Publication Type

Conference Proceeding Article

Version

publishedVersion

Publication Date

6-2026

Abstract

Text-to-image (T2I) generative models are increasingly used to produce content for education, media, and public-facing communication, and are starting to be integrated into higher-impact pipelines. Since generated images tend to reinforce stereotypes, producing representational erasure via “default” depictions and shaping perceptions of who belongs in certain roles, a growing body of work has proposed metrics to quantify gender bias in T2I outputs. Yet existing evaluations remain fragmented. Metrics are often reported without a shared view of what they measure, what assumptions they entail, or how their results should be interpreted under different deployment contexts. This limits the usefulness of gender bias measurement for both technical auditing and emerging governance discussions. We propose a risk-aligned auditing framework for gender bias in T2I models composed of three constituents that connects risk categories, evaluation metrics, and harms. First, we identify risk-tiered use-case profiles aligned with the EU AI Act's risk categories to motivate why auditing expectations may vary with deployment contexts and stakeholder exposure. Second, we construct a metric catalog that consolidates gender-bias evaluation methods and organizes them in three measurement categories: gender prediction, embedding similarity, and downstream task. Third, we introduce a harm typology that maps context-dependent harm categories (e.g., representational, quality-of-service) to specific risk-tired scenarios. Finally, we introduce THUMB cards (Text-to-image Harms-informed Use-case-aligned Metrics of gender Bias) that help formulate auditing systematically by the incorporation of context, scenario and bias manifestation, harm hypotheses, and audit strategy. Thus, by linking measurements and harms to specific risk-tiered deployment contexts, our framework supports a more reusable format for better auditing reporting.

Keywords

accountability, AI governance, evaluation metrics, gender bias evaluation, generative AI auditing, risk-based governance, sociotechnical harms

Discipline

Artificial Intelligence and Robotics

Research Areas

Intelligent Systems and Optimization

Areas of Excellence

Digital transformation

Publication

FAccT '26: The 2026 ACM Conference on Fairness, Accountability, and Transparency, Montreal, Canada, June 25-28

First Page

1454

Last Page

1475

ISBN

9798400725968

Identifier

10.1145/3805689.3812261

Publisher

ACM

City or Country

New York

Additional URL

https://doi.org/10.1145/3805689.3812261

Download

Share

COinS