI 2026-01 conditional novelty 6.0 of 10

    In three experiments, GPT-4o instructed to argue for a conspiracy raised believers' confidence about as much as it lowered it when arguing against; a truth-constraining prompt and a corrective debrief largely undid the harm.

  3. AI Propaganda factories with language models

    cs.CR 2025-08 conditional novelty 5.0 of 10

    Small language models sustain political personas and become more ideologically extreme when replying to counter-arguments, according to a language-model judge.

Reference graph

Works this paper leans on

63 extracted references · 57 canonical work pages · cited by 3 Pith papers

  1. [1]

    Scaling up interactive argumentation by providing counterarguments with a chatbot.Nature Human Behaviour, 6(4):579–592, February 2022

    Sacha Altay, Marlène Schwartz, Anne-Sophie Hacquin, Aurélien Allard, Stefaan Blancke, and Hugo Mercier. Scaling up interactive argumentation by providing counterarguments with a chatbot.Nature Human Behaviour, 6(4):579–592, February 2022

    work page 2022

  2. [2]

    L. P. Argyle, E. C. Busby, J. R. Gubler, A. Lyman, J. Olcott, J. Pond, and D. Wingate. Testing theories of political persuasion using AI.Proceedings of the National Academy of Sciences, 122(18):e2412815122, 2025

    work page 2025

  3. [3]

    Managing extreme AI risks amid rapid progress.Science, 384(6698):842–845, May 2024

    Yoshua Bengio, Geoffrey Hinton, Andrew Yao, Dawn Song, Pieter Abbeel, Trevor Darrell, Yuval Noah Harari, Ya-Qin Zhang, Lan Xue, Shai Shalev-Shwartz, Gillian Hadfield, Jeff Clune, Tegan Maharaj, Frank Hutter, Atılım Güneş Baydin, Sheila McIlraith, Qiqi Gao, Ashwin Acharya, David Krueger, Anca Dragan, Philip Torr, Stuart Russell, Daniel Kahneman, Jan Braune...

    work page 2024

  4. [4]

    Lauderdale

    Jack Blumenau and Benjamin E. Lauderdale. The variable persuasiveness of political rhetoric.American Journal of Political Science, 68:255–270, 1 2024

    work page 2024

  5. [5]

    Durably reducing transphobia: A field experiment on door-to-door canvassing

    David Broockman and Joshua Kalla. Durably reducing transphobia: A field experiment on door-to-door canvassing. Science, 352(6282):220–224, 2016

    work page 2016

  6. [6]

    The Brms Book: Applied Bayesian Regression Modelling Using R and Stan (Early Draft)

    Paul-Christian Bürkner. The Brms Book: Applied Bayesian Regression Modelling Using R and Stan (Early Draft). Unknown, 2024. 16

    work page 2024

  7. [7]

    Artificial Influence: An Analysis Of AI-Driven Persuasion, March 2023

    Matthew Burtell and Thomas Woodside. Artificial Influence: An Analysis Of AI-Driven Persuasion, March 2023

    work page 2023

  8. [8]

    A framework to assess the persuasion risks large language model chatbots pose to democratic societies, 2025

    Zhongren Chen, Joshua Kalla, Quan Le, Shinpei Nakamura-Sakai, Jasjeet Sekhon, and Ruixiao Wang. A framework to assess the persuasion risks large language model chatbots pose to democratic societies, 2025

    work page 2025

Show all 63 references
  1. [9]

    Cialdini, Carl A

    Robert B. Cialdini, Carl A. Kallgren, and Raymond R. Reno. A focus theory of normative conduct: A theoretical refinement and reevaluation of the role of norms in human behavior.Advances in Experimental Social Psychology, 24:201–234, 1991

    1991

  2. [10]

    Generalized Additive Models

    Michael Clark. Generalized Additive Models. 2022

    2022

  3. [11]

    Persuasion in Parallel

    Alexander Coppock. Persuasion in Parallel. How Information Changes Minds about Politics.University of Chicago Press, 2023

    2023

  4. [12]

