AI-Care: A Conversational Agentic System for Task Coordination in Alzheimer's Disease Care
Authors: Preyash Yadav, Michelle Cohn, Priyanka Koppolu et al.
Summary
The authors built a voice-first chatbot for people with Alzheimer's to manage calendars and to-do lists through natural conversation, layered on top of a caregiving platform. The system uses a state-machine approach (LangGraph) that sanitizes input, classifies intent, loads context, checks safety constraints, collects missing information through follow-up questions, executes actions, and composes responses. Safety-critical facts like medications come from caregiver-verified records, not model generation, and the system never makes autonomous medical decisions. A pilot with four users showed they found it trustworthy and could complete coordination tasks by talking.
Main takeaways:
- Designed to reduce cognitive load for people with memory/thinking impairments by handling multi-step tasks (calendar, reminders, lists) through simple conversation
- Uses a deterministic state machine for orchestration: every request goes through sanitization, intent classification, context loading, safety checks, slot-filling dialogue, tool execution, and response generation
- Safety-critical information (medications, allergies) is retrieved from caregiver-verified records rather than generated by the language model
- Handles ambiguous or incomplete requests through controlled multi-turn clarification instead of guessing or failing silently
- Preliminary user testing (n=4, mild-to-moderate Alzheimer's) found the system trustworthy, competent, and usable for coordination tasks
Relevance
Not directly related to my persona/midtraining work—this is an application paper about agentic tool use for a vulnerable population, not about how behaviors are installed or what conditional triggers look like inside the model.
Threat model
Potential threat/caveat for clean result "Fine-tuning one persona on a two-marker chunk and another on the start marker plants the end marker at every donor answer's end, not chained to the start (LOW confidence)": this item discusses failure, evaluation.
Abstract
arXiv:2605.08480v1 Announce Type: new Abstract: Individuals with Alzheimer's disease (AD) and Alzheimer's disease-related dementia (ADRD) experience memory and thinking changes that impact their ability to use digital daily management tools. For example, adding an event to a digital calendar requires multiple steps that may act as barriers to independent use for individuals with AD/ADRD. This paper presents AI-Care, a conversational agentic artificial intelligence (AI) layer built on top of a remote caregiving platform co-designed with people with AD/ADRD. AI-Care is designed to reduce the cognitive load on individuals with AD/ADRD when managing everyday tasks such as setting calendar reminders and organizing to-do lists through natural-language interaction with a voice-first chatbot. The system uses a LangGraph-based stateful orchestration approach in which each request passes through sanitization, intent classification, context loading, safety checks, deterministic slot collection, tool execution, and response composition. Safety-critical responses, particularly around medications and allergies, are grounded in caregiver-verified records rather than free-form model generation. The system does not make autonomous medical or treatment decisions. Incomplete or ambiguous requests are handled through controlled multi-turn clarification rather than silent failure or guessing. The system supports both typed and spoken input, with voice output through ElevenLabs text-to-speech. Longer responses are chunked before synthesis to avoid rushed playback. A preliminary pilot with four individuals with mild-to-moderate