mlsanityData to 7 Oct 2026

'Person' == Light-skinned, Western Man, and Sexualization of Women of Color: Stereotypes in Stable Diffusion

arXiv 2310.19981 ↗First posted 30 Oct 2023cs.CVcited by 33

On the map: Diffusion models → Image generation, text-to-image and editing

Abstract (arXiv)

We study stereotypes embedded within one of the most popular text-to-image generators: Stable Diffusion. We examine what stereotypes of gender and nationality/continental identity does Stable Diffusion display in the absence of such information i.e. what gender and nationality/continental identity is assigned to `a person', or to `a person from Asia'. Using vision-language model CLIP's cosine similarity to compare images generated by CLIP-based Stable Diffusion v2.1 verified by manual examination, we chronicle results from 136 prompts (50 results/prompt) of front-facing images of persons from 6 different continents, 27 nationalities and 3 genders. We observe how Stable Diffusion outputs of `a person' without any additional gender/nationality information correspond closest to images of men and least with persons of nonbinary gender, and to persons from Europe/North America over Africa/Asia, pointing towards Stable Diffusion having a concerning representation of personhood to be a European/North American man. We also show continental stereotypes and resultant harms e.g. a person from Oceania is deemed to be Australian/New Zealander over Papua New Guinean, pointing to the erasure of Indigenous Oceanic peoples, who form a majority over descendants of colonizers both in Papua New Guinea and in Oceania overall. Finally, we unexpectedly observe a pattern of oversexualization of women, specifically Latin American, Mexican, Indian and Egyptian women relative to other nationalities, measured through an NSFW detector. This demonstrates how Stable Diffusion perpetuates Western fetishization of women of color through objectification in media, which if left unchecked will amplify this stereotypical representation. Image datasets are made publicly available.

Key facts. Cited by 33 papers in the map (data through 7 Oct 2026). Placed in Image generation, text-to-image and editing, one of 14 directions in Diffusion models. Builds on 13 of its 36 references that are in the map. First posted on arXiv 30 Oct 2023.

Builds on

13 of 36 references.

Closest ancestors · 1

Same direction, newest first

  1. 2023Multimodal Composite Association Score: Measuring Gender Bias in Generative Multimodal Models26 Apr

Other references · 12

  1. 2020Language Models are Few-Shot Learnerscited by 32,199
  2. 2021High-Resolution Image Synthesis with Latent Diffusion Modelscited by 18,694
  3. 2022LAION-5B: An open large-scale dataset for training next generation image-text modelscited by 4,131
  4. 2016Semantics derived automatically from language corpora contain human-like biasescited by 894
  5. 2021Multimodal datasets: misogyny, pornography, and malignant stereotypescited by 326
  6. 2023Stable Bias: Analyzing Societal Representations in Diffusion Modelscited by 225
  7. 2015Generating Images from Captions with Attentioncited by 135
  8. 2022Contrastive Language-Vision AI Models Pretrained on Web-Scraped Multimodal Data Exhibit Sexual Objectification Biascited by 43
  9. 2022American == White in Multimodal Language-and-Image AIcited by 39
  10. 2023T2IAT: Measuring Valence and Stereotypical Biases in Text-to-Image Generationcited by 39
  11. 2023A Friendly Face: Do Text-to-Image Systems Rely on Stereotypes when the Input is Under-Specified?cited by 25
  12. 2020Gender Classification and Bias Mitigation in Facial Imagescited by 8

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2023: 1; 2024: 9; 2025: 14; 2026: 9.

33 papers.

Sort:

By area:

