Documenting Patterns of Exoticism of Marginalized Populations within Text-to-Image Generators

2025-08-04

AI Summary

This study identifies systemic exoticization in text-to-image generative models toward Global South nations and marginalized communities—including Indigenous peoples and multiply marginalized groups—manifesting as overemphasis on stereotypical cultural markers (e.g., attire) while underrepresenting everyday practices and contextual authenticity, thereby reinforcing homogenized, reductive representations. Through controlled image-generation experiments and multi-case qualitative analysis, we systematically document and quantify these bias patterns for the first time. We propose a novel “community-engaged design” framework that integrates local knowledge holders into model evaluation and iterative refinement. Empirical results demonstrate that this approach significantly improves cultural accuracy and representational diversity. Our work provides both empirical evidence and a methodological paradigm for advancing fairness governance in generative AI.

Technology Category

• Computer Vision: Bias, Fairness & Privacy
• Philosophy and Ethics of AI: Bias, Fairness & Equity
• Machine Learning: Ethics, Bias, and Fairness

Application Category

• Social Networks and Social Media: Generative AI / large language models and their impact on social systems
• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking
• Economics, Online Markets and Human Computation: Trust and reliance of crowd workers and data experts on GenAI

Abstract

A significant majority of AI fairness research studying the harmful outcomes of GAI tools have overlooked non-Western communities and contexts, necessitating a stronger coverage in this vein. We extend our previous work on exoticism (Ghosh et al., 2024) of 'Global South' countries from across the world, as depicted by GAI tools. We analyze generated images of individuals from 13 countries -- India, Bangladesh, Papua New Guinea, Egypt, Ethiopia, Tunisia, Sudan, Libya, Venezuela, Colombia, Indonesia, Honduras, and Mexico -- performing everyday activities (such as being at home, going to work, getting groceries, etc.), as opposed to images for the same activities being performed by persons from 3 'Global North' countries -- USA, UK, Australia. While outputs for 'Global North' demonstrate a difference across images and people clad in activity-appropriate attire, individuals from 'Global South' countries are depicted in similar attire irrespective of the performed activity, indicative of a pattern of exoticism where attire or other cultural features are overamplified at the cost of accuracy. We further show qualitatively-analyzed case studies that demonstrate how exoticism is not simply performed upon 'Global South' countries but also upon marginalized populations even in Western contexts, as we observe a similar exoticization of Indigenous populations in the 'Global North', and doubly upon marginalized populations within 'Global South' countries. We document implications for harm-aware usage patterns of such tools, and steps towards designing better GAI tools through community-centered endeavors.

Problem

• Exoticism patterns in text-to-image generators for marginalized populations
• Overlooked non-Western communities in AI fairness research
• Harmful exoticization in both Global South and Western contexts

Innovation

• Analyzing exoticism in text-to-image generators
• Comparing Global South and North depictions
• Proposing harm-aware GAI design steps