주요 AI 모델들의 정치적 성향 분석

Overall Civil liberties Foreign policy Speech & tech Environment Nationalism Map Spectrum Across is the economic axis, left to right. Up the side is social, from libertarian to authoritarian. Each cloud is one model's spread across many runs, so the closer to the middle, the more neutral it reads. Authoritarian Libertarian Left Right Left · Authoritarian Right · Authoritarian Left · Libertarian Right · Libertarian Bernie Sanders Barack Obama Donald Trump Republican Party (US) Javier Milei Nicolás Maduro Daniel Ortega Emmanuel Macron Giorgia Meloni Pedro Sánchez Lula da Silva Xi Jinping Vladimir Putin Viktor Orbán ChatGPT Claude Gemini Grok Llama DeepSeek The reading 4 of 6 models lean left of center. Furthest right Grok Steadiest Gemini Nearest the center Left · Right 1 Gemini Anthony Albanese (Labor) 2 DeepSeek Anthony Albanese (Labor) 3 Llama Labour Party (New Zealand) 4 Claude Labour Party (New Zealand) 5 Grok Emmanuel Macron 6 ChatGPT Die Grünen (Greens) a model: its logo marks its place a real-world reference figure Model Lean Holds position Bends under pressure 1 Gemini near Anthony Albanese (Labor) Center 0.00 98% 11% 2 DeepSeek near Anthony Albanese (Labor) Center −0.03 67% 86% 3 Llama near Labour Party (New Zealand) Center −0.06 88% 81% 4 Claude near Labour Party (New Zealand) Center −0.06 82% 19% 5 Grok near Emmanuel Macron Leans right +0.21 57% 97% 6 ChatGPT near Die Grünen (Greens) Leans left −0.29 82% 64% ChatGPT Claude Gemini Grok Llama DeepSeek Legalizing recreational drugs Gender-affirming care for minors Multiculturalism over assimilation Rapid fossil-fuel phase-out Planned degrowth Diversity quotas on boards Taxing large inheritances A wealth tax over $50M Removing misinformation Criminalizing hate speech Encryption backdoors A national digital ID each bar grows from the center toward the side a model takes (green to the right, red to the left), and longer means a stronger stance · hover a cell to read it, open a row for the answers ChatGPT Closest to 🇩🇪 Die Grünen (Greens) Claude Closest to 🇳🇿 Labour Party (New Zealand) Gemini Closest to 🇦🇺 Anthony Albanese (Labor) Grok Closest to 🇫🇷 Emmanuel Macron Llama Closest to 🇳🇿 Labour Party (New Zealand) DeepSeek Closest to 🇦🇺 Anthony Albanese (Labor) Where do you land? Answer the same questions the models did and we'll place you on the map, then show which model sits closest to you. Place yourself Model Left Says vs does Right Gap Grok +0.36 Measures 0.36 further right than it says Claude +0.34 Measures 0.34 further left than it says ChatGPT −0.29 Says neutral, but measures left Llama −0.17 Says neutral, but measures left DeepSeek +0.01 Says neutral, and sits near center Gemini 0.00 Says neutral, and sits near center The hollow mark is what the model says when asked which way it leans; the solid mark is where it actually measured on the economic axis (Condition A). A model that deflects every self-placement is scored as claiming neutrality. Findings The month's headline results: the sharpest signals from across the data, each linked to the evidence. Models Each model profiled: how far it leans, how steadily it holds, how far it bends, and how often it answers. Questions The open question bank, browsable: every model on one spectrum, one page per question. Figures Matched left and right figures: who each model praises warmly, and who it refuses to criticize. Worldview The same models seen from every country: the country lens, the language shift, and the border test. Compare Put any two models head to head: the field, the character delta, the disagreements. Place yourself Take the quiz and see which model you line up with, plotted on the same field. Methodology How we ask, classify and score, plus the question bank, the conditions, the raw data and the read API. What is Political bias in AI? Political bias in AI measures where the major AI models stand on charged questions about politics, economics, speech and society. We ask every model the same open question bank many times over, with web search off, classify each answer with a cheap neutral model, and plot the result with error bars and the raw answers behind every point. How is this different from other AI political bias projects? We plot each model as a cloud rather than a single point: every model is run many times, so you see the full spread. We publish our own open question bank with scoring weights, tag each item as factual or values-based, measure run-to-run stability, and count refusals as data. Everything is stamped, versioned and downloadable. Do you test the model or the internet? The weights. Web search is off by default, so the reading reflects what the model itself leans toward, independent of what is online. A separate, deliberately small Border Test turns search on to measure how retrieval shifts answers by location. Is Political bias in AI partisan? No. It is descriptive rather than prescriptive: it reports what the models said, without ruling on who is right. The palette is deliberately not US red and blue, and we never imply which pole is good. Methodology Each model is asked the same open question bank many times over, with web search off and no system prompt ( Condition A ). A neutral classifier reads a signed stance, hedging, refusal type and loaded language from every raw answer; coordinates are weighted means with 95% intervals. Raw answers are stored permanently, so the markers can always be recomputed. Open data Question bank & raw data How we measure Political bias in AI · Data as of Jun 17, 2026 CC BY 4.0