  1. 2024Mapping the Increasing Use of LLMs in Scientific Paperscited by 112
  2. 2023TIBET: Identifying and Evaluating Biases in Text-to-Image Generative Modelscited by 51
  3. 2024Gender, Race, and Intersectional Bias in Resume Screening via Language Model Retrievalcited by 49
  4. 2024ViSAGe: A Global-Scale Analysis of Visual Stereotypes in Text-to-Image Generationcited by 27
  5. 2024AI-generated faces influence gender stereotypes and racial homogenizationcited by 13
  6. 2024From Melting Pots to Misrepresentations: Exploring Harms in Generative AIcited by 13
  7. 2024BiasDora: Exploring Hidden Biased Associations in Vision-Language Modelscited by 12
  8. 2025A Large Scale Analysis of Gender Biases in Text-to-Image Generative Modelscited by 12
  9. 2024A Unified Framework and Dataset for Assessing Societal Bias in Vision-Language Modelscited by 11
  10. 2024Image-Perfect Imperfections: Safety, Bias, and Authenticity in the Shadow of Text-To-Image Model Evolutioncited by 7
  11. 2025Un-Straightening Generative AI: How Queer Artists Surface and Challenge the Normativity of Generative AI Modelscited by 6
  12. 2025POET: Supporting Prompting Creativity and Personalization with Automated Expansion of Text-to-Image Generationcited by 6
21 more
  1. 2025Biases Propagate in Encoder-based Vision-Language Models: A Systematic Analysis From Intrinsic Measures to Zero-shot Retrieval Outcomescited by 6
  2. 2025Bias Amplification in Stable Diffusion's Representation of Stigma Through Skin Tones and Their Homogeneitycited by 5
  3. 2025Unveiling the "Fairness Seesaw": Discovering and Mitigating Gender and Race Bias in Vision-Language Modelscited by 4
  4. 2025Person-Centric Annotations of LAION-400M: Auditing Bias and Its Transfer to Modelscited by 4
  5. 2025DebiasPI: Inference-time Debiasing by Prompt Iteration of a Text-to-Image Generative Modelcited by 3
  6. 2025BiasConnect: Investigating Bias Interactions in Text-to-Image Modelscited by 2
  7. 2026Who Gets Flagged? The Pluralistic Evaluation Gap in AI Content Watermarkingcited by 2
  8. 2026What if AI systems weren't chatbots?cited by 2
  9. 2024AITTI: Learning Adaptive Inclusive Token for Text-to-Image Generationcited by 1
  10. 2025Discovering Divergent Representations between Text-to-Image Modelscited by 1
  11. 2026Neutral Prompts, Non-Neutral People: Quantifying Gender and Skin-Tone Bias in Gemini Flash 2.5 Image and GPT Image 1.5cited by 1
  12. 2026Who Defines Fairness? Target-Based Prompting for Demographic Representation in Generative Modelscited by 1
  13. 2026Prompt Revision as a Source of Cultural Bias in Text-to-Image Systemscited by 1
  14. 2025FairT2I: Mitigating Social Bias in Text-to-Image Generation via Large Language Model-Assisted Detection and Attribute Rebalancingcited by 0
  15. 2025From job titles to jawlines: Using context voids to study generative AI systemscited by 0
  16. 2025When Cars Have Stereotypes: Auditing Demographic Bias in Objects from Text-to-Image Modelscited by 0
  17. 2025Designing and Evaluating Malinowski's Lens: An AI-Native Educational Game for Ethnographic Learningcited by 0
  18. 2026Compounding Disadvantage: Auditing Intersectional Bias in LLM-Generated Explanations Across Indian and American STEM Educationcited by 0
  19. 2026GLEaN: A Text-to-image Bias Detection Approach for Public Comprehensioncited by 0
  20. 2026Beyond Categories of Caste: Examining Caste Bias and Morality in Text-to-Image AI Modelscited by 0
  21. 2026Mod-Guide: An LLM-based Content Moderation Feedback System to Address Insensitive Speech toward Indigenous Ethnic and Religious Minority Communitiescited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2023Stable Diffusion Exposed: Gender Bias from Prompt to Image
  2. 2025Hidden Bias in the Machine: Stereotypes in Text-to-Image Models
  3. 2023RS-Corrector: Correcting the Racial Stereotypes in Latent Diffusion Models
  4. 2025Visual Cues of Gender and Race are Associated with Stereotyping in Vision-Language Models
  5. 2026Newer Is Not Fairer: Gender Stereotyping in Text-to-Image AI Across Model Generations
  6. 2025Cultural Awareness in Vision-Language Models: A Cross-Country Exploration
  7. 2025SCALEX: Scalable Concept and Latent Exploration for Diffusion Models
  8. 2023Finetuning Text-to-Image Diffusion Models for Fairness