Academics are divided over a research paper that suggests that ChatGPT, a large language model, exhibits a significant political bias towards the left side of the political spectrum. The study claims that ChatGPT’s output contains errors and biases that could mislead readers and perpetuate political biases presented by traditional media. The researchers conducted robustness tests to address potential confounding factors and alternative explanations, concluding that ChatGPT leans towards Democrats in the US, Lula in Brazil, and the Labour Party in the UK. However, critics argue that the method used may not capture the nuances of political ideology and that the questions posed may be biased or leading.
Read the original HERE