This contribution presents a playful yet critical provocation, exploring how Large Language Models (LLMs) can generate synthetic personas inspired by the spirit of Barbie Land—a utopian world where empowerment, inclusivity, and imagination are central themes. Grounded in feminist and queer perspectives, the speculative approach emphasizes the value of fictional narratives as sources of inspiration for researchers, framing LLM-generated synthetic personas as tools for fostering critical dialogue, challenging systemic inequities, and advancing equity-driven design in research and practice. As part of on-going research, this contribution shows the initial findings of a multi-faceted speculative investigation, revealing biases present in AI-generated synthetic personas, and examining pathways for harnessing their creative potential to support more inclusive and socially conscious technological futures.
Synthetic Dreams in Barbie Land: Speculative Queer Adventures with Feminist LLM-Generated Personas / Haxvig, H. A.; D'Andrea, V.; Teli, M.. - (2025), pp. 379-385. ( 2025 ACM Conference on Designing Interactive Systems, DIS 2025 Companion prt 2025) [10.1145/3715668.3736361].
Synthetic Dreams in Barbie Land: Speculative Queer Adventures with Feminist LLM-Generated Personas
Haxvig H. A.;D'Andrea V.;Teli M.
2025-01-01
Abstract
This contribution presents a playful yet critical provocation, exploring how Large Language Models (LLMs) can generate synthetic personas inspired by the spirit of Barbie Land—a utopian world where empowerment, inclusivity, and imagination are central themes. Grounded in feminist and queer perspectives, the speculative approach emphasizes the value of fictional narratives as sources of inspiration for researchers, framing LLM-generated synthetic personas as tools for fostering critical dialogue, challenging systemic inequities, and advancing equity-driven design in research and practice. As part of on-going research, this contribution shows the initial findings of a multi-faceted speculative investigation, revealing biases present in AI-generated synthetic personas, and examining pathways for harnessing their creative potential to support more inclusive and socially conscious technological futures.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione



