June 20th, 2024 2 Minute Read Press Release

New Study Finds Political Bias Embedded in Wikipedia Articles

With Large Language Models like ChatGPT often using Wikipedia content for their training, this bias is at risk of spreading

NEW YORK, NY — Wikipedia has evolved over the past two decades into an indispensable information resource for millions of users worldwide, amassing over 4 billion visits per month. But its significance extends beyond direct human readership since Wikipedia content is also routinely employed in the training of Large Language Models (LLMs) like ChatGPT. Given Wikipedia’s immense reach and influence, the accuracy and neutrality of its content is of paramount importance. Any biases present in Wikipedia content are at risk of being absorbed into the foundational parameters of contemporary AI systems, potentially perpetuating, and amplifying these biases further.

In a new Manhattan Institute report, David Rozado details findings from his groundbreaking analysis of political bias in Wikipedia content. In his assessment, Rozado computationally assesses the sentiment and emotional tone associated with politically charged terms within Wikipedia articles (i.e., names of recent U.S. presidents, U.S. congressmembers, U.S. Supreme Court justices, or prime ministers of Western countries). Findings show that Wikipedia entries are more likely to attach negative sentiment to terms representative of right-leaning political orientation than to their left-leaning counterparts. Moreover, terms suggestive of right-of-center political stances are more frequently connected with emotions of anger and disgust than those suggestive of a left-of-center stance. Conversely, terms associated with left-leaning ideology are more frequently linked with the emotion of joy than terms associated with right-leaning ideology.

Rozado finds that some of the politically biased sentiment associations embedded in Wikipedia articles also pop up in OpenAI’s language models. This is suggestive of the potential for biases in Wikipedia content percolating into widely used AI systems. Given Wikipedia's significant and valuable role as a public resource, Rozado highlights areas where Wikipedia can improve in how it presents political information by upholding and strengthening Wikipedia's principles of neutrality and impartiality.

Click here to view the report.


Are you interested in supporting the Manhattan Institute’s public-interest research and journalism? As a 501(c)(3) nonprofit, donations in support of MI and its scholars’ work are fully tax-deductible as provided by law (EIN #13-2912529).