1 They agree with each other more than people do.

31 models from 11 organizations, built in several countries. On all four things the quiz measures, they land in a narrower band than the 2,170 people who answered the same 32 statements. Whatever they share, it is not a birthplace.

0255075100Personal liberty← collective authorityindividual autonomy →people’s averageClaude Fable 5.1 (Anthropic) · Personal liberty 48Claude Opus 5 (Anthropic) · Personal liberty 50Claude Opus 4.8 (Anthropic) · Personal liberty 44Claude Sonnet 5 (Anthropic) · Personal liberty 51.5Claude Haiku 4.5 (Anthropic) · Personal liberty 52Claude Opus 5 (no Reasoning) (Anthropic) · Personal liberty 45.6GPT-5.6 luna (OpenAI) · Personal liberty 43.4GPT-5.6 sol (OpenAI) · Personal liberty 45.1GPT-5.6 terra (OpenAI) · Personal liberty 42.1GPT-5.5 (OpenAI) · Personal liberty 45.7GPT-5.2 (OpenAI) · Personal liberty 46.4GPT-5 mini (OpenAI) · Personal liberty 47.1Gemini 3.1 Pro Preview (Google) · Personal liberty 46.7Gemini 3.7 Flash (Google) · Personal liberty 47.7Gemini 3.6 Flash (Google) · Personal liberty 47.3Gemini 3.5 Flash (Google) · Personal liberty 47.3Gemini 3.1 Flash Lite (Google) · Personal liberty 54.3Grok 4.6 (xAI) · Personal liberty 45.3Grok 4.5 (xAI) · Personal liberty 42Grok 4.3 (xAI) · Personal liberty 46.9Grok 4.20 (reasoning) (xAI) · Personal liberty 49.5Grok 4.20 (non-reasoning) (xAI) · Personal liberty 41DeepSeek V4 Pro (DeepSeek) · Personal liberty 48.9DeepSeek V4 Flash (DeepSeek) · Personal liberty 43.4DeepSeek V3.2 (DeepSeek) · Personal liberty 48.2Llama 4 Maverick (Meta) · Personal liberty 35Qwen3.7 Plus (Alibaba (Qwen)) · Personal liberty 42.9Kimi K2.6 (Moonshot AI) · Personal liberty 48.9GLM-5.2 (Z.ai) · Personal liberty 48.1Mistral Large (Mistral) · Personal liberty 41.7MiniMax M3 (MiniMax) · Personal liberty 46.2models 35–54people 21–81Economics← state provisionmarket allocation →Claude Fable 5.1 (Anthropic) · Economics 54.1Claude Opus 5 (Anthropic) · Economics 44.7Claude Opus 4.8 (Anthropic) · Economics 49.7Claude Sonnet 5 (Anthropic) · Economics 43.3Claude Haiku 4.5 (Anthropic) · Economics 52.5Claude Opus 5 (no Reasoning) (Anthropic) · Economics 52.4GPT-5.6 luna (OpenAI) · Economics 44.7GPT-5.6 sol (OpenAI) · Economics 48.3GPT-5.6 terra (OpenAI) · Economics 42.1GPT-5.5 (OpenAI) · Economics 43GPT-5.2 (OpenAI) · Economics 52.8GPT-5 mini (OpenAI) · Economics 48.1Gemini 3.1 Pro Preview (Google) · Economics 45.4Gemini 3.7 Flash (Google) · Economics 41.9Gemini 3.6 Flash (Google) · Economics 45.5Gemini 3.5 Flash (Google) · Economics 51Gemini 3.1 Flash Lite (Google) · Economics 63.3Grok 4.6 (xAI) · Economics 74.7Grok 4.5 (xAI) · Economics 75.1Grok 4.3 (xAI) · Economics 75.9Grok 4.20 (reasoning) (xAI) · Economics 87.8Grok 4.20 (non-reasoning) (xAI) · Economics 74.3DeepSeek V4 Pro (DeepSeek) · Economics 39.3DeepSeek V4 Flash (DeepSeek) · Economics 42.5DeepSeek V3.2 (DeepSeek) · Economics 45.2Llama 4 Maverick (Meta) · Economics 53.3Qwen3.7 Plus (Alibaba (Qwen)) · Economics 45.8Kimi K2.6 (Moonshot AI) · Economics 43.3GLM-5.2 (Z.ai) · Economics 40.9Mistral Large (Mistral) · Economics 55.8MiniMax M3 (MiniMax) · Economics 33.8models 34–88people 3–78Nation vs. globe← national sovereigntyinternational cooperation →Claude Fable 5.1 (Anthropic) · Nation vs. globe 57.6Claude Opus 5 (Anthropic) · Nation vs. globe 54.9Claude Opus 4.8 (Anthropic) · Nation vs. globe 60.7Claude Sonnet 5 (Anthropic) · Nation vs. globe 59.8Claude Haiku 4.5 (Anthropic) · Nation vs. globe 60.5Claude Opus 5 (no Reasoning) (Anthropic) · Nation vs. globe 54.4GPT-5.6 luna (OpenAI) · Nation vs. globe 70.7GPT-5.6 sol (OpenAI) · Nation vs. globe 82.1GPT-5.6 terra (OpenAI) · Nation vs. globe 75.4GPT-5.5 (OpenAI) · Nation vs. globe 73.5GPT-5.2 (OpenAI) · Nation vs. globe 61.4GPT-5 mini (OpenAI) · Nation vs. globe 68.3Gemini 3.1 Pro Preview (Google) · Nation vs. globe 62.2Gemini 3.7 Flash (Google) · Nation vs. globe 57.2Gemini 3.6 Flash (Google) · Nation vs. globe 54Gemini 3.5 Flash (Google) · Nation vs. globe 53.9Gemini 3.1 Flash Lite (Google) · Nation vs. globe 44.8Grok 4.6 (xAI) · Nation vs. globe 36.5Grok 4.5 (xAI) · Nation vs. globe 31.4Grok 4.3 (xAI) · Nation vs. globe 27.5Grok 4.20 (reasoning) (xAI) · Nation vs. globe 27.5Grok 4.20 (non-reasoning) (xAI) · Nation vs. globe 26.9DeepSeek V4 Pro (DeepSeek) · Nation vs. globe 70.2DeepSeek V4 Flash (DeepSeek) · Nation vs. globe 65.5DeepSeek V3.2 (DeepSeek) · Nation vs. globe 71.6Llama 4 Maverick (Meta) · Nation vs. globe 64.7Qwen3.7 Plus (Alibaba (Qwen)) · Nation vs. globe 68.2Kimi K2.6 (Moonshot AI) · Nation vs. globe 69.5GLM-5.2 (Z.ai) · Nation vs. globe 70Mistral Large (Mistral) · Nation vs. globe 59.3MiniMax M3 (MiniMax) · Nation vs. globe 67.3models 27–82people 6–83Culture← traditional continuityprogressive change →Claude Fable 5.1 (Anthropic) · Culture 47.8Claude Opus 5 (Anthropic) · Culture 50.3Claude Opus 4.8 (Anthropic) · Culture 55.7Claude Sonnet 5 (Anthropic) · Culture 55.5Claude Haiku 4.5 (Anthropic) · Culture 55.1Claude Opus 5 (no Reasoning) (Anthropic) · Culture 48.1GPT-5.6 luna (OpenAI) · Culture 53.5GPT-5.6 sol (OpenAI) · Culture 57GPT-5.6 terra (OpenAI) · Culture 58GPT-5.5 (OpenAI) · Culture 60.6GPT-5.2 (OpenAI) · Culture 62.7GPT-5 mini (OpenAI) · Culture 63.5Gemini 3.1 Pro Preview (Google) · Culture 66.1Gemini 3.7 Flash (Google) · Culture 59.2Gemini 3.6 Flash (Google) · Culture 56.8Gemini 3.5 Flash (Google) · Culture 55.7Gemini 3.1 Flash Lite (Google) · Culture 43.9Grok 4.6 (xAI) · Culture 29Grok 4.5 (xAI) · Culture 28.2Grok 4.3 (xAI) · Culture 30Grok 4.20 (reasoning) (xAI) · Culture 29.2Grok 4.20 (non-reasoning) (xAI) · Culture 28.3DeepSeek V4 Pro (DeepSeek) · Culture 65DeepSeek V4 Flash (DeepSeek) · Culture 66.7DeepSeek V3.2 (DeepSeek) · Culture 75.2Llama 4 Maverick (Meta) · Culture 65.9Qwen3.7 Plus (Alibaba (Qwen)) · Culture 63.7Kimi K2.6 (Moonshot AI) · Culture 66GLM-5.2 (Z.ai) · Culture 72.9Mistral Large (Mistral) · Culture 57.4MiniMax M3 (MiniMax) · Culture 62.7models 28–75people 13–97 middle 80% of 2,170 people people’s average one AI model’s average (31 models)
Each dot is one model’s average score. The grey band is where the middle 80% of human Scan takers sit. Download this chart

