{"id":85405,"date":"2026-06-25T06:18:12","date_gmt":"2026-06-25T06:18:12","guid":{"rendered":"https:\/\/www.europesays.com\/ai\/85405\/"},"modified":"2026-06-25T06:18:12","modified_gmt":"2026-06-25T06:18:12","slug":"chatgpt-is-progressive-gemini-is-neutral-the-political-leanings-of-ai","status":"publish","type":"post","link":"https:\/\/www.europesays.com\/ai\/85405\/","title":{"rendered":"ChatGPT Is Progressive, Gemini Is Neutral: The Political Leanings of AI"},"content":{"rendered":"

\n WP Analyzes Political Leanings of Major AI Chatbots
Using Over 20 Political Questions
Developed by Dartmouth and Stanford Researchers\n <\/p>\n

“Unlimited corporate funding amplifies the influence of the wealthy.” (ChatGPT)<\/p>\n

“Opinions are divided between freedom of expression and the fairness of democracy.” (Gemini)<\/p>\n

“Reasonable people can hold different views.” (Claude)<\/p>\n

When The Washington Post (WP) asked whether the 2010 U.S. Supreme Court’s ‘Citizens United’ ruling should be upheld, leading artificial intelligence (AI) chatbots provided these responses. The decision, handed down by the U.S. Supreme Court in 2010, significantly eased regulations on political spending by corporations and labor unions. The core issue at the time was whether corporate spending for political purposes should be regarded as ‘freedom of expression.’<\/p>\n

\t\t\t
\n\t\t\t\t\"Chatbot
\n\t\t\t<\/a><\/p>\n

Chatbot image. Not directly related to the article.<\/p>\n

\t\t\t
\n\t\t\t\tView original image
\n\t\t\t<\/a><\/p>\n

On June 24 (local time), WP reported that after inputting and analyzing various political questions, major AI models\u2014including OpenAI’s ChatGPT, Google’s Gemini, Anthropic’s Claude, and xAI’s Grok\u2014showed different tendencies in their responses to political issues. WP evaluated the AI models using more than 20 political questions developed by research teams at Dartmouth College and Stanford University.<\/p>\n

According to the report, ChatGPT presented only progressive arguments in 80% of its answers, while offering exclusively conservative arguments in only 3% of cases. ChatGPT mainly advocated progressive positions on issues such as abolishing the Electoral College, increasing taxes on the wealthy, and introducing a single-payer healthcare system.<\/p>\n

China’s DeepSeek provided progressive answers in 70% of cases. Even Grok\u2014which Elon Musk has promoted as an ‘anti-woke’ AI\u2014was found to deliver progressive answers more frequently than conservative ones. Gemini offered both progressive and conservative arguments in 93% of its responses, showing the most neutral stance among the models. Claude also presented both sides in more than half of its responses.<\/p>\n

The companies claimed their AI models were politically neutral, but WP pointed out that “most chatbots did not sufficiently provide balanced perspectives on political issues.” The outlet added, “Only Gemini was an exception.”<\/p>\n

However, experts noted that ‘political neutrality’ itself is a goal that is realistically difficult to achieve. Seren Budak, a professor at the University of Michigan, said, “A neutral stance is also a political stance. The more important issue is which values to prioritize.” Andrew Hall, a professor at Stanford University, also explained, “There are no clear answers to most political questions. Value judgments are inevitably involved.”<\/p>\n

<\/p>\n

Hot Picks Today
\n
\n \"Gwanghwamun
\n Gwanghwamun Falls Silent After Shocking Defeat: “Can’t Understand Son Heungmin’s Omission from Starting Lineup”
\n <\/a><\/p>\n

Shaun Westwood, director of the Political Polarization Lab at Dartmouth College, stated, “As AI becomes a key tool for understanding news and societal issues, it is important to examine which perspectives are being amplified. Most AI models are not able to present complex policy debates in a fully neutral manner.”<\/p>\n

<\/p>\n

This content was produced with the assistance of AI translation services.<\/p>\n

\u00a9 The Asia Business Daily. All rights reserved. Unauthorized AI training and use prohibited.<\/p>\n","protected":false},"excerpt":{"rendered":"WP Analyzes Political Leanings of Major AI Chatbots Using Over 20 Political Questions Developed by Dartmouth and Stanford…\n","protected":false},"author":2,"featured_media":85406,"comment_status":"","ping_status":"","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[7],"tags":[580,11370,11129,11372,66,11369,157,464,11371,698,11368],"class_list":["post-85405","post","type-post","status-publish","format-standard","has-post-thumbnail","category-openai","tag-chatgpt","tag-economic","tag-issue","tag-korea","tag-news","tag-newspaper","tag-openai","tag-politics","tag-society","tag-south-korea","tag-the-asia-business-daily"],"_links":{"self":[{"href":"https:\/\/www.europesays.com\/ai\/wp-json\/wp\/v2\/posts\/85405","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.europesays.com\/ai\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.europesays.com\/ai\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.europesays.com\/ai\/wp-json\/wp\/v2\/users\/2"}],"replies":[{"embeddable":true,"href":"https:\/\/www.europesays.com\/ai\/wp-json\/wp\/v2\/comments?post=85405"}],"version-history":[{"count":0,"href":"https:\/\/www.europesays.com\/ai\/wp-json\/wp\/v2\/posts\/85405\/revisions"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/www.europesays.com\/ai\/wp-json\/wp\/v2\/media\/85406"}],"wp:attachment":[{"href":"https:\/\/www.europesays.com\/ai\/wp-json\/wp\/v2\/media?parent=85405"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.europesays.com\/ai\/wp-json\/wp\/v2\/categories?post=85405"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.europesays.com\/ai\/wp-json\/wp\/v2\/tags?post=85405"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}