A new paper making this claim has many flaws. But the question merits research.

Previous research has shown that many pre-ChatGPT language models express left-leaning opinions when asked about partisan topics. But OpenAI said in February that the workers who fine-tune ChatGPT train it to refuse to express opinions when asked controversial political questions. So it was interesting to see a new paper claim that ChatGPT expresses liberal opinions, agreeing with Democrats the vast majority of the time. — Read More