AI 기반 사전 의료 계획 대리인 연구

📝 Abstract

Loss of decisional capacity, coupled with the increasing absence of reliable human proxies, raises urgent questions about how individuals’ values can be represented in Advance Care Planning (ACP). To probe this fraught design space of high-risk, high-subjectivity decision support, we built an experience prototype (ACPAgent) and asked 15 participants in 4 workshops to train it to be their personal ACP proxy. We analysed their coping strategies and feature requests and mapped the results onto axes of agent autonomy and human control. Our findings show a surprising 86.7% agreement with ACPAgent, arguing for a potential new role of AI in ACP where agents act as personal advocates for individuals, building mutual intelligibility over time. We propose that the key areas of future risk that must be addressed are the moderation of users’ expectations and designing accountability and oversight over agent deployment and cutoffs.

💡 Analysis

Loss of decisional capacity, coupled with the increasing absence of reliable human proxies, raises urgent questions about how individuals’ values can be represented in Advance Care Planning (ACP). To probe this fraught design space of high-risk, high-subjectivity decision support, we built an experience prototype (ACPAgent) and asked 15 participants in 4 workshops to train it to be their personal ACP proxy. We analysed their coping strategies and feature requests and mapped the results onto axes of agent autonomy and human control. Our findings show a surprising 86.7% agreement with ACPAgent, arguing for a potential new role of AI in ACP where agents act as personal advocates for individuals, building mutual intelligibility over time. We propose that the key areas of future risk that must be addressed are the moderation of users’ expectations and designing accountability and oversight over agent deployment and cutoffs.

📄 Content

Words to Describe What I’m Feeling: Exploring the Potential of AI Agents for High Subjectivity Decisions in Advance Care Planning Kellie Yu Hui Sim kellie_sim@mymail.sutd.edu.sg Singapore University of Technology and Design Singapore, Singapore Pin Sym Foong pinsym@nus.edu.sg Telehealth Core National University of Singapore Singapore, Singapore Chenyu Zhao cinderella.zchenyu0912@gmail.com Singapore University of Technology and Design Singapore, Singapore Melanie Yi Ning Quek melanieqyn@gmail.com National University of Singapore Singapore, Singapore Swarangi Subodh Mehta swarangi_mehta@sutd.edu.sg Singapore University of Technology and Design Singapore, Singapore Kenny Tsu Wei Choo kenny_choo@sutd.edu.sg Singapore University of Technology and Design Singapore, Singapore Figure 1: Summary of current challenges, study design and findings. Abstract Loss of decisional capacity, coupled with the increasing absence of reliable human proxies, raises urgent questions about how individ- uals’ values can be represented in Advance Care Planning (ACP). To probe this fraught design space of high-risk, high-subjectivity decision support, we built an experience prototype (ACPAgent) and asked 15 participants in 4 workshops to train it to be their per- sonal ACP proxy. We analysed their coping strategies and feature requests and mapped the results onto axes of agent autonomy and human control. Our findings show a surprising 86.7% agreement with ACPAgent, arguing for a potential new role of AI in ACP where agents act as personal advocates for individuals, building mutual intelligibility over time. We propose that the key areas of future risk that must be addressed are the moderation of users’ expectations and designing accountability and oversight over agent deployment and cutoffs. This work is licensed under a Creative Commons Attribution 4.0 International License. CHI ’26, Barcelona, Spain © 2026 Copyright held by the owner/author(s). ACM ISBN 979-8-4007-2278-3/2026/04 https://doi.org/10.1145/3772318.3791335 CCS Concepts • Human-centered computing →Empirical studies in HCI; • Applied computing →Consumer health. Keywords Advance Care Planning, Large Language Models, Proxy Decision- Making, Advocates, Delegation ACM Reference Format: Kellie Yu Hui Sim, Pin Sym Foong, Chenyu Zhao, Melanie Yi Ning Quek, Swarangi Subodh Mehta, and Kenny Tsu Wei Choo. 2026. Words to De- scribe What I’m Feeling: Exploring the Potential of AI Agents for High Subjectivity Decisions in Advance Care Planning. In Proceedings of the 2026 CHI Conference on Human Factors in Computing Systems (CHI ’26), April 13–17, 2026, Barcelona, Spain. ACM, New York, NY, USA, 34 pages. https://doi.org/10.1145/3772318.3791335 1 Introduction Societies worldwide are rapidly ageing. Longer lifespans and declin- ing fertility have made older adults a fast-growing segment of the population, placing new demands on healthcare, caregiving, and decision-making systems [52, 54, 56]. A central challenge arises when older adults lose decisional capacity: who should speak for them? Traditionally, family caregivers have assumed this proxy role. Yet demographic shifts have shrunk the caregiver pool, in- creasing the likelihood that older adults have no proxy or must arXiv:2512.11276v2 [cs.HC] 22 Jan 2026 CHI ’26, April 13–17, 2026, Barcelona, Spain Sim et al. rely on unfamiliar strangers, such as public guardians, for critical medical decisions [58, 72, 74]. One widely promoted way to give patients a voice before they lose capacity [41, 50] has been Advance Care Planning (ACP), in which individuals articulate their preferences for future end-of-life scenarios [2]. Yet completing an ACP is far from straightforward. The decisions involved–such as whether to undergo resuscitation– are high-risk (involving life and death) and highly subjective (rooted in personal values and priorities) [12]. Although ACP is often described in strategy and policy research as an ongoing, conversation-centred process [22], in practice it often becomes a once-off, documentation-focused event, with lim- ited revisiting or revision over time [40]. Several process issues contribute: ACP documents are hard to retrieve and share across care settings, making iterative updating impractical. For example, clinicians frequently cannot access forms completed elsewhere, creating barriers to ongoing review [36]. Public awareness is also limited–many do not know ACP documents can be updated [48], and often revise them only during acute threats such as COVID- 19 [39]. Together, these factors render ACP documents into static commitments made for unpredictable futures, raising questions about their relevance for evolving clinical circumstances, health conditions, and personal priorities. Advances in artificial intelligence (AI) have prompted interest in their role in complex, high-risk judgements. AI systems now excel at handling large datasets and generating predictions, with applications in domains such as autonomous driv

This content is AI-processed based on ArXiv data.