Presentation
AI vs. Human: A Framework for Responsible Qualitative Research in the Age of AI
DescriptionGenerative AI is rapidly making its way into UX research, with tools that promise to plan workshops, generate personas, analyze interviews, and even stand in for participants. But are these tools really equipped to deliver actionable insights, or just convincing illusions of them?
That was the question we posed in the AI vs. Human project, a comparative study that pitted the best of these AI tools against trained UX researchers in a variety of qualitative research tasks including drafting discussion guides, moderating interviews, and analyzing data. The study revealed successes on both sides - human and AI - but also surfaced a deeper, perhaps counterintuitive, pattern of findings that speaks to GenAI’s strengths and weaknesses. Most of us think of software as perfect for repetitive, rule-based tasks, but this mental model just doesn’t work for software built to predict, approximate, and create. In this session, I’ll share findings from the AI vs. Human project and offer a framework for using GenAI tools responsibly. By shifting our mental model - from rule-based precision to probabilistic prediction - we can better understand where AI adds value, where it falls short, and how to engage with it thoughtfully, without mistaking imitation for insight.
That was the question we posed in the AI vs. Human project, a comparative study that pitted the best of these AI tools against trained UX researchers in a variety of qualitative research tasks including drafting discussion guides, moderating interviews, and analyzing data. The study revealed successes on both sides - human and AI - but also surfaced a deeper, perhaps counterintuitive, pattern of findings that speaks to GenAI’s strengths and weaknesses. Most of us think of software as perfect for repetitive, rule-based tasks, but this mental model just doesn’t work for software built to predict, approximate, and create. In this session, I’ll share findings from the AI vs. Human project and offer a framework for using GenAI tools responsibly. By shifting our mental model - from rule-based precision to probabilistic prediction - we can better understand where AI adds value, where it falls short, and how to engage with it thoughtfully, without mistaking imitation for insight.
Contributor
Event Type
Special Session
UX Miniconference
TimeTuesday, October 14th10:40am - 11am CDT
LocationGrand E/F
User Experience
UX Mini-Conference