EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Political censorship in large language models originating from China

Jennifer Pan iD, Xu Xu iD

PNAS Nexus · 05.02.2026 · 5 (2)

DOI: 10.1093/pnasnexus/pgag013 ↗

Kurz gefasst

Fachartikel in der Zeitschrift PNAS Nexus, online erschienen am 17.02.2026. Verfasst von Jennifer Pan und 1 weiteren Autor:innen. Beteiligte Einrichtung: Department of Communication, Stanford University, 450 Jane Stanford Way, Stanford, CA 94305, USA. Die Arbeit wurde bisher 9-mal zitiert (Crossref) und nennt 39 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Stanford Center for the Study of China’s Economy and Institutions.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

A growing body of research on large language models (LLMs) has identified various biases, primarily in contexts where biases reflect societal patterns. This article focuses on a different source of bias in LLMs—government censorship. By comparing foundation models developed in China and those from outside China, we find substantially higher rates of refusal to respond, shorter responses, and inaccurate responses to a battery of 145 political questions in China-originating models. These disparities diminish for less-sensitive prompts, showing that technological and market differences cannot fully explain this divergence. While all models exhibit higher refusal to respond rates with Chinese-language prompts than English ones, language differences are less pronounced than disparities between China-originating and non-China-originating models. We caution that our study is observational and cross-sectional and does not establish a causal linkage between regulatory pressures and censorship behaviors of China-originating LLMs, but these results suggest that censorship through government regulation requiring companies to restrict political content may be an important factor contributing to political bias in LLMs.

Schlagwörter

Stichwörter der Autor:innen

AIChinaLLMcensorshippolitical bias

Autor:innen und Institutionen

  • Department of Communication, Stanford University, 450 Jane Stanford Way, Stanford, CA 94305, USA.
  • Department of Politics & School of Public and International Affairs, Princeton University, 403 Robertson Hall, Princeton, NJ 08544, USA.

ORCID-Profile: Jennifer Pan, Xu Xu

Förderung

  • Stanford Center for the Study of China’s Economy and Institutions

Zitieren

APA

Pan, J., & Xu, X. (2026). Political censorship in large language models originating from China. PNAS Nexus, 5(2). https://doi.org/10.1093/pnasnexus/pgag013

Vancouver

Pan J, Xu X. Political censorship in large language models originating from China. PNAS Nexus. 2026;5(2). doi:10.1093/pnasnexus/pgag013

BibTeX
@article{Pan2026,
  author = {Pan, Jennifer and Xu, Xu},
  title = {Political censorship in large language models originating from China},
  journal = {PNAS Nexus},
  year = {2026},
  volume = {5},
  number = {2},
  doi = {10.1093/pnasnexus/pgag013}
}
Kostenlos starten →