Propaganda is all you need

TL;DR

This paper explores the political biases embedded in large language models, examining how alignment processes influence societal perceptions and the potential implications for political decision-making and societal structure.

Contribution

It introduces a socio-political framework to analyze political biases in LLMs and highlights the impact of alignment on embedding spaces and societal influence.

Findings

Alignment affects political notions in LLMs' embedding space

Biases may reflect the dominant ideology as per Marxist theory

Potential societal impacts include increased uniformity or disguised extremism

Abstract

As Machine Learning (ML) is still a recent field of study, especially outside the realm of abstract Mathematics and Computer Science, few works have been conducted on the political aspect of large Language Models (LLMs), and more particularly about the alignment process and its political dimension. This process can be as simple as prompt engineering but is also very complex and can affect completely unrelated notions. For example, politically directed alignment has a very strong impact on an LLM's embedding space and the relative position of political notions in such a space. Using special tools to evaluate general political bias and analyze the effects of alignment, we can gather new data to understand its causes and possible consequences on society. Indeed, by taking a socio-political approach, we can hypothesize that most big LLMs are aligned with what Marxist philosophy calls the 'dominant ideology.' As AI's role in political decision-making, at the citizen's scale but also in government agencies, such biases can have huge effects on societal change, either by creating new and insidious pathways for societal uniformity or by allowing disguised extremist views to gain traction among the people.