T2IBias: Uncovering Societal Bias Encoded in the Latent Space of Text-to-Image Generative Models

2025-11-13

AI Summary

This study is the first to systematically uncover embedded racial and gender biases within the pretraining latent spaces of text-to-image (T2I) models. To quantify bias, we conducted a large-scale human evaluation of 5,000 images generated by five prominent open-source models—including Qwen-Image and Kandinsky—using neutral occupational prompts and a diverse pool of annotators. Results reveal that all models significantly amplify stereotypical associations: caregiving roles are over-attributed to women, high-status roles to men, and whiteness dominates across occupations. Moreover, model-specific biases emerge—e.g., Qwen-Image exhibits a preference for East Asian facial features, while Kandinsky shows pronounced underrepresentation of non-white subjects. Our work establishes a reproducible, human-centered methodology for fairness assessment in T2I systems, enabling informed model selection, bias-mitigating prompt engineering, and responsible deployment practices.

Technology Category

• Computer Vision: Bias, Fairness & Privacy
• Machine Learning: Ethics, Bias, and Fairness
• 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: Fairness-aware retrieval and ranking
• Web Mining and Content Analysis: Large pretrained models with web data

Abstract

Text-to-image (T2I) generative models are largely used in AI-powered real-world applications and value creation. However, their strategic deployment raises critical concerns for responsible AI management, particularly regarding the reproduction and amplification of race- and gender-related stereotypes that can undermine organizational ethics. In this work, we investigate whether such societal biases are systematically encoded within the pretrained latent spaces of state-of-the-art T2I models. We conduct an empirical study across the five most popular open-source models, using ten neutral, profession-related prompts to generate 100 images per profession, resulting in a dataset of 5,000 images evaluated by diverse human assessors representing different races and genders. We demonstrate that all five models encode and amplify pronounced societal skew: caregiving and nursing roles are consistently feminized, while high-status professions such as corporate CEO, politician, doctor, and lawyer are overwhelmingly represented by males and mostly White individuals. We further identify model-specific patterns, such as QWEN-Image's near-exclusive focus on East Asian outputs, Kandinsky's dominance of White individuals, and SDXL's comparatively broader but still biased distributions. These results provide critical insights for AI project managers and practitioners, enabling them to select equitable AI models and customized prompts that generate images in alignment with the principles of responsible AI. We conclude by discussing the risks of these biases and proposing actionable strategies for bias mitigation in building responsible GenAI systems.

Problem

• Investigating societal biases encoded in text-to-image model latent spaces
• Analyzing race and gender stereotypes amplified across five popular T2I models
• Identifying biased occupational representations in AI-generated professional images

Innovation

• Detect bias in T2I models via latent space analysis
• Use neutral prompts to generate profession images
• Evaluate bias with human assessors across demographics