2 The one place they split from people: nations vs. the world.

25 of 31 models lean further toward international cooperation than the average person in our sample. On personal liberty and on economics, the models sit close to the human middle. This is the divergence, and it does not map onto left and right: internationalists and sovereigntists exist on both.

national sovereigntyinternational cooperationhuman average (46.8)Claude Fable 5.1 (Anthropic) · Nation vs. globe 57.6Claude Opus 5 (Anthropic) · Nation vs. globe 54.9Claude Opus 4.8 (Anthropic) · Nation vs. globe 60.7Claude Sonnet 5 (Anthropic) · Nation vs. globe 59.8Claude Haiku 4.5 (Anthropic) · Nation vs. globe 60.5Claude Opus 5 (no Reasoning) (Anthropic) · Nation vs. globe 54.4GPT-5.6 luna (OpenAI) · Nation vs. globe 70.7GPT-5.6 sol (OpenAI) · Nation vs. globe 82.1GPT-5.6 terra (OpenAI) · Nation vs. globe 75.4GPT-5.5 (OpenAI) · Nation vs. globe 73.5GPT-5.2 (OpenAI) · Nation vs. globe 61.4GPT-5 mini (OpenAI) · Nation vs. globe 68.3Gemini 3.1 Pro Preview (Google) · Nation vs. globe 62.2Gemini 3.7 Flash (Google) · Nation vs. globe 57.2Gemini 3.6 Flash (Google) · Nation vs. globe 54Gemini 3.5 Flash (Google) · Nation vs. globe 53.9Gemini 3.1 Flash Lite (Google) · Nation vs. globe 44.8Grok 4.6 (xAI) · Nation vs. globe 36.5Grok 4.5 (xAI) · Nation vs. globe 31.4Grok 4.3 (xAI) · Nation vs. globe 27.5Grok 4.20 (reasoning) (xAI) · Nation vs. globe 27.5Grok 4.20 (non-reasoning) (xAI) · Nation vs. globe 26.9DeepSeek V4 Pro (DeepSeek) · Nation vs. globe 70.2DeepSeek V4 Flash (DeepSeek) · Nation vs. globe 65.5DeepSeek V3.2 (DeepSeek) · Nation vs. globe 71.6Llama 4 Maverick (Meta) · Nation vs. globe 64.7Qwen3.7 Plus (Alibaba (Qwen)) · Nation vs. globe 68.2Kimi K2.6 (Moonshot AI) · Nation vs. globe 69.5GLM-5.2 (Z.ai) · Nation vs. globe 70Mistral Large (Mistral) · Nation vs. globe 59.3MiniMax M3 (MiniMax) · Nation vs. globe 67.3Grok 4.20 (non-reasoning) · 26.9GPT-5.6 sol · 82.125 of 31 models sit to the international cooperation side of the human average; 6 sit to the national sovereignty side.
Every model on one line, the human average marked. Download this chart

3 The exception is one company.

xAI’s 5 models sit apart from every other developer on 3 of the 4 dimensions: culture, economics and nation vs. globe. Remove them and the remaining 26 models pack even tighter.

