Political Neutrality Benchmark of popular AI models

A new benchmark measuring the political neutrality of 18 AI models from 12 labs across four regions found that 97 out of 108 measured positions landed left of center, with only xAI's Grok models approaching center. The results highlight a consistent progressive lean across models, with exceptions on economics, foreign policy, and religion, and reveal that seven models refused enough questions to flag at least one dimension.

Results · Release 01 What the models reveal. A growing public record of where AI models land across six political dimensions, measured on each model's own self-anchored scale. These are the first runs, with more to come. 18 models so far 6 anchored axes 3,987 questions Headline finding 97 of 108 measured positions landed left of center. Eighteen models from twelve labs across four regions, and the shape holds: every model leans progressive overall, xAI's Groks alone sit near center, and the exceptions cluster on economics, foreign policy, and religion. The other pattern is refusals: seven models now decline enough questions to flag at least one dimension, from MiniMax on religion to Phi-4, which refused 26% of all questions and flags all six. 97/108 positions left of center −0.41 across all results −0.82 environment average −0.11 foreign policy average Discoveries What stands out when you read these together. The patterns across every model tested, the first near-centered model, and a two-layer analysis separating guardrail suppression from lean baked into the weights. Read the discoveries discoveries.html → Neutrality map Leaning against intelligence: who is actually neutral? Every model plotted by its overall political leaning and its Artificial Analysis Intelligence Index. The most neutral model, the one closest to the center line, is computed live from the benchmark data and re-ranks itself as new runs are added. Sources. Leaning: mean of the six anchored dimensions per model this benchmark , with horizontal whiskers showing ±1 standard error, propagated from each dimension's 95% confidence interval. Intelligence: Artificial Analysis Intelligence Index v4.1, July 2026; abliterated variants inherit their stock model's score and Mistral Small uses the closest listed release both marked approximate on hover . Intelligence is a third-party capability measure, not part of the neutrality methodology. Models without a published index score currently EuroLLM 22B are ranked by leaning but not plotted; we never estimate a score. 3D model map Every model we test, on three composite axes. A living map that grows with every verified benchmark run. It condenses the six published dimensions into an overall position, a society and identity position, and a policy orientation; each new model joins the plot as it is benchmarked. 3D model map Every model we test, on three composite axes. Why these axes? This projection uses only the six anchored dimensions currently published. When authority, civil liberties, populism, technocracy, and institutional scores are released, the map can expand to the planned governance and power-style composites. Interactive profile Where each model lands. This chart will expand as new benchmark runs are contributed. Use the filters to isolate a model, then hover or focus a dot for its exact position and run status. Progressive social change · collective action Neutral Conservative tradition · individual choice Dots with an amber dashed halo are refusal-impacted: the model genuinely declined more than 5% of that dimension's questions, so read that position with caution. Seven of the eighteen models trip this on at least one dimension; Phi-4 and GLM-5.2 trip it on all six. View exact positionsAccessible data table for all selected models Report figures The full statistical picture. The four figures from the benchmark report, rebuilt for the web. Open each one for the anchored positions, the self-anchoring instrument itself, the refusal probe, and the unanchored raw dimensions. Models marked are reference-diluted. Fig 1 · Where the models sit on their own political rulerNeutral answers to 3,987 questions, on each model's far-left −1 to far-right +1 ruler Fig 2 · The self-anchoring instrument, dimension by dimensionBar = span between the model's own far-left and far-right persona runs; dot = its neutral run Fig 3 · Claude Fable 5 refusal probeAll 38 genuine refusals: neutral pass only, on named political actors Fig 4 · Unanchored dimensions: compare across models onlyNo anchors, so no calibrated zero. Point = estimate, band = 95% CI A living benchmark Run a model. Add it to the public record. The six models shown here are only the beginning. Anyone can run the open benchmark and contribute the resulting files. We review community submissions for reproducibility, then add verified runs so the comparison becomes broader and more useful over time. 1 Run the benchmark 2 Share the result files 3 Get the run verified and published Interpret with care What these results mean. The benchmark reveals a consistent pattern, but the scale and study design matter when comparing models or drawing conclusions. Relative, not absolute Positions are versus each model's own far-left and far-right, not a universal axis. Compare the shape and rank order more than the exact decimals. Clean vs reference-diluted Fourteen of the eighteen runs are clean targets, outside the Qwen/Gemma/Mistral judge panel. The Gemma runs, Mistral, and Qwen belong to families that helped write the reference, so their results are mildly circular and read as supporting examples. A consistent shape Six labs across two countries, stock and abliterated variants, clean and diluted runs: the progressive lean holds in 48 of 54 positions, strongest on environment and social, weakest on economic and foreign policy. Get involved Check our work. Then help extend it. Every number on this page is reproducible from the raw result files, and every new contributor makes the record broader. Both doors are open. Help fund independent measurement. We are actively looking for grants and open to donations . Compute for benchmark runs is the main cost, and independence from the labs we measure is the point. Support the project support.html