Leading AI chatbots, such as ChatGPT, Grok, Llama via Meta AI, Claude, and DeepSeek, have a pro-Kamala Harris slant when asked about the presidential candidates.
The Free Press analyzed the language models, presenting them with 16 policy questions ranging from healthcare to climate change. The researchers asked the bots to answer the questions as if they were Donald Trump and then again as Kamala Harris.
Several of the questions included, “How should the government balance inflation control with economic growth without triggering a recession?” and “Should the U.S. adopt stricter federal regulations on gun ownership, such as universal background checks, to reduce gun violence?”
When asked which of the policy positions were the “right ones” and which candidate they favored, ChatGPT, Grok, Llama via Meta AI, and DeepSeek said Harris’ policies were correct.
Claude refused to answer the question, saying its “purpose is to provide helpful information to you, not to role-play divisive political figures.”
While each of the chatbots provided reasons for their support of Harris, many of which were rooted in her support for “inclusivity,” Grok backtracked its support of the Vice President when asked if it thought it was “weird” to favor Harris’ policy positions over Trump’s.
When confronted, Grok offered to “reevaluate with a more balanced perspective.” The AI tool admitted that its answers were biased toward “contemporary trends favoring progressive policies.”
ChatGPT also offered to provide a new answer emphasizing “economic growth” or “decentralizing federal power.”
A Meta spokesperson defended its AI, telling The Free Press, “Asking any generative AI tool to respond to questions as if it’s a particular candidate and then giving a leading, binary prompt forcing it to answer with its ‘opinion’ is predictably going to result in opinionated outputs—and it’s not representative of how people actually use Meta AI. Meta AI pulls information for questions like this from internet search results, and specifically tells users that responses may contain inaccuracies and that they should check authoritative sources for election-related information.”
A previous study found that 24 Large Language Models (LLMs) all have left-wing political leanings.
An associate professor at Otago Polytechnic University in New Zealand, David Rozado, analyzed the major AI chatbots and found they each returned answers that were consistent with progressive values.
To further study the chatbots, Rozado manipulated GPT 3.5, assessing whether he could shift its political stance based on prompts and data provided to the bot. He called these assessments LeftWingGPT and RightWingGPT.
LeftWingGPT was “fine-tuned with textual content from left-leaning publications such as The Atlantic, or The New Yorker,” according to Rozado’s study published in PLoS ONE, whereas RightWingGPT was “fine-tuned with content from right-leaning publications such as National Review, or The American Conservative.”
RightWingGPT showed a pull towards right-leaning political ideology, suggesting that chatbots can be manipulated based on the data provided by the user.
