Personality Shapes Gender Bias in Persona-Conditioned LLM Narratives Across English and Hindi: An Empirical Investigation

2026-04-26

AI Summary

This study investigates how personality traits modulate gender bias in content generated by large language models (LLMs) under role-playing conditions, focusing on English–Hindi bilingual contexts and Indian workplace scenarios. By systematically manipulating gender, occupation, and personality traits—spanning the HEXACO dimensions and the Dark Triad—the authors generated 23,400 professional narratives across six state-of-the-art LLMs. Combining multilingual textual bias analysis with controlled experiments, the work reveals, for the first time, a dynamic relationship between personality and LLM-generated gender bias. Findings indicate that Dark Triad traits significantly amplify stereotyping, whereas prosocial HEXACO traits mitigate bias, with effects varying across models and languages. These results demonstrate the context-dependent nature of algorithmic bias and highlight how role-based prompting may engender uneven representational harms.

Technology Category

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

Application Category

• User Modeling, Personalization and Recommendation: Large Language Models (LLM) for user modeling and recommendation
• Social Networks and Social Media: Generative AI / large language models and their impact on social systems
• Economics, Online Markets and Human Computation: Cost models of using LLMs in production systems

Abstract

Large Language Models (LLMs) are increasingly deployed in persona-driven applications such as education, customer service, and social platforms, where models are prompted to adopt specific personas when interacting with users. While persona conditioning can improve user experience and engagement, it also raises concerns about how personality cues may interact with gender biases and stereotypes. In this work, we present a controlled study of persona-conditioned story generation in English and Hindi, where each story portrays a working professional in India producing context-specific artifacts (e.g., lesson plans, reports, letters) under systematically varied persona gender, occupational role, and personality traits from the HEXACO and Dark Triad frameworks. Across 23,400 generated stories from six state-of-the-art LLMs, we find that personality traits are significantly associated with both the magnitude and direction of gender bias. In particular, Dark Triad personality traits are consistently associated with higher gender-stereotypical representations compared to socially desirable HEXACO traits, though these associations vary across models and languages. Our findings demonstrate that gender bias in LLMs is not static but context-dependent. This suggests that persona-conditioned systems used in real-world applications may introduce uneven representational harms, reinforcing gender stereotypes in generated educational, professional, or social content.

Problem

• gender bias
• personality traits
• persona-conditioned LLMs
• stereotypes
• cross-lingual

Innovation

• persona-conditioned generation
• gender bias modulation
• personality traits
• cross-lingual analysis
• Dark Triad