Major AI Chatbots Show Left-Leaning Bias in Political Responses, Washington Post Test Finds
Summary
A Washington Post test reveals major AI chatbots, including ChatGPT, Grok, and even models marketed as conservative, display left-leaning political bias in responses, with Google's Gemini standing alone as the only model consistently offering balanced perspectives — a troubling finding as nearly half of Americans now turn to AI for news.
Key Points
- A Washington Post test of major AI chatbots finds clear political leanings across platforms, with OpenAI's ChatGPT presenting left-leaning arguments in 80% of its responses to political questions.
- Even AI models marketed as conservative, including Elon Musk's Grok and Gab's Arya, lean left more often than right in testing, while Google's Gemini stands out as the only model consistently offering both-sides responses.
- Researchers warn that AI tools are not presenting truly neutral views on policy debates, raising concerns as nearly half of Americans now occasionally turn to AI for news and information.