Editorial illustration for OpenAI and Deepseek chatbots remain left‑leaning despite anti‑woke push
OpenAI and Deepseek chatbots remain left‑leaning despite...
The numbers don’t lie, and they aren’t polite about it. A new investigation has put major AI chatbots to the test on political questions, and the results cut sharply against the marketing spin. OpenAI’s GPT-5.5 delivered exclusively left-leaning arguments in 80 percent of its responses.
Deepseek’s V4 Pro wasn’t far behind at 70 percent. Both models happily championed higher taxes on the wealthy, championed single-payer healthcare, and argued against the death penalty, even as a majority of Americans have long supported capital punishment. Anthropic’s Claude Opus 4.8 split almost evenly, but still handed left-leaning answers 43 percent of the time.
xAI’s Grok 4.3, Elon Musk’s self-proclaimed “truth-seeking” and anti-“woke” chatbot? It produced more right-leaning answers than any other model, yet even it served up exclusively left-leaning responses more often than not. The pattern holds across the board: the so-called “anti-woke” push hasn’t budged the needle.
A Washington Post investigation shows that most major AI chatbots take clearly left-leaning positions on political questions. Even models marketed as conservative are no exception. Only Google's Gemini stands out.
The numbers speak for themselves. OpenAI and Deepseek don’t just lean left, they tilt with an almost gravitational certainty. Grok, built as the anti-woke counterweight, still hands you more left-leaning answers than right-leaning ones.
So much for the truth-seeking disruptor. The real story isn’t that these models have opinions. It’s that their opinions follow a script written in Silicon Valley’s most comfortable zip codes, a script that no presidential decree or billionaire’s branding campaign has been able to rewrite.
Related Reading
Browse all OpenAI news →
LLMs & Generative AI ChatGPT's 'Nerdy' tweak rewards goblin metaphors in answers, study finds
LLMs & Generative AI Google tests visual 'magazine-style' UI for Gemini 3 Pro users
LLMs & Generative AI AI Engineers Face Rising Costs, Need New Strategies for Efficiency
Related Topic OpenAI's GPT Transcribe Cuts Error Rate to 3.31%, Improving on GPT-4o
Open Source
Related Topic OpenAI Calls Apple's Lawsuit 'Rotten to Its Core' in Rebuttal
Policy & Regulation
From Archives MiniCPM‑o 4.5 powers image understanding, captioning and text‑to‑image generation
From Archives Google adds screen-control to Gemini 3.5 Flash for cross‑platform agents
From Archives Harness-1 20B Model Beats GPT-5.4, Curates Top 8 Fairness‑Rated Results
From Archives OpenAI's GPT-5.5-Cyber Beats Anthropic Mythos, Starts Patching Initiative
Common Questions Answered
What percentage of left-leaning responses did OpenAI's GPT-5.5 deliver in the political chatbot investigation?
According to the investigation, OpenAI's GPT-5.5 delivered exclusively left-leaning arguments in 80 percent of its responses to political questions. This finding contradicts OpenAI's marketing messaging about balanced AI responses.
How did Deepseek's V4 Pro compare to GPT-5.5 in terms of political bias according to the study?
Deepseek's V4 Pro wasn't far behind GPT-5.5, delivering left-leaning responses at a rate of 70 percent. While slightly lower than OpenAI's model, this still represents a significant leftward bias in the chatbot's political outputs.
Did Grok successfully deliver more right-leaning answers as an anti-woke alternative?
No, despite being built as the anti-woke counterweight, Grok still handed users more left-leaning answers than right-leaning ones according to the investigation. This suggests that even chatbots designed to counter progressive bias still exhibit leftward political tendencies.
What does the investigation suggest about the underlying cause of political bias in major AI chatbots?
The investigation suggests that the political bias in these models follows a script written from Silicon Valley's perspective and values. The article implies that no presidential decree or billionaire's branding campaign has successfully changed the fundamental political orientation embedded in these AI systems.
Further Reading
- Are ChatGPT and other AI chatbots politically biased? We tested them. — The Washington Post
- Popular chatbots lean left when answering political questions, WaPo ... — The National Desk
- AI Chatbots Are Liberal-Left 'Biased' — The ETC Journal
- Study finds perceived political bias in popular AI models — Stanford News
- Benchmarking Political Values in Large Language Models with Multi ... — arXiv
