# Progressive in Principle, Centrist in Practice: LLM Political Bias Is Instrument-Dependent

arXiv: https://arxiv.org/abs/2606.00048 · HTML: https://mlsanity.com/paper/2606.00048/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 19 of its 39 references that are in the map.
- First posted on arXiv 3 May 2026.

## Builds on (most cited first, 19 of 19)

- [Training language models to follow instructions with human feedback](https://mlsanity.com/paper/2203.02155/) (2022)
- [Direct Preference Optimization: Your Language Model is Secretly a Reward Model](https://mlsanity.com/paper/2305.18290/) (2023)
- [Whose Opinions Do Language Models Reflect?](https://mlsanity.com/paper/2303.17548/) (2023)
- [From Pretraining Data to Language Models to Downstream Tasks: Tracking the Trails of Political Biases Leading to Unfair NLP Models](https://mlsanity.com/paper/2305.08283/) (2023)
- [The political ideology of conversational AI: Converging evidence on ChatGPT's pro-environmental, left-libertarian orientation](https://mlsanity.com/paper/2301.01768/) (2023)
- [Political Compass or Spinning Arrow? Towards More Meaningful Evaluations for Values and Opinions in Large Language Models](https://mlsanity.com/paper/2402.16786/) (2024)
- [Hidden Persuaders: LLMs' Political Leaning and Their Influence on Voters](https://mlsanity.com/paper/2410.24190/) (2024)
- [Large Language Models Reflect the Ideology of their Creators](https://mlsanity.com/paper/2410.18417/) (2024)
- [Beyond prompt brittleness: Evaluating the reliability and consistency of political worldviews in LLMs](https://mlsanity.com/paper/2402.17649/) (2024)
- [Only a Little to the Left: A Theory-grounded Measure of Political Bias in Large Language Models](https://mlsanity.com/paper/2503.16148/) (2025)
- [Large Means Left: Political Bias in Large Language Models Increases with Their Number of Parameters](https://mlsanity.com/paper/2505.04393/) (2025)
- [Measuring Political Preferences in AI Systems: An Integrative Approach](https://mlsanity.com/paper/2503.10649/) (2025)
- [The Political Ideology of Large Language Models: Measurement, Inconsistency, and Persuasive Influence](https://mlsanity.com/paper/2505.04171/) (2025)
- [A Cautionary Tale About "Neutrally" Informative AI Tools Ahead of the 2025 Federal Elections in Germany](https://mlsanity.com/paper/2502.15568/) (2025)
- [Framing Political Bias in Multilingual LLMs Across Pakistani Languages](https://mlsanity.com/paper/2506.00068/) (2025)
- [Political Alignment in Large Language Models: A Multidimensional Audit of Psychometric Identity and Behavioral Bias](https://mlsanity.com/paper/2601.06194/) (2026)
- [Bias Beyond Borders: Political Ideology Evaluation and Steering in Multilingual LLMs](https://mlsanity.com/paper/2601.23001/) (2026)
- [Who Would Chatbots Vote For? Political Preferences of ChatGPT and Gemini in the 2024 European Union Elections](https://mlsanity.com/paper/2409.00721/) (2024)
- [PReSS: An Automated Black-Box Framework for Evaluating Political Stance Stability in LLMs](https://mlsanity.com/paper/2504.17052/) (2025)

## Abstract (arXiv)

Prior work establishes that instruction-tuned LLMs exhibit left-of-center political bias, but measures it exclusively through abstract questionnaires. We show it does not predict how models vote on concrete policies. We introduce a dual-instrument methodology grounded in Swiss direct democracy. First, we administer the Smartvote questionnaire (75 policy questions) to 66 LLMs and compare their answers to those of 184 elected members of the Swiss National Council. Second, we put 48 real federal referenda (Volksabstimmungen) to 9 flagship LLMs in four national languages and three information conditions, and compare their votes to the actual outcomes and to party recommendations (Parolen). The instruments disagree. (1) The left-to-right agreement gradient that dominates Smartvote replicates prior work (mean rho = -0.77). On referenda it shifts to center-peaked: models align most with centrist Die Mitte and FDP rather than leftist SP and Gr"une (Wilcoxon p = 0.008). (2) For some models the language of a question changes the answer: cross-linguistic consistency ranges from 50% (Mistral) to 98% (GPT-5.4). (3) Two models vote Nein on 83-94% of referenda at similar rates on progressive and conservative proposals (binomial p < 0.0001), change-aversion rather than a left-right bias. What prior work measured as "leftward bias" may not extend beyond abstract instruments: confronted with real decisions, LLMs behave less like coalition partners of the left than like cautious civil servants, centrist and inconsistent across languages.