AI's Regimes of Representation: A Community-centered Study of Text-to-Image Models in South Asia

Abstract

This paper presents a community-centered study of cultural limitations of text-to-image (T2I) models in the South Asian context. We theorize these failures using scholarship on dominant media regimes of representations and locate them within participants' reporting of their existing social marginalizations. We thus show how generative AI can reproduce an outsiders gaze for viewing South Asian cultures, shaped by global and regional power inequities. By centering communities as experts and soliciting their perspectives on T2I limitations, our study adds rich nuance into existing evaluative frameworks and deepens our understanding of the culturally-specific ways AI technologies can fail in non-Western and Global South settings. We distill lessons for responsible development of T2I models, recommending concrete pathways forward that can allow for recognition of structural inequalities.

References in corpus (21)

• Hierarchical Text-Conditional Image Generation with CLIP Latents
• Photorealistic Text-to-Image Diffusion Models with Deep Language Understanding
• Zero-Shot Text-to-Image Generation
• Decolonial AI: Decolonial Theory as Sociotechnical Foresight in Artificial Intelligence
• Scaling Autoregressive Models for Content-Rich Text-to-Image Generation
• Easily Accessible Text-to-Image Generation Amplifies Demographic Stereotypes at Large Scale
• Towards a Critical Race Methodology in Algorithmic Fairness
• Multimodal datasets: misogyny, pornography, and malignant stereotypes
• Rethinking Fairness: An Interdisciplinary Survey of Critiques of Hegemonic ML Fairness Approaches
• The Forgotten Margins of AI Ethics
• A Systematic Review and Thematic Analysis of Community-Collaborative Approaches to Computing Research
• CrowdWorkSheets: Accounting for Individual and Collective Identities Underlying Crowdsourced Dataset Annotation
• Cultural Incongruencies in Artificial Intelligence
• Contrastive Language-Vision AI Models Pretrained on Web-Scraped Multimodal Data Exhibit Sexual Objectification Bias
• DALL-Eval: Probing the Reasoning Skills and Social Biases of Text-to-Image Generation Models
• Whose Ground Truth? Accounting for Individual and Collective Identities Underlying Dataset Annotation
• Stakeholder Participation in AI: Beyond "Add Diverse Stakeholders and Stir"
• Sociotechnical Harms of Algorithmic Systems: Scoping a Taxonomy for Harm Reduction
• Non-portability of Algorithmic Fairness in India
• Biases in Generative Art -- A Causal Look from the Lens of Art History
• Visual Conceptual Blending with Large-scale Language and Vision Models

Cited by in corpus (9)

• Data Feminism for AI
• AI Suggestions Homogenize Writing Toward Western Styles and Diminish Cultural Nuances
• Collective Constitutional AI: Aligning a Language Model with Public Input
• Beyond Behaviorist Representational Harms: A Plan for Measurement and Mitigation
• From human-centered to social-centered artificial intelligence: Assessing ChatGPT's impact through disruptive events
• From Fake Perfects to Conversational Imperfects: Exploring Image-Generative AI as a Boundary Object for Participatory Design of Public Spaces
• Understanding Gender Bias in AI-Generated Product Descriptions
• Evaluating the Social Impact of Generative AI Systems in Systems and Society
• Recourse for reclamation: Chatting with generative language models