Unequal Voices: How LLMs Construct Constrained Queer Narratives

TL;DR

This paper investigates how large language models (LLMs) often produce limited, stereotyped, and marginalizing narratives about queer identities, highlighting significant constraints in their representations.

Contribution

It introduces a framework to analyze constrained queer representations in LLMs and provides empirical evidence of their limited and stereotyped portrayals.

Findings

LLMs produce significantly stereotyped queer narratives

LLMs exhibit discursive othering of queer identities

Representations of queer personas are notably constrained

Abstract

One way social groups are marginalized in discourse is that the narratives told about them often default to a narrow, stereotyped range of topics. In contrast, default groups are allowed the full complexity of human existence. We describe the constrained representations of queer people in LLM generations in terms of harmful representations, narrow representations, and discursive othering and formulate hypotheses to test for these phenomena. Our results show that LLMs are significantly limited in their portrayals of queer personas.