Yet another algorithmic bias: A Discursive Analysis of Large Language Models Reinforcing Dominant Discourses on Gender and Race

2025-08-13

AI Summary

This study examines how large language models (LLMs) reproduce racial and gender inequities through narrative generation: Black women are systematically associated with ancestry, trauma, and resistance, whereas white women are granted individuated, introspective subjectivity. Addressing the limitations of quantitative bias detection—which fails to capture ideological mechanisms—this paper develops a qualitative discourse analysis framework integrating critical social theory to conduct close, comparative readings of LLM-generated short stories. Findings reveal that mainstream “bias mitigation” strategies merely alter surface-level lexical choices without disrupting deeper exclusionary discursive structures. The framework pioneers the integration of Foucauldian discourse analysis with algorithmic critique, offering a transferable methodological pathway for identifying and intervening in structural biases embedded in LLMs. (149 words)

Technology Category

• Machine Learning: Large Multimodal Models (LMMs)
• Natural Language Processing: (Large) Language Models
• Philosophy and Ethics of AI: Bias, Fairness & Equity

Application Category

• Social Networks and Social Media: Fairness and bias in social network and social media analysis
• User Modeling, Personalization and Recommendation: Large Language Models (LLM) for user modeling and recommendation
• Economics, Online Markets and Human Computation: Humans versus LLMs for data annotation and labeling

Abstract

With the advance of Artificial Intelligence (AI), Large Language Models (LLMs) have gained prominence and been applied in diverse contexts. As they evolve into more sophisticated versions, it is essential to assess whether they reproduce biases, such as discrimination and racialization, while maintaining hegemonic discourses. Current bias detection approaches rely mostly on quantitative, automated methods, which often overlook the nuanced ways in which biases emerge in natural language. This study proposes a qualitative, discursive framework to complement such methods. Through manual analysis of LLM-generated short stories featuring Black and white women, we investigate gender and racial biases. We contend that qualitative methods such as the one proposed here are fundamental to help both developers and users identify the precise ways in which biases manifest in LLM outputs, thus enabling better conditions to mitigate them. Results show that Black women are portrayed as tied to ancestry and resistance, while white women appear in self-discovery processes. These patterns reflect how language models replicate crystalized discursive representations, reinforcing essentialization and a sense of social immobility. When prompted to correct biases, models offered superficial revisions that maintained problematic meanings, revealing limitations in fostering inclusive narratives. Our results demonstrate the ideological functioning of algorithms and have significant implications for the ethical use and development of AI. The study reinforces the need for critical, interdisciplinary approaches to AI design and deployment, addressing how LLM-generated discourses reflect and perpetuate inequalities.

Problem

• Analyzing gender and racial biases in LLM outputs
• Proposing qualitative methods to detect nuanced biases
• Addressing limitations in fostering inclusive AI narratives

Innovation

• Qualitative discursive framework for bias detection
• Manual analysis of LLM-generated gender and racial biases
• Critical interdisciplinary approaches to AI design