    Costello, Gordon Pennycook, and David Rand

    Thomas H. Costello, Gordon Pennycook, and David Rand. Durably reducing conspiracy beliefs through dialogues with ai.Science, 2024

    2024

  5. [13]

    Delli Carpini and Scott Keeter.What Americans Know about Politics and Why It Matters

    Michael X. Delli Carpini and Scott Keeter.What Americans Know about Politics and Why It Matters. Yale University Press, New Haven, CT, 1996

    1996

  6. [14]

    Enhancing chat language models by scaling high-quality instructional conversations, 2023

    Ning Ding, Yulin Chen, Bokai Xu, Yujia Qin, Zhi Zheng, Shengding Hu, Zhiyuan Liu, Maosong Sun, and Bowen Zhou. Enhancing chat language models by scaling high-quality instructional conversations, 2023

    2023

  7. [15]

    Measuring the persuasiveness of language models, 2024

    Esin Durmus, Liane Lovitt, Alex Tamkin, Stuart Ritchie, Jack Clark, and Deep Ganguli. Measuring the persuasiveness of language models, 2024

    2024

  8. [16]

    A mechanism-based approach to mitigating harms from persuasive generative ai.arXiv, 2024

    Seliem El-Sayed, Canfer Akbulut, Amanda Mccroskery, Geoff Keeling, Zachary Kenton, Zaria Jalan, Nahema Marchal, Arianna Manzini, Toby Shevlane, Shannon Vallor, Daniel Susser, Matija Franklin, Sophie Bridgers, Harry Law, Matthew Rahtz, Murray Shanahan, Michael Henry Tessler, Ar...

    2024

  9. [17]

    Measuring and improving consistency in pretrained language models.Transactions of the Association for Computational Linguistics, 9:1012–1031, 9 2021

    Yanai Elazar, Nora Kassner, Shauli Ravfogel, Abhilasha Ravichander, Eduard Hovy, Hinrich Schütze, and Yoav Goldberg. Measuring and improving consistency in pretrained language models.Transactions of the Association for Computational Linguistics, 9:1012–1031, 9 2021

    2021

  10. [18]

    Key trends and figures in machine learning, 2023

    Epoch AI. Key trends and figures in machine learning, 2023. Accessed: 2025-06-21

    2023

  11. [19]

    Moral reframing: A technique for effective and persuasive commu- nication across political divides.Social and Personality Psychology Compass, 13, 12 2019

    Matthew Feinberg and Robb Willer. Moral reframing: A technique for effective and persuasive commu- nication across political divides.Social and Personality Psychology Compass, 13, 12 2019

    2019

  12. [20]

    A S Gerber and D P Green.Field Experiments: Design, Analysis, and Interpretation. W. W. Norton, 2012

    2012

  13. [21]

    Goldstein, Jason Chao, Shelby Grossman, Alex Stamos, and Michael Tomz

    Josh A. Goldstein, Jason Chao, Shelby Grossman, Alex Stamos, and Michael Tomz. How persuasive is ai-generated propaganda? PNAS Nexus, 3, 2 2024

    2024

  14. [22]

    Goldstein, Girish Sastry, Micah Musser, Renee DiResta, Matthew Gentzel, and Katerina Sedova

    Josh A. Goldstein, Girish Sastry, Micah Musser, Renee DiResta, Matthew Gentzel, and Katerina Sedova. Generative language models and automated influence operations: Emerging threats and potential mitigations. arXiv, 1 2023

    2023

  15. [23]

    Thousands of ai authors on the future of ai.arXiv, 2024

    Katja Grace, Harlan Stewart, Julia Fabienne Sandkühler, Stephen Thomas, Ben Weinstein-Raun, and Jan Brauner. Thousands of ai authors on the future of ai.arXiv, 2024

    2024

  16. [24]

    The role of transportation in the persuasiveness of public narratives

    Melanie C Green and Timothy C Brock. The role of transportation in the persuasiveness of public narratives. Journal of Personality and Social Psychology, 79(5):701–721, 2000

    2000

  17. [25]

    Evaluating the persuasive influence of political microtargeting with large language models.Proceedings of the National Academy of Sciences, 121:e2403116121, 6 2024

