The levers of political persuasion with conversational artificial intelligence

There are widespread fears that conversational artificial intelligence (AI) could soon exert unprecedented influence over human beliefs. In this work, in three large-scale experiments (<i>N</i> = 76,977 participants), we deployed 19 large language models (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. 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, which had smaller effects. We further show that these methods increased persuasion by exploiting LLMs' ability to rapidly access and strategically deploy information and that, notably, where they increased AI persuasiveness, they also systematically decreased factual accuracy.

Authors

Institutions

Publication Details

Journal
Science
Published
2025-12-04
DOI
https://doi.org/10.1126/science.aea3884
Citations
10
Primary Topic
Misinformation and Its Impacts
Type
article
Field-Weighted Citation Impact
42.02

Funders

Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
article

The levers of political persuasion with conversational artificial intelligence

Ben M Tappin, Hause Lin, Christopher Summerfield, Helen Margetts et al.
10 citations
Science
Misinformation and Its Impacts
42.02
article

The levers of political persuasion with conversational artificial intelligence

Ben M Tappin, Hause Lin, Christopher Summerfield, Helen Margetts, Luke Hewitt, David G. Rand, Kobi Hackenburg, E. Dale Saunders, Stewart Black, Catherine Fist
article en
10 citations

Abstract

There are widespread fears that conversational artificial intelligence (AI) could soon exert unprecedented influence over human beliefs. In this work, in three large-scale experiments (<i>N</i> = 76,977 participants), we deployed 19 large language models (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. 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, which had smaller effects. We further show that these methods increased persuasion by exploiting LLMs' ability to rapidly access and strategically deploy information and that, notably, where they increased AI persuasiveness, they also systematically decreased factual accuracy.

ScienceVol. 390(6777)
Cornell University (US), University of Oxford (GB), New York State College of Agriculture & Life Sciences (US), Research Institute in Science of Cyber Security (GB), Massachusetts Institute of Technology (US), London School of Economics and Political Science (GB), Stanford University (US)
Government of the United Kingdom, UK Research and Innovation, Leverhulme Trust, University of Bristol, Science and Technology Facilities Council
Openalex Percentile: Top 0%
Misinformation and Its Impacts
42.02
DOIOpen Access PDFOpenAlex
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.