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

• URL: http://arxiv.org/abs/2511.10089v2
• Date: Sat, 15 Nov 2025 11:37:58 GMT
• Title: T2IBias: Uncovering Societal Bias Encoded in the Latent Space of Text-to-Image Generative Models
• Authors: Abu Sufian, Cosimo Distante, Marco Leo, Hanan Salam,
• Abstract summary: Text-to-image (T2I) generative models are largely used in AI-powered real-world applications and value creation.<n>We investigate whether societal biases are systematically encoded within the pretrained latent spaces of state-of-the-art T2I models.
• Score: 5.565960549039278
• License: http://creativecommons.org/licenses/by-nc-sa/4.0/
• 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. The code and Data Repository: https://github.com/Sufianlab/T2IBias

Related papers

• Prompting Away Stereotypes? Evaluating Bias in Text-to-Image Models for Occupations [9.58968557546246]
  We frame representational societal bias assessment as an image curation and evaluation task.<n>Using five state-of-the-art models, we compare neutral baseline prompts against fairness-aware controlled prompts.<n>Results show that prompting can substantially shift demographic representations, but with highly model-specific effects.
  arXiv  Detail & Related papers  (2025-08-31T13:46:16Z)
• Adultification Bias in LLMs and Text-to-Image Models [55.02903075972816]
  We study bias along axes of race and gender in young girls.<n>We focus on "adultification bias," a phenomenon in which Black girls are presumed to be more defiant, sexually intimate, and culpable than their White peers.
  arXiv  Detail & Related papers  (2025-06-08T21:02:33Z)
• Can we Debias Social Stereotypes in AI-Generated Images? Examining Text-to-Image Outputs and User Perceptions [6.87895735248661]
  This paper proposes a theory-driven bias detection rubric and a Social Stereotype Index (SSI) to evaluate social biases in T2I outputs.<n>We audited three major T2I model outputs using 100 queries across three categories -- geocultural, occupational, and adjectival.<n>Our findings reveal a key tension -- although prompt refinement can mitigate stereotypes, it can limit contextual alignment.
  arXiv  Detail & Related papers  (2025-05-27T04:01:03Z)
• Could AI Trace and Explain the Origins of AI-Generated Images and Text? [53.11173194293537]
  AI-generated content is increasingly prevalent in the real world.<n> adversaries might exploit large multimodal models to create images that violate ethical or legal standards.<n>Paper reviewers may misuse large language models to generate reviews without genuine intellectual effort.
  arXiv  Detail & Related papers  (2025-04-05T20:51:54Z)
• Using complex prompts to identify fine-grained biases in image generation through ChatGPT-4o [0.0]
  Two dimensions of bias can be revealed through the study of large AI models.<n>Not only bias in training data or the products of an AI, but also bias in society.<n>I briefly discuss how we can use complex prompts to image generation AI to investigate either dimension of bias.
  arXiv  Detail & Related papers  (2025-04-01T03:17:35Z)
• A Large Scale Analysis of Gender Biases in Text-to-Image Generative Models [47.16682882493828]
  This paper presents a large-scale study on gender bias in text-to-image (T2I) models, focusing on everyday situations.<n>We create a dataset of 3,217 gender-neutral prompts and generate 200 images over 5 prompt variations per prompt from five leading T2I models.<n>We automatically detect the perceived gender of people in the generated images and filter out images with no person or multiple people of different genders.<n>Our analysis shows that T2I models reinforce traditional gender roles and reflect common gender stereotypes in household roles.
  arXiv  Detail & Related papers  (2025-03-30T11:11:51Z)
• Exploring Bias in over 100 Text-to-Image Generative Models [49.60774626839712]
  We investigate bias trends in text-to-image generative models over time, focusing on the increasing availability of models through open platforms like Hugging Face.<n>We assess bias across three key dimensions: (i) distribution bias, (ii) generative hallucination, and (iii) generative miss-rate.<n>Our findings indicate that artistic and style-transferred models exhibit significant bias, whereas foundation models, benefiting from broader training distributions, are becoming progressively less biased.
  arXiv  Detail & Related papers  (2025-03-11T03:40:44Z)
• Bias in Generative AI [2.5830293457323266]
  This study analyzed images generated by three popular generative artificial intelligence (AI) tools to investigate potential bias in AI generators. All three AI generators exhibited bias against women and African Americans. Women were depicted as younger with more smiles and happiness, while men were depicted as older with more neutral expressions and anger.
  arXiv  Detail & Related papers  (2024-03-05T07:34:41Z)
• The Male CEO and the Female Assistant: Evaluation and Mitigation of Gender Biases in Text-To-Image Generation of Dual Subjects [58.27353205269664]
  We propose the Paired Stereotype Test (PST) framework, which queries T2I models to depict two individuals assigned with male-stereotyped and female-stereotyped social identities.<n>PST queries T2I models to depict two individuals assigned with male-stereotyped and female-stereotyped social identities.<n>Using PST, we evaluate two aspects of gender biases -- the well-known bias in gendered occupation and a novel aspect: bias in organizational power.
  arXiv  Detail & Related papers  (2024-02-16T21:32:27Z)
• CBBQ: A Chinese Bias Benchmark Dataset Curated with Human-AI Collaboration for Large Language Models [52.25049362267279]
  We present a Chinese Bias Benchmark dataset that consists of over 100K questions jointly constructed by human experts and generative language models. The testing instances in the dataset are automatically derived from 3K+ high-quality templates manually authored with stringent quality control. Extensive experiments demonstrate the effectiveness of the dataset in detecting model bias, with all 10 publicly available Chinese large language models exhibiting strong bias in certain categories.
  arXiv  Detail & Related papers  (2023-06-28T14:14:44Z)
• Fairness And Bias in Artificial Intelligence: A Brief Survey of Sources, Impacts, And Mitigation Strategies [11.323961700172175]
  This survey paper offers a succinct, comprehensive overview of fairness and bias in AI. We review sources of bias, such as data, algorithm, and human decision biases. We assess the societal impact of biased AI systems, focusing on the perpetuation of inequalities and the reinforcement of harmful stereotypes.
  arXiv  Detail & Related papers  (2023-04-16T03:23:55Z)
• Stable Bias: Analyzing Societal Representations in Diffusion Models [72.27121528451528]
  We propose a new method for exploring the social biases in Text-to-Image (TTI) systems. Our approach relies on characterizing the variation in generated images triggered by enumerating gender and ethnicity markers in the prompts. We leverage this method to analyze images generated by 3 popular TTI systems and find that while all of their outputs show correlations with US labor demographics, they also consistently under-represent marginalized identities to different extents.
  arXiv  Detail & Related papers  (2023-03-20T19:32:49Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.