When Cars Have Stereotypes: Auditing Demographic Bias in Objects from Text-to-Image Models

TL;DR

This paper introduces SODA, a framework for detecting demographic biases in objects generated by text-to-image models, revealing stereotypes and disparities across multiple models and object categories.

Contribution

The paper presents a novel systematic auditing framework, SODA, to measure demographic biases in object generation from state-of-the-art text-to-image models.

Findings

Strong associations between demographic cues and visual attributes

Models generate less diverse outputs, amplifying biases

Biases reflect and reinforce stereotypes in generated objects

Abstract

While prior research on text-to-image generation has predominantly focused on biases in human depictions, we investigate a more subtle yet pervasive phenomenon: demographic bias in generated objects (e.g., cars). We introduce SODA (Stereotyped Object Diagnostic Audit), a novel framework for systematically measuring such biases. Our approach compares visual attributes of objects generated with demographic cues (e.g., "for young people'') to those from neutral prompts, across 2,700 images produced by three state-of-the-art models (GPT Image-1, Imagen 4, and Stable Diffusion) in five object categories. Through a comprehensive analysis, we uncover strong associations between specific demographic groups and visual attributes, such as recurring color patterns prompted by gender or ethnicity cues. These patterns reflect and reinforce not only well-known stereotypes but also more subtle and unintuitive biases. We also observe that some models generate less diverse outputs, which in turn amplifies the visual disparities compared to neutral prompts. Our proposed auditing framework offers a practical approach for testing, revealing how stereotypes still remain embedded in today's generative models. We see this as an essential step toward more systematic and responsible AI development.