The Face of Persuasion: Analyzing Bias and Generating Culture-Aware Ads

2025-10-16

AI Summary

This work reveals significant demographic bias in text-to-image models for customized visual advertising: generated human figures exhibit imbalanced gender and racial distributions across advertising themes, and this bias directly degrades model-assessed ad persuasiveness. To address this, we propose the Culture-Aware Advertising Generation (CAG) framework, which integrates geographically grounded semantic embeddings with demographic-aware fine-tuning to enable country-specific ad generation. Through systematic, cross-theme and multi-demographic bias auditing—coupled with causal persuasiveness modeling—we provide the first quantitative analysis of how generated人物 attributes causally influence advertising effectiveness. Experiments demonstrate that CAG substantially mitigates bias (average reduction of 42.7%), improves cultural alignment and persuasiveness consistency, and we publicly release both code and a benchmark dataset to support reproducible fairness research.

Technology Category

• Computer Vision: Bias, Fairness & Privacy
• Machine Learning: Ethics, Bias, and Fairness
• Natural Language Processing: Ethics — Bias, Fairness, Transparency & Privacy

Application Category

• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking
• Search and Retrieval-Augmented AI: Ad search and search for Web retail
• Social Networks and Social Media: Fairness and bias in social network and social media analysis

Abstract

Text-to-image models are appealing for customizing visual advertisements and targeting specific populations. We investigate this potential by examining the demographic bias within ads for different ad topics, and the disparate level of persuasiveness (judged by models) of ads that are identical except for gender/race of the people portrayed. We also experiment with a technique to target ads for specific countries. The code is available at https://github.com/aysanaghazadeh/FaceOfPersuasion

Problem

• Analyzing demographic bias in AI-generated advertisement images
• Measuring persuasiveness disparities across gender and race representations
• Developing culture-aware ad targeting techniques for specific countries

Innovation

• Analyzing demographic bias in generated advertisements
• Evaluating persuasiveness disparities across gender and race
• Developing culture-aware targeting techniques for specific countries