# Political censorship in large language models originating from China

## TL;DR

This study finds that large language models from China show more political censorship compared to those from other countries, likely due to government regulation.

## Contribution

The paper introduces evidence that government censorship influences the behavior of large language models in politically sensitive contexts.

## Key findings

- China-originating models show higher refusal rates and shorter, less accurate responses to political questions.
- Disparities in responses decrease for less-sensitive prompts, indicating regulatory influence.
- Chinese-language prompts result in higher refusal rates, but differences are smaller than between China and non-China models.

## Abstract

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.