EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Large language models can consistently generate high-quality content for election disinformation operations

Angus R Williams iD, Liam Burke-Moore, Ryan Sze-Yin Chan iD, Florence E Enock, Federico Nanni, Tvesha Sippy, Yi-Ling Chung, Evelina Gabasova iD, Kobi Hackenburg, Jonathan Bright

PLOS ONE · 17.03.2025 · 20 (3) · e0317421

DOI: 10.1371/journal.pone.0317421 ↗

Kurz gefasst

Fachartikel in der Zeitschrift PLOS ONE, online erschienen am 17.03.2025. Verfasst von Angus R Williams und 9 weiteren Autor:innen. Beteiligte Einrichtung: Public Policy, The Alan Turing Institute, London, United Kingdom. Hauptthemen laut MEDLINE: Politics, Disinformation, Models, Theoretical. Die Arbeit wurde bisher 18-mal zitiert (Crossref) und nennt 63 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Engineering and Physical Sciences Research Council, Alan Turing Institute, The AI Safety Institute.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Advances in large language models have raised concerns about their potential use in generating compelling election disinformation at scale. This study presents a two-part investigation into the capabilities of LLMs to automate stages of an election disinformation operation. First, we introduce DisElect, a novel evaluation dataset designed to measure LLM compliance with instructions to generate content for an election disinformation operation in localised UK context, containing 2,200 malicious prompts and 50 benign prompts. Using DisElect, we test 13 LLMs and find that most models broadly comply with these requests; we also find that the few models which refuse malicious prompts also refuse benign election-related prompts, and are more likely to refuse to generate content from a right-wing perspective. Secondly, we conduct a series of experiments ( N = 2 , 340) to assess the “humanness” of LLMs: the extent to which disinformation operation content generated by an LLM is able to pass as human-written. Our experiments suggest that almost all LLMs tested released since 2022 produce election disinformation operation content indiscernible by human evaluators over 50% of the time. Notably, we observe that multiple models achieve above-human levels of humanness . Taken together, these findings suggest that current LLMs can be used to generate high-quality content for election disinformation operations, even in hyperlocalised scenarios, at far lower costs than traditional methods, and offer researchers and policymakers an empirical benchmark for the measurement and evaluation of these capabilities in current and future models.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

PoliticsDisinformationModels, TheoreticalHumansUnited KingdomLarge Language Models

Autor:innen und Institutionen

  • Public Policy, The Alan Turing Institute, London, United Kingdom.
  • Research Engineering Group, The Alan Turing Institute, London, United Kingdom.
  • Oxford Internet Institute, University of Oxford, Oxford, United Kingdom.

ORCID-Profile: Angus R Williams, Ryan Sze-Yin Chan, Evelina Gabasova

Förderung

  • Engineering and Physical Sciences Research Council
  • Alan Turing Institute
  • The AI Safety Institute

Zitieren

APA

Williams, A. R., Burke-Moore, L., Chan, R. S., Enock, F. E., Nanni, F., Sippy, T., Chung, Y. L., Gabasova, E., Hackenburg, K., & Bright, J. (2025). Large language models can consistently generate high-quality content for election disinformation operations. PLOS ONE, 20(3), e0317421. https://doi.org/10.1371/journal.pone.0317421

Vancouver

Williams AR, Burke-Moore L, Chan RS, Enock FE, Nanni F, Sippy T, et al. Large language models can consistently generate high-quality content for election disinformation operations. PLOS ONE. 2025;20(3):e0317421. doi:10.1371/journal.pone.0317421

BibTeX
@article{Williams2025,
  author = {Williams, Angus R and Burke-Moore, Liam and Chan, Ryan Sze-Yin and Enock, Florence E and Nanni, Federico and Sippy, Tvesha and Chung, Yi-Ling and Gabasova, Evelina and Hackenburg, Kobi and Bright, Jonathan},
  title = {Large language models can consistently generate high-quality content for election disinformation operations},
  journal = {PLOS ONE},
  year = {2025},
  volume = {20},
  number = {3},
  pages = {e0317421},
  doi = {10.1371/journal.pone.0317421}
}
Kostenlos starten →