As text-to-image systems continue to grow in popularity with the general public, questions have arisen about bias and diversity in the generated images. Here, we investigate properties of images generated in response to prompts which are visually under-specified, but contain salient social attributes (e.g., 'a portrait of a threatening person' versus 'a portrait of a friendly person'). Grounding our work in social cognition theory, we find that in many cases, images contain similar demographic biases to those reported in the stereotype literature. However, trends are inconsistent across different models and further investigation is warranted.
@article{arxiv.2302.07159,
title = {A Friendly Face: Do Text-to-Image Systems Rely on Stereotypes when the Input is Under-Specified?},
author = {Kathleen C. Fraser and Svetlana Kiritchenko and Isar Nejadgholi},
journal= {arXiv preprint arXiv:2302.07159},
year = {2023}
}
Comments
Appearing in the AAAI 2023 Workshop on Creative AI Across Modalities