A Tale of Two Identities: An Ethical Audit of AI-Crafted Synthetic Personas
Proceedings of the AAAI Conference on Artificial Intelligence · 14.03.2026 · 40 (44) · 37765-37774
DOI: 10.1609/aaai.v40i44.41112 ↗Kurz gefasst
Fachartikel in der Zeitschrift Proceedings of the AAAI Conference on Artificial Intelligence, online erschienen am 14.03.2026. Verfasst von Pranav Narayanan Venkit und 4 weiteren Autor:innen. Die Arbeit wurde bisher 1-mal zitiert (Crossref).
Abstract (Originalsprache)
As LLMs (large language models) are increasingly used to generate synthetic personas, particularly in data-limited domains such as health, privacy, and HCI, it becomes necessary to understand how these narratives represent identity, especially that of minority communities. In this paper, we audit synthetic personas generated by 3 LLMs (GPT4o, Gemini 1.5 Pro, Deepseek v2.5) through the lens of representational harm, focusing specifically on racial identity. Using a mixed-methods approach combining close reading, lexical analysis, and a parameterized creativity framework, we compare 1,512 LLM-generated persona to human-authored responses. Our findings reveal that LLMs disproportionately foreground racial markers, overproduce culturally coded language, and construct personas that are syntactically elaborate yet narratively reductive. These patterns result in a range of sociotechnical harms, including stereotyping, exoticism, erasure, and benevolent bias, that are often obfuscated by superficially positive narrations. We formalize this phenomenon as algorithmic othering, where minoritized identities are rendered hypervisible but less authentic.
Zitieren
APA
Venkit, P. N., Li, J., Zhou, Y., Rajtmajer, S., & Wilson, S. (2026). A Tale of Two Identities: An Ethical Audit of AI-Crafted Synthetic Personas. Proceedings of the AAAI Conference on Artificial Intelligence, 40(44), 37765-37774. https://doi.org/10.1609/aaai.v40i44.41112
Vancouver
Venkit PN, Li J, Zhou Y, Rajtmajer S, Wilson S. A Tale of Two Identities: An Ethical Audit of AI-Crafted Synthetic Personas. Proceedings of the AAAI Conference on Artificial Intelligence. 2026;40(44):37765-37774. doi:10.1609/aaai.v40i44.41112
BibTeX
@article{Venkit2026,
author = {Venkit, Pranav Narayanan and Li, Jiayi and Zhou, Yingfan and Rajtmajer, Sarah and Wilson, Shomir},
title = {A Tale of Two Identities: An Ethical Audit of AI-Crafted Synthetic Personas},
journal = {Proceedings of the AAAI Conference on Artificial Intelligence},
year = {2026},
volume = {40},
number = {44},
pages = {37765--37774},
doi = {10.1609/aaai.v40i44.41112}
}
EUVIMED