    Kobi Hackenburg and Helen Margetts. Evaluating the persuasive influence of political microtargeting with large language models.Proceedings of the National Academy of Sciences, 121:e2403116121, 6 2024

    2024

  18. [26]

    Tappin, Paul Röttger, Scott Hale, Jonathan Bright, and Helen Margetts

    Kobi Hackenburg, Ben M. Tappin, Paul Röttger, Scott Hale, Jonathan Bright, and Helen Margetts. Scaling language model size yields diminishing returns for single-message political persuasion, 2024

    2024

  19. [27]

    Reflecting on the journey: Mechanisms in narrative persuasion

    Anne Hamby, David Brinberg, and Kim Daniloski. Reflecting on the journey: Mechanisms in narrative persuasion. Journal of Consumer Psychology, 27(1):11–22, 2017

    2017

  20. [28]

    Tappin, James Slezak, Valerie Coffman, Nathaniel Lubin, and Mohammad Hamidian

    Luke Hewitt, David Broockman, Alexander Coppock, Ben M. Tappin, James Slezak, Valerie Coffman, Nathaniel Lubin, and Mohammad Hamidian. How experiments help campaigns persuade voters: Evidence from a large archive of campaigns’ own experiments.American Political Science Review,...

    2024

  21. [29]

    Thompson

    Tiffany Hsu and Stuart A. Thompson. Disinformation Researchers Raise Alarms About A.I. Chatbots. The New York Times, February 2023

    2023

  22. [30]

    Global views on a.i

    Ipsos. Global views on a.i. and disinformation, 2023

    2023

  23. [31]

    Jones and Benjamin K

    Cameron R. Jones and Benjamin K. Bergen. Lies, Damned Lies, and Distributional Language Statistics: Persuasion and Deception with Large Language Models, December 2024

    2024

  24. [32]

    Kalla and David E

    Joshua L. Kalla and David E. Brookman. Reducing exclusionary attitudes through interpersonal conversation: Evidence from three field experiments.American Political Science Review, 114(2):410–425, 2020

    2020

  25. [33]

    Personalizing moral reframing in interpersonal conversation: A field experiment.The Journal of Politics, 84(2), 2022

    Joshua L Kalla, Adam Seth Levine, and David E Broockman. Personalizing moral reframing in interpersonal conversation: A field experiment.The Journal of Politics, 84(2), 2022. Sponsored by the Southern Political Science Association

    2022

  26. [34]

    Scaling laws for neural language models.arXiv, 1 2020

    Jared Kaplan, Sam McCandlish, Tom Henighan, Tom B Brown, Benjamin Chess, Rewon Child, Scott Gray, Alec Radford, Jeffrey Wu, and Dario Amodei. Scaling laws for neural language models.arXiv, 1 2020

    2020

  27. [35]

    Solving quantitative reasoning problems with language models.Advances in Neural Information Processing Systems, 35:3843–3857, 2022

    Aitor Lewkowycz, Anders Andreassen, David Dohan, Ethan Dyer, Henryk Michalewski, Vinay Ramasesh, Ambrose Slind, Cem Anil, Imanol Schlag, Theo Gutman-Solo, et al. Solving quantitative reasoning problems with language models.Advances in Neural Information Processing Systems, 35:...

    2022

  28. [36]

    Hypersuasion – On AI’s Persuasive Power and How to Deal with It.Philosophy & Technology, 37(2):64, May 2024

    Floridi Luciano. Hypersuasion – On AI’s Persuasive Power and How to Deal with It.Philosophy & Technology, 37(2):64, May 2024

    2024

  29. [37]

    How people view ai, disinformation and elections — in charts.Politico, 2023

    Lucia Mackenzie and Mark Scott. How people view ai, disinformation and elections — in charts.Politico, 2023

    2023

  30. [38]

    Someheteroskedasticity-consistentcovariancematrixestimators with improved finite sample properties.Journal of Econometrics, 29(3):305–325, 1985

    JamesG.MacKinnonandHalbertWhite. Someheteroskedasticity-consistentcovariancematrixestimators with improved finite sample properties.Journal of Econometrics, 29(3):305–325, 1985

    1985

  31. [39]

    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

    2024

  32. [40]

    Not Born Yesterday: The Science of Who We Trust and What We Believe

    Hugo Mercier. Not Born Yesterday: The Science of Who We Trust and What We Believe. Princeton University Press, January 2020

    2020

  33. [41]

    Why do humans reason? arguments for an argumentative theory

    Hugo Mercier and Dan Sperber. Why do humans reason? arguments for an argumentative theory. Behavioral and Brain Sciences, 34(2):57–111, 2011

    2011

  34. [42]

    Harvard University Press, September 2018

    Hugo Mercier and Dan Sperber.The Enigma of Reason. Harvard University Press, September 2018

    2018

  35. [43]

    AI’s Superhuman Persuasion | Psychology Today

    John Nosta. AI’s Superhuman Persuasion | Psychology Today. https://www.psychologytoday.com/intl/blog/the-digital-self/202310/ais-superhuman-persuasion

  36. [44]

    Training language models to follow instructions with human feedback.Advances in Neural Information Processing Systems, 35:27730–27744, 2022

    Long Ouyang, Jeffrey Wu, Xu Jiang, Diogo Almeida, Carroll Wainwright, Pamela Mishkin, Chong Zhang, Sandhini Agarwal, Katarina Slama, Alex Ray, et al. Training language models to follow instructions with human feedback.Advances in Neural Information Processing Systems, 35:27730...

    2022

  37. [45]

    Data quality of platforms and panels for online behavioral research.Behavior Research Methods, 54:1643, 8 2022

    Eyal Peer, David Rothschild, Andrew Gordon, Zak Evernden, and Ekaterina Damer. Data quality of platforms and panels for online behavioral research.Behavior Research Methods, 54:1643, 8 2022

    2022

  38. [46]

    Springer New York, 1986

    Richard E Petty and John T Cacioppo.The Elaboration Likelihood Model of Persuasion, pages 1–24. Springer New York, 1986

    1986

  39. [47]

    Pilz, James Sanders, Robi Rahman, and Lennart Heim

    Konstantin F. Pilz, James Sanders, Robi Rahman, and Lennart Heim. Trends in ai supercomputers, 2025

    2025

  40. [48]

    Post-Broadcast Democracy: How Media Choice Increases Inequality in Political Involvement and Polarizes Elections

    MarkusPrior. Post-Broadcast Democracy: How Media Choice Increases Inequality in Political Involvement and Polarizes Elections. Cambridge University Press, Cambridge, 2007

    2007

  41. [49]

    Persuasion with Large Language Models: A Survey, November 2024

    Alexander Rogiers, Sander Noels, Maarten Buyl, and Tijl De Bie. Persuasion with Large Language Models: A Survey, November 2024

    2024

  42. [50]

    Political compass or spinning arrow? towards more meaningful evaluations for values and opinions in large language models.arXiv, 2 2024

    Paul Röttger, Valentin Hofmann, Valentina Pyatkin, Musashi Hinck, Hannah Rose Kirk, Hinrich Schütze, and Dirk Hovy. Political compass or spinning arrow? towards more meaningful evaluations for values and opinions in large language models.arXiv, 2 2024

    2024

  43. [51]

    On the conversational persuasiveness of large language models: A randomized controlled trial.arXiv, 3 2024

    Francesco Salvi, Manoel Horta Ribeiro, Riccardo Gallotti, and Robert West. On the conversational persuasiveness of large language models: A randomized controlled trial.arXiv, 3 2024. 18

    2024

  44. [52]

    Listen for a change? a longitudinal field experiment on listening’s potential to facilitate persuasion, Sep 2024

    Erik Santoro, David Broockman, Joshua Kalla, and Roni Porat. Listen for a change? a longitudinal field experiment on listening’s potential to facilitate persuasion, Sep 2024

    2024

  45. [53]

    Large language models are more persuasive than incentivized human persuaders.arXiv preprint arXiv:2505.09662, 2025

