Towards Considerate Embodied AI: Co-Designing Situated Multi-Site Healthcare Robots from Abstract Concepts to High-Fidelity Prototypes

TL;DR

This study addresses how to ground embodied AI in high-stakes healthcare by implementing a 14-week, multi-context co-design program that moves from abstract ideation to high-fidelity prototypes. It demonstrates that sustained cross-disciplinary collaboration, iterative prototyping, and structured educational scaffolding produce more deployable, context-aware robotic solutions across emergency, sleep, and long-term care settings. The authors distill eight guidelines organized into four embodied-AI design dimensions—needs grounding, feasibility, embodied literacy, and design-space expansion—to support considerate deployment. The work highlights how prototypes function as thinking tools, how workflow and social dynamics shape design, and the importance of cross-context learning for scalable, acceptable healthcare robotics.

Abstract

Co-design is essential for grounding embodied artificial intelligence (AI) systems in real-world contexts, especially high-stakes domains such as healthcare. While prior work has explored multidisciplinary collaboration, iterative prototyping, and support for non-technical participants, few have interwoven these into a sustained co-design process. Such efforts often target one context and low-fidelity stages, limiting the generalizability of findings and obscuring how participants' ideas evolve. To address these limitations, we conducted a 14-week workshop with a multidisciplinary team of 22 participants, centered around how embodied AI can reduce non-value-added task burdens in three healthcare settings: emergency departments, long-term rehabilitation facilities, and sleep disorder clinics. We found that the iterative progression from abstract brainstorming to high-fidelity prototypes, supported by educational scaffolds, enabled participants to understand real-world trade-offs and generate more deployable solutions. We propose eight guidelines for co-designing more considerate embodied AI: attuned to context, responsive to social dynamics, mindful of expectations, and grounded in deployment. Project Page: https://byc-sophie.github.io/Towards-Considerate-Embodied-AI/

Figures (3)

• Figure 1: Overview of the workshop. Weekly activities spanned diverse formats, including educational sessions (E), need identification (N), milestone activities (M) (e.g., storyboard, cardboard prototype, full-scale prototype), and a public showcase (S). Colored tags on the top-left corners of each activity card indicate the key activity types. Additionally, we annotated each week with relevant data collection types, including interviews (I), artifact analysis (A), and post-educational-session discussions (D), to highlight the integration of research through co-creation.
• Figure 2: An overview of how long-term engagement, researcher involvement, and healthcare expertise were operationalized during the study: (1) four participant-retention strategies that supported continuity, re-engagement, and sustained momentum across sessions; (2) four roles enacted by the research team (i.e., facilitator, consultant, data collector, and co-learner) that structured involvement and maintained appropriate boundaries; and (3) the pre-design, during-design, and post-design integration of healthcare domain knowledge that guided design reasoning and informed prototype refinement.
• Figure 3: Evolution of co-ideation across three stages: storyboards illustrating robot-assisted scenarios in ED (top), SDC (middle), and LTR (bottom); $\rightarrow$ Cardboard prototypes elaborating each context's envisioned robot roles, user groups, capabilities, and interaction modalities for ED (top), SDC (middle), and LTR (bottom); $\rightarrow$ full-scale prototypes developed by participants for ED (left), SDC (middle), and LTR (right). Comparison of robot roles across healthcare settings: 1) Delivery roles were envisioned in ED, SDC and LTR (shared, with distinct purposes); 2) Storage roles appeared in ED, SDC and LTR (shared, with distinct purposes); 3) Tour guide roles emerged in ED, SDC and LTR (shared but contextually distinct with different levels of emphasis); 4) Comfort/entertainment roles were unique to LTR.