Designing for Human-AI Interaction: Comparing Intermittent, Continuous, and Proactive Interactions for a Music Application

Designing effective and user-centred interactions between humans and AI systems poses fundamental challenges. The behaviour of AI systems is complex and uncertain, making it difficult to envision and craft optimal user experiences. Improved frameworks are needed to guide the design of human-AI interaction. In this paper, we develop and evaluate prototypes for a music application, representing three distinct paradigms of human-AI interaction: Intermittent, Continuous, and Proactive. Through qualitative user interviews with 12 participants, we compare the user experience across these prototypes, shedding light on potential challenges and opportunities for the paradigms represented. We found that the three prototypes exhibit distinct characteristics in terms of supported goals and user control. This case study contributes to a deeper understanding of the complexities involved in designing AI systems and offers insights for the development of more user-centred AI applications.