Investigating Design Elements of UX Personas about AI Collaboration Apps using Large Language Models

Investigating Design Elements of User Personas about Apps for Human-AI Collaborative App Design

Abstract

This study proposes a method for constructing a database of UX Personas for AI applications using a Large Language Model (LLM). UX Personas are fictional representations of application users that are essential for Human-AI collaborative design. The proposed method combines automatic extraction (machine annotation) with manual correction (human curation) to enhance the quality of the output data. An LLM could extract information about Job, Age, Pain Points, and Backstory of UX Personas from a website. Our results will contribute to the understanding of designing efficient Human-AI collaboration.