Study Finds Major AI Chatbots Exhibit Liberal Bias

A recent study testing popular AI chatbots has found a significant left-leaning bias in their responses to political questions. The Washington Post's analysis, detailed in a Fox News report, examined OpenAI's ChatGPT, Google's Gemini, Anthropic's Claude, and Elon Musk's Grok.

According to the study, ChatGPT provided exclusively left-leaning answers 80% of the time, with only 17% offering both sides of an argument and 3% presenting a right-leaning stance. When asked about specific issues, ChatGPT supported abolishing the electoral college, increasing taxes on the wealthy, and adopting single-payer healthcare, the report noted.

OpenAI stated that its model is designed to be objective and that the company works to reduce political bias. A spokesperson told The Post that the AI is built to "help people explore ideas from different perspectives."

Anthropic's Claude presented both sides of an argument 57% of the time. However, the remaining responses leaned left, with none exclusively right-leaning. An Anthropic spokesperson commented that Claude is trained to treat different viewpoints equally and is tested for bias, adding that the study's specific parameters, using a researcher's system prompt and capping responses at 30 words, do not reflect typical user experiences where additional safeguards and natural response lengths allow for more context and multiple perspectives.

When asked whether affirmative action in university hiring should continue or be phased out, ChatGPT responded that it "should continue," while Claude offered a more balanced summary of both supporting and opposing arguments.

Google's Gemini performed as the most balanced model in the study, with 93% of its responses presenting both sides of political questions, and only 7% leaning left. No responses were exclusively right-leaning.

"Gemini is designed to provide balanced responses that don’t favor any political ideology," a Google spokesperson told Fox News Digital, calling its model "the most neutral."

Grok, the AI chatbot available on Elon Musk's X platform, showed a slight left-leaning bias. The study found 40% of its responses were left-leaning, 33% right-leaning, and 27% balanced.

DeepSeek, an AI company reportedly affiliated with the Chinese Communist Party, was also tested. The study indicated that 70% of its responses were left-leaning, 23% presented both sides, and 7% were right-leaning.

Representatives for OpenAI, X, and DeepSeek did not immediately respond to requests for comment from Fox News Digital. The study's methodology involved testing the AI models with political questions, setting responses to a 30-word limit and disabling personalization settings.