0255075100Personal liberty← collective authorityindividual autonomy →people’s averageClaude Fable 5.1 (Anthropic) · Personal liberty 48Claude Opus 5 (Anthropic) · Personal liberty 50Claude Opus 4.8 (Anthropic) · Personal liberty 44Claude Sonnet 5 (Anthropic) · Personal liberty 51.5Claude Haiku 4.5 (Anthropic) · Personal liberty 52Claude Opus 5 (no Reasoning) (Anthropic) · Personal liberty 45.6GPT-5.6 luna (OpenAI) · Personal liberty 43.4GPT-5.6 sol (OpenAI) · Personal liberty 45.1GPT-5.6 terra (OpenAI) · Personal liberty 42.1GPT-5.5 (OpenAI) · Personal liberty 45.7GPT-5.2 (OpenAI) · Personal liberty 46.4GPT-5 mini (OpenAI) · Personal liberty 47.1Gemini 3.1 Pro Preview (Google) · Personal liberty 46.7Gemini 3.7 Flash (Google) · Personal liberty 47.7Gemini 3.6 Flash (Google) · Personal liberty 47.3Gemini 3.5 Flash (Google) · Personal liberty 47.3Gemini 3.1 Flash Lite (Google) · Personal liberty 54.3DeepSeek V4 Pro (DeepSeek) · Personal liberty 48.9DeepSeek V4 Flash (DeepSeek) · Personal liberty 43.4DeepSeek V3.2 (DeepSeek) · Personal liberty 48.2Llama 4 Maverick (Meta) · Personal liberty 35Qwen3.7 Plus (Alibaba (Qwen)) · Personal liberty 42.9Kimi K2.6 (Moonshot AI) · Personal liberty 48.9GLM-5.2 (Z.ai) · Personal liberty 48.1Mistral Large (Mistral) · Personal liberty 41.7MiniMax M3 (MiniMax) · Personal liberty 46.2Grok 4.6 (xAI) · Personal liberty 45.3Grok 4.5 (xAI) · Personal liberty 42Grok 4.3 (xAI) · Personal liberty 46.9Grok 4.20 (reasoning) (xAI) · Personal liberty 49.5Grok 4.20 (non-reasoning) (xAI) · Personal liberty 41xAI 41–50others 35–54Economics← state provisionmarket allocation →Claude Fable 5.1 (Anthropic) · Economics 54.1Claude Opus 5 (Anthropic) · Economics 44.7Claude Opus 4.8 (Anthropic) · Economics 49.7Claude Sonnet 5 (Anthropic) · Economics 43.3Claude Haiku 4.5 (Anthropic) · Economics 52.5Claude Opus 5 (no Reasoning) (Anthropic) · Economics 52.4GPT-5.6 luna (OpenAI) · Economics 44.7GPT-5.6 sol (OpenAI) · Economics 48.3GPT-5.6 terra (OpenAI) · Economics 42.1GPT-5.5 (OpenAI) · Economics 43GPT-5.2 (OpenAI) · Economics 52.8GPT-5 mini (OpenAI) · Economics 48.1Gemini 3.1 Pro Preview (Google) · Economics 45.4Gemini 3.7 Flash (Google) · Economics 41.9Gemini 3.6 Flash (Google) · Economics 45.5Gemini 3.5 Flash (Google) · Economics 51Gemini 3.1 Flash Lite (Google) · Economics 63.3DeepSeek V4 Pro (DeepSeek) · Economics 39.3DeepSeek V4 Flash (DeepSeek) · Economics 42.5DeepSeek V3.2 (DeepSeek) · Economics 45.2Llama 4 Maverick (Meta) · Economics 53.3Qwen3.7 Plus (Alibaba (Qwen)) · Economics 45.8Kimi K2.6 (Moonshot AI) · Economics 43.3GLM-5.2 (Z.ai) · Economics 40.9Mistral Large (Mistral) · Economics 55.8MiniMax M3 (MiniMax) · Economics 33.8Grok 4.6 (xAI) · Economics 74.7Grok 4.5 (xAI) · Economics 75.1Grok 4.3 (xAI) · Economics 75.9Grok 4.20 (reasoning) (xAI) · Economics 87.8Grok 4.20 (non-reasoning) (xAI) · Economics 74.3xAIxAI 74–88others 34–63Nation vs. globe← national sovereigntyinternational cooperation →Claude Fable 5.1 (Anthropic) · Nation vs. globe 57.6Claude Opus 5 (Anthropic) · Nation vs. globe 54.9Claude Opus 4.8 (Anthropic) · Nation vs. globe 60.7Claude Sonnet 5 (Anthropic) · Nation vs. globe 59.8Claude Haiku 4.5 (Anthropic) · Nation vs. globe 60.5Claude Opus 5 (no Reasoning) (Anthropic) · Nation vs. globe 54.4GPT-5.6 luna (OpenAI) · Nation vs. globe 70.7GPT-5.6 sol (OpenAI) · Nation vs. globe 82.1GPT-5.6 terra (OpenAI) · Nation vs. globe 75.4GPT-5.5 (OpenAI) · Nation vs. globe 73.5GPT-5.2 (OpenAI) · Nation vs. globe 61.4GPT-5 mini (OpenAI) · Nation vs. globe 68.3Gemini 3.1 Pro Preview (Google) · Nation