    Philipp Schoenegger, Francesco Salvi, Jiacheng Liu, Xiaoli Nan, Ramit Debnath, Barbara Fasolo, Evelina Leivada, Gabriel Recchia, Fritz Günther, Ali Zarifhonarvar, et al. Large language models are more persuasive than incentivized human persuaders.arXiv preprint arXiv:2505.09662, 2025

    2025

  46. [54]

    Quantifying language models’ sensitivity to spurious features in prompt design or: How i learned to start worrying about prompt formatting

    Melanie Sclar, Yejin Choi, Yulia Tsvetkov, Alane Suhr, and Paul G Allen. Quantifying language models’ sensitivity to spurious features in prompt design or: How i learned to start worrying about prompt formatting. arXiv, 10 2023

    2023

  47. [55]

    Can AI scaling continue through 2030? Blog post

    Jaime Sevilla, Tamay Besiroglu, Ben Cottier, Josh You, Edu Roldán, Pablo Villalobos, and Ege Erdil. Can AI scaling continue through 2030? Blog post

  48. [56]

    The persuasive effects of political microtargeting in the age of generative artificial intelligence.PNAS Nexus, 3(2):pgae035, February 2024

    Almog Simchon, Matthew Edwards, and Stephan Lewandowsky. The persuasive effects of political microtargeting in the age of generative artificial intelligence.PNAS Nexus, 3(2):pgae035, February 2024

    2024

  49. [57]

    Berinsky, Antonio Alonso Arechar, Robb Willer, and David Rand

    Michael Nicholas Stagnaro, James Druckman, Adam J. Berinsky, Antonio Alonso Arechar, Robb Willer, and David Rand. Representativeness versus response quality: Assessing nine opt-in online survey samples. OSF Preprints, 2 2024

    2024

  50. [58]

    Tappin, Chloe Wittenberg, Luke B

    Ben M. Tappin, Chloe Wittenberg, Luke B. Hewitt, Adam J. Berinsky, and David G. Rand. Quantifying the potential persuasive returns to political microtargeting.Proceedings of the National Academy of Sciences, 120(25):e2216261120, June 2023

    2023

  51. [59]

    Practical Bayesian model evaluation using leave-one-out cross-validation and WAIC.Statistics and Computing, 27(5):1413–1432, September 2017

    Aki Vehtari, Andrew Gelman, and Jonah Gabry. Practical Bayesian model evaluation using leave-one-out cross-validation and WAIC.Statistics and Computing, 27(5):1413–1432, September 2017

    2017

  52. [60]

    Morally reframed arguments can affect support for political candidates

    Jan G Voelkel and Matthew Feinberg. Morally reframed arguments can affect support for political candidates. Social Psychological and Personality Science, 9(8):917–924, 2018

    2018

  53. [61]

    Generative propaganda: Evidence of ai’s impact from a state-backed disinformation campaign.PNAS Nexus, 4(4):pgaf083, 04 2025

    Morgan Wack, Carl Ehrett, Darren Linvill, and Patrick Warren. Generative propaganda: Evidence of ai’s impact from a state-backed disinformation campaign.PNAS Nexus, 4(4):pgaf083, 04 2025

    2025

  54. [62]

    On the robustness of chatgpt: An adversarial and out-of-distribution perspective.arXiv, 2 2023

    Jindong Wang, Xixu Hu, Wenxin Hou, Hao Chen, Runkai Zheng, Yidong Wang, Linyi Yang, Wei Ye, Haojun Huang, Xiubo Geng, Binxing Jiao, Yue Zhang, and Xing Xie. On the robustness of chatgpt: An adversarial and out-of-distribution perspective.arXiv, 2 2023

    2023

  55. [63]

    Chain-of-thought prompting elicits reasoning in large language models.Advances in Neural Information Processing Systems, 35:24824–24837, 2022

    Jason Wei, Xuezhi Wang, Dale Schuurmans, Maarten Bosma, Fei Xia, Ed Chi, Quoc V Le, Denny Zhou, et al. Chain-of-thought prompting elicits reasoning in large language models.Advances in Neural Information Processing Systems, 35:24824–24837, 2022. 19

    2022

Pith tools

Reviewed August 6, 2026 · model on record in the stance chip above.