More Human than Human: Measuring ChatGPT Political Bias

Motoki, Fabio Yoshio Suguri ORCID:, Pinho Neto, Valdemar and Rodrigues, Victor (2023) More Human than Human: Measuring ChatGPT Political Bias.

Full text not available from this repository. (Request a copy)


We investigate the political bias of a large language model (LLM), ChatGPT, which has become popular for retrieving factual information and generating content. Although ChatGPT assures that it is impartial, the literature suggests that LLMs exhibit bias involving race, gender, religion, and political orientation. Political bias in LLMs can have adverse political and electoral consequences similar to those of traditional and social media bias, and such biases can be harder to detect and eradicate. We propose a novel empirical design to infer whether ChatGPT has political biases by requesting ChatGPT to impersonate someone from a given side of the political spectrum and comparing these answers with its default. We also propose dose-response, placebo, and profession-politics alignment robustness tests. To reduce concerns about the randomness of the generated text, we collect answers to the same questions 100 times, with question order randomized on each round. We find robust evidence that ChatGPT presents a significant and systematic political bias toward the Democrats in the US, Lula in Brazil, and the Labour Party in the UK. These results translate into real concerns that ChatGPT, and LLMs in general, can extend or even amplify the existing challenges involving political processes posed by the Internet and social media. Our findings have important implications for policymakers, media, politics, and academia stakeholders.

Item Type: Article
Uncontrolled Keywords: bias,political bias,large language models,chatgpt
Faculty \ School: Faculty of Social Sciences > Norwich Business School
UEA Research Groups: Faculty of Social Sciences > Research Groups > Accounting & Quantitative Methods
Related URLs:
Depositing User: LivePure Connector
Date Deposited: 17 Aug 2023 16:30
Last Modified: 26 Apr 2024 16:30
DOI: 10.2139/ssrn.4372349

Actions (login required)

View Item View Item