LLM-Ideoplasticity: Measuring Ideological Plasticity in the Political Behavior of LLMs as a Context-Conditioned Distribution

2026-05-26

AI Summary

This study addresses the common oversimplification in existing research that treats large language models’ (LLMs’) political orientations as fixed points, thereby neglecting their contextual dependence. The work proposes the first systematic framework grounded in the VAA-CHES political space projection model, integrating multi-context experiments, chain-of-thought prompting, and cross-linguistic/rhetorical perturbations to demonstrate that LLM political stances are better characterized as conditional distributions rather than static coordinates. Through multi-trait multi-method (MTMM) validation, the study establishes the novel concept of “ideological plasticity,” showing that persuasive phrasing and low-resource languages can shift ideological positions by up to 0.57 and 0.52 units, respectively. It further reveals that the aggregate political space occupied by current LLMs spans only about one-third of that covered by major European political parties, advocating for a distributional approach to modeling LLM political behavior and offering a new paradigm for AI ideological assessment.

Abstract

We argue, with systematic empirical evidence, that a large language model's political ideology is not a fixed point, but a conditional distribution $\mathbb{P}($position$\mid$context$)$ over a real political space. We evaluate nine current LLMs using a unified measurement framework anchored by VAA-CHES projection models, which map responses onto three validated dimensions (lrgen, lrecon, galtan) across six contextual axes. Our findings reveal high sensitivity to context: persuasive framing and under-represented languages displace coordinates by up to 0.57 and 0.52 units, respectively, while chain-of-thought reasoning often amplifies rather than dampens paraphrase instability. Despite this local plasticity, the model cohort occupies a remarkably narrow Overton envelope overall, occupying roughly one-third the spread of major European parties. Supported by a multi-trait multi-method (MTMM) analysis, we conclude that a single point cannot summarize LLM political behavior; it must be characterized as a shape. Our code and data are publicly available at https://github.com/sakhadib/LLM-Ideoplasticity.

Problem

• ideological plasticity
• political behavior
• large language models
• context-conditioned distribution
• Overton window

Innovation

• ideological plasticity
• context-conditioned distribution
• large language models
• political behavior
• VAA-CHES projection