A Tale of Two Identities: An Ethical Audit of Human and AI-Crafted Personas
This addresses ethical risks in AI-generated personas for minority communities in domains like health and HCI, though it is incremental as it builds on existing audit methods.
The paper audited synthetic personas generated by three large language models (GPT4o, Gemini 1.5 Pro, Deepseek 2.5) compared to human-authored ones, finding that LLMs disproportionately foreground racial markers and produce culturally coded language, leading to harms like stereotyping and erasure, formalized as algorithmic othering.
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 2.5) 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 1512 LLM generated personas 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. Based on these findings, we offer design recommendations for narrative-aware evaluation metrics and community-centered validation protocols for synthetic identity generation.