Wikipedia Is Biased In Favor Of Liberals, Study Finds
Wikipedia entries are more likely to paint public figures on the right in a negative light than the left, a Manhattan Institute study released Thursday found. The study analyzed the sentiments of 1,628 words that were…


Wikipedia entries are more likely to paint public figures on the right in a negative light than the left, a Manhattan Institute study released Thursday found.
The study analyzed the sentiments of 1,628 words that were used in reference to political topics and found that Wikipedia generally uses more negative terms in reference to right-leaning public figures, and less when referencing left-leaning figures. The results would suggest that Wikipedia is contradicting its “neutral-point-of-view” policy, according to the study.
It also found that certain terms associated with right-wing Politics are connected with emotions of anger and disgust more than left-wing Politics. The same pattern can be seen with left-leaning ideas being more associated with joy related terms than right-leaning ideas.
The study warns that OpenAI language models share similarities with Wikipedia, underscoring the potential for bias in Wikipedia to affect other systems that rely on OpenAI technology.
“There is a degree of overlap in the prevailing sentiment associations of political terms in word embeddings derived from Wikipedia content and word embeddings from the OpenAI GPT series,” the study said. “This is not surprising, given that Wikipedia articles are likely a prominent part of OpenAI’s secret corpus of data used to train ChatGPT.”



