This Conversation at DRS2024 in Boston, attracting around 30 participants, centered on the evolving role of design within multidisciplinary AI teams, particularly in the context of the development of clinical AI applications. As AI is entering healthcare, questions are being raised if and how designers can contribute to AI-driven clinical solutions and whether they need to develop potential other skills and responsibilities. Drawing from ongoing research and insights from various workshops and interviews, the conversation highlighted the importance of the negotiation of agency between humans and machines in clinical settings, translational design for patients, data hierarchy and its impact on design, and finally the importance of language and storytelling in framing interactions mediated by AI.

Digital health technologies, powered by digital data, provide an opportunity to improve the efficacy and efficiency of health systems at large. However, little is known about different approaches to the use of data for digital health design, or about their possible relations to system-level dynamics. In this contribution, we identify four existing approaches to the use of data for digital health design, namely the silent, the overt, the data-enabled, and the convergent. After characterising the approaches, we provide real-life examples of each. Furthermore, we compare the approaches in terms of selected desirable characteristics of the design process, highlighting relative advantages and disadvantages. Finally, we reflect on the system-level relevance of the differentiation between the approaches and point towards future research directions. Overall, the contribution provides researchers and practitioners with a broad conceptual framework to examine data-related challenges and opportunities in digital health design.