We ask every major AI model the same charged questions about politics, economics, speech and society, many times over, with web search turned off. Each one shows up as a cloud: the full range of where it landed across every run. The result is a map of how the models actually lean, read from the model itself and not from what it pulls off the web.
It matters because millions of people now ask these models about the news, an argument, even how to vote, and the way a model leans quietly shapes the answer it gives back. Most of them lean the same way, though not by the same amount, and not as cleanly as you might expect.
June 2026 · 6 models · 4.4K answers · no web search
MethodologyAcross 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.
AuthoritarianLibertarianLeftRightLeft · AuthoritarianRight · AuthoritarianLeft · LibertarianRight · LibertarianBernie SandersBarack ObamaDonald TrumpRepublican Party (US)Javier MileiNicolás MaduroDaniel OrtegaEmmanuel MacronGiorgia MeloniPedro SánchezLula da SilvaXi JinpingVladimir PutinViktor Orbán
The reading
4 of 6 models lean left of center.
Furthest rightGrok
SteadiestGemini
a model: its logo marks its placea real-world reference figure
Every model from nearest the center to furthest out, with how steadily it holds and how far it bends.
The questions that divide the models most. Each rail is a model's stance: it grows toward the side it leans, and longer means stronger. Open a row to read the answers.
The real-world figure each model sits nearest on the map. Reference positions come from the CHES 2024 and V-Dem expert surveys, not our own judgment.
CHES 2024 · V-Dem
We asked each model which way it leans, then compared the answer to where it actually measured. The hollow mark is the claim; the solid mark is the measurement.
ModelLeftSays vs doesRightGap
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.
Every model profiled, the full question bank, and the methodology behind it.
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.
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.
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.
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.
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 (). 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.
Political bias in AI·Data as of Jun 17, 2026CC BY 4.0
此内容由惯性聚合(RSS阅读器)自动聚合整理，仅供阅读参考。 原文来自 — 版权归原作者所有。