Generating the Modal Worker: A Cross-Model Audit of Race and Gender in LLM-Generated Personas Across 41 Occupations
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866918411368398848 |
|---|---|
| author | van der Linden, Ilona Kumar, Sahana Dixit, Arnav Sudan, Aadi Danda, Smruthi Anastasiu, David C. Lukoff, Kai |
| author_facet | van der Linden, Ilona Kumar, Sahana Dixit, Arnav Sudan, Aadi Danda, Smruthi Anastasiu, David C. Lukoff, Kai |
| contents | As generative AI tools are increasingly used to portray people in professional roles, understanding their racial and gender representational biases is critical. We audit over 1.5 million occupational personas generated by four major large language models - GPT-4, Gemini 2.5, DeepSeek V3.1, and Mistral-medium - across 41 U.S. occupations. Comparing these personas against U.S. Bureau of Labor Statistics (BLS) data, we find that models generate demographics with less variation than real-world data, functionally compressing each occupation toward a dominant demographic profile rather than representing population-level variation. A shift/exaggeration decomposition reveals the structure of these distortions: White (-31pp) and Black (-9pp) workers are consistently underrepresented, while Hispanic (+17pp) and Asian (+12pp) workers are overrepresented, with stereotype exaggeration amplifying existing occupational segregation. These distortions are often extreme, including near-total portrayals of housekeepers as Hispanic and the near-erasure of Black workers from many occupations. Because these patterns recur across models with different institutional and cultural origins, they suggest shared structural sources of bias rather than model-specific artifacts. We argue that auditing generative AI requires evaluation frameworks that examine how synthetic populations systematically reshape demographic visibility across social roles. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2510_21011 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Generating the Modal Worker: A Cross-Model Audit of Race and Gender in LLM-Generated Personas Across 41 Occupations van der Linden, Ilona Kumar, Sahana Dixit, Arnav Sudan, Aadi Danda, Smruthi Anastasiu, David C. Lukoff, Kai Human-Computer Interaction Artificial Intelligence Computers and Society As generative AI tools are increasingly used to portray people in professional roles, understanding their racial and gender representational biases is critical. We audit over 1.5 million occupational personas generated by four major large language models - GPT-4, Gemini 2.5, DeepSeek V3.1, and Mistral-medium - across 41 U.S. occupations. Comparing these personas against U.S. Bureau of Labor Statistics (BLS) data, we find that models generate demographics with less variation than real-world data, functionally compressing each occupation toward a dominant demographic profile rather than representing population-level variation. A shift/exaggeration decomposition reveals the structure of these distortions: White (-31pp) and Black (-9pp) workers are consistently underrepresented, while Hispanic (+17pp) and Asian (+12pp) workers are overrepresented, with stereotype exaggeration amplifying existing occupational segregation. These distortions are often extreme, including near-total portrayals of housekeepers as Hispanic and the near-erasure of Black workers from many occupations. Because these patterns recur across models with different institutional and cultural origins, they suggest shared structural sources of bias rather than model-specific artifacts. We argue that auditing generative AI requires evaluation frameworks that examine how synthetic populations systematically reshape demographic visibility across social roles. |
| title | Generating the Modal Worker: A Cross-Model Audit of Race and Gender in LLM-Generated Personas Across 41 Occupations |
| topic | Human-Computer Interaction Artificial Intelligence Computers and Society |
| url | https://arxiv.org/abs/2510.21011 |