AI-based user adaptive interaction

Context

During my master's at Kumoh National Institute of Technology I worked in a research project, funded by the National Research Foundation of Korea, on how people accept and collaborate with AI-infused systems — and how interaction can adapt to the person using it.

Challenge

Acceptance of AI is usually explained with behavioural models such as the technology acceptance model (TAM). Those models often overlook the technical characteristics of a system and the differences between users, even though both shape whether people trust and adopt AI.

My role

I led the literature review and the technology-acceptance studies as first author, and contributed as a co-author to an experiment on emotion in voice-based interaction.

Process

• Systematic review. I analysed 85 peer-reviewed journal articles on the factors that influence the adoption of AI devices, across areas such as autonomous driving, healthcare and robotics.
• Extended acceptance model. I proposed an extended TAM that adds technological factors and user differences, and tested it with an online survey of 81 residents of South Korea using structural equation modelling (18 of 25 hypotheses supported).
• Voice interaction. In a separate experiment, the team examined how human factors and voice design variables affect users' emotional response to a voice-based intelligent system.

Deliverables

• A systematic review published in Human Factors, later named a Clarivate ESI Highly Cited Paper (2025)
• The extended technology acceptance model and its survey studies, continued in a 2026 article with 509 respondents in Transportation Research Part F