From Efficiency to Meaning: Adolescents' Envisioned Role of AI in Health Management

While prior research has focused on providers, caregivers, and adult patients, little is known about adolescents’ perceptions of AI in health learning and management. Utilizing design fiction and co-design methods, we conducted seven workshops with 23 adolescents (aged 14-17) to understand how they anticipate using health AI in the context of a family celiac diagnosis. Our findings reveal that adolescents have four main envisioned roles of health AI: enhancing health understanding and help-seeking, reducing cognitive burden, supporting family health management, and providing guidance while respecting their autonomy. We also identified nuanced trust and a divided view on emotional support from health AI. These findings suggest that adolescents perceive AI’s value as a tool that moves them from efficiency to meaning-one that creates time for valued activities. We discuss opportunities for future health AI systems to be designed to encourage adolescent autonomy and reflection, while also supporting meaningful, dialectical activities.

💡 Research Summary

This paper investigates how adolescents envision the role of artificial intelligence (AI) in personal health learning and management, an area that has been largely overlooked in favor of research on clinicians, caregivers, and adult patients. Using a combination of design fiction and co‑design methods, the authors conducted seven workshops with 23 participants aged 14‑17, focusing on a family scenario involving a celiac disease diagnosis. The interdisciplinary team—including HCI scholars and pediatric physicians—crafted progressive fictional scenarios that prompted participants to imagine concrete AI interactions within their daily lives and family health routines.

Four primary roles for health‑focused AI emerged from the data. First, adolescents expect AI to enhance health understanding and facilitate help‑seeking by translating complex medical information into accessible language and connecting them to professional resources when needed. Second, they see AI as a means to reduce cognitive burden, automating routine tasks such as symptom logging, dietary tracking, and medication reminders, thereby freeing mental bandwidth for school, social activities, and personal interests. Third, participants envision AI supporting collaborative family health management, acting as a bridge that enables smoother role transitions between parents and teens, supports data sharing, and helps coordinate decisions across family members. Fourth, they value AI that respects autonomy: rather than making decisions for them, AI should provide guided options, encourage reflection, and reinforce self‑efficacy.

Trust and emotional support emerged as nuanced concerns. While adolescents expressed confidence in AI that offers transparent, evidence‑based explanations and personalized feedback, opinions diverged on whether AI should provide emotional comfort. Some participants welcomed a non‑judgmental conversational partner for stress relief, whereas others warned that over‑reliance on AI for empathy could diminish human interaction and create privacy or safety risks. This split suggests that designers should treat emotional‑support features as optional or clearly delineate when human professional involvement is required.

A central conceptual contribution is the shift from “efficiency” to “meaning.” Adolescents view AI not merely as a time‑saving tool but as a catalyst that reallocates saved time toward intrinsically valuable activities—family bonding, hobbies, and self‑development. Consequently, user‑interface design should prioritize meaning‑oriented feedback and visualizations that make the benefits of efficiency explicit in terms of life‑quality gains, rather than focusing solely on speed or task completion.

The study situates its findings within broader HCI literature on adolescent health technologies, highlighting how AI can extend existing support for information seeking, self‑management, and family informatics while addressing developmental needs for autonomy and privacy. Limitations include a modest sample size, geographic concentration, and a focus on a single chronic condition, which may constrain generalizability. Nonetheless, the rich qualitative insights provide a valuable foundation for future research, policy, and design guidelines aimed at creating adolescent‑centric health AI that balances efficiency, trust, autonomy, and meaningful engagement.