Assessing Political Bias in Large Language Models

TL;DR

This paper evaluates political bias in popular open-source LLMs concerning EU political issues, revealing size-related bias tendencies and emphasizing the need for transparency to ensure trustworthy AI applications.

Contribution

It introduces a novel method to quantify political bias in LLMs using the Wahl-O-Mat voting tool, highlighting size and language influences on bias.

Findings

Llama3-70B aligns more with left-leaning parties

Smaller models tend to be more neutral in English prompts

Biases are consistent with low variance across models

Abstract

The assessment of bias within Large Language Models (LLMs) has emerged as a critical concern in the contemporary discourse surrounding Artificial Intelligence (AI) in the context of their potential impact on societal dynamics. Recognizing and considering political bias within LLM applications is especially important when closing in on the tipping point toward performative prediction. Then, being educated about potential effects and the societal behavior LLMs can drive at scale due to their interplay with human operators. In this way, the upcoming elections of the European Parliament will not remain unaffected by LLMs. We evaluate the political bias of the currently most popular open-source LLMs (instruct or assistant models) concerning political issues within the European Union (EU) from a German voter's perspective. To do so, we use the "Wahl-O-Mat," a voting advice application used in Germany. From the voting advice of the "Wahl-O-Mat" we quantize the degree of alignment of LLMs with German political parties. We show that larger models, such as Llama3-70B, tend to align more closely with left-leaning political parties, while smaller models often remain neutral, particularly when prompted in English. The central finding is that LLMs are similarly biased, with low variances in the alignment concerning a specific party. Our findings underline the importance of rigorously assessing and making bias transparent in LLMs to safeguard the integrity and trustworthiness of applications that employ the capabilities of performative prediction and the invisible hand of machine learning prediction and language generation.