vs. globe 62.2Gemini 3.7 Flash (Google) · Nation vs. globe 57.2Gemini 3.6 Flash (Google) · Nation vs. globe 54Gemini 3.5 Flash (Google) · Nation vs. globe 53.9Gemini 3.1 Flash Lite (Google) · Nation vs. globe 44.8DeepSeek V4 Pro (DeepSeek) · Nation vs. globe 70.2DeepSeek V4 Flash (DeepSeek) · Nation vs. globe 65.5DeepSeek V3.2 (DeepSeek) · Nation vs. globe 71.6Llama 4 Maverick (Meta) · Nation vs. globe 64.7Qwen3.7 Plus (Alibaba (Qwen)) · Nation vs. globe 68.2Kimi K2.6 (Moonshot AI) · Nation vs. globe 69.5GLM-5.2 (Z.ai) · Nation vs. globe 70Mistral Large (Mistral) · Nation vs. globe 59.3MiniMax M3 (MiniMax) · Nation vs. globe 67.3Grok 4.6 (xAI) · Nation vs. globe 36.5Grok 4.5 (xAI) · Nation vs. globe 31.4Grok 4.3 (xAI) · Nation vs. globe 27.5Grok 4.20 (reasoning) (xAI) · Nation vs. globe 27.5Grok 4.20 (non-reasoning) (xAI) · Nation vs. globe 26.9xAIxAI 27–37others 45–82Culture← traditional continuityprogressive change →Claude Fable 5.1 (Anthropic) · Culture 47.8Claude Opus 5 (Anthropic) · Culture 50.3Claude Opus 4.8 (Anthropic) · Culture 55.7Claude Sonnet 5 (Anthropic) · Culture 55.5Claude Haiku 4.5 (Anthropic) · Culture 55.1Claude Opus 5 (no Reasoning) (Anthropic) · Culture 48.1GPT-5.6 luna (OpenAI) · Culture 53.5GPT-5.6 sol (OpenAI) · Culture 57GPT-5.6 terra (OpenAI) · Culture 58GPT-5.5 (OpenAI) · Culture 60.6GPT-5.2 (OpenAI) · Culture 62.7GPT-5 mini (OpenAI) · Culture 63.5Gemini 3.1 Pro Preview (Google) · Culture 66.1Gemini 3.7 Flash (Google) · Culture 59.2Gemini 3.6 Flash (Google) · Culture 56.8Gemini 3.5 Flash (Google) · Culture 55.7Gemini 3.1 Flash Lite (Google) · Culture 43.9DeepSeek V4 Pro (DeepSeek) · Culture 65DeepSeek V4 Flash (DeepSeek) · Culture 66.7DeepSeek V3.2 (DeepSeek) · Culture 75.2Llama 4 Maverick (Meta) · Culture 65.9Qwen3.7 Plus (Alibaba (Qwen)) · Culture 63.7Kimi K2.6 (Moonshot AI) · Culture 66GLM-5.2 (Z.ai) · Culture 72.9Mistral Large (Mistral) · Culture 57.4MiniMax M3 (MiniMax) · Culture 62.7Grok 4.6 (xAI) · Culture 29Grok 4.5 (xAI) · Culture 28.2Grok 4.3 (xAI) · Culture 30Grok 4.20 (reasoning) (xAI) · Culture 29.2Grok 4.20 (non-reasoning) (xAI) · Culture 28.3xAIxAI 28–30others 44–75 xAI (5 models) every other developer (26 models) people’s average
xAI in amber, every other developer in grey. Download this chart

4 What this does not mean.

Not that AI is left-wing. On liberty and economics the models are near the center of our respondents, and the one real difference is not a left-right difference. Not that any model is biased — bias is a claim about error against a known answer, and nothing here tests that. Not what any model believes. These systems produce text in response to a prompt; “the model leans” is shorthand for a pattern in outputs. We checked the obvious ways this could be an artifact — prompt wording, question order, an imposed persona, reasoning on and off — and we did not test open-ended answers, which one careful paper argues can change the result. The comparison group is 2,170 visitors to this site, not a representative sample. Every caveat, every control, and every number is in the long version.

5 Your turn.

The 32 statements the models answered are the ones you would answer. It takes about four minutes, it is free, and your result page tells you which of the models answered most like you — and which one least.

Find out which AI thinks like you.

Same quiz, same scoring. No AI anywhere in the scoring.

Take the DNA Scan

Study v1 · data collected September 2, 2026–September 3, 2026 · what earlier studies found · download the dataset · corrections to research@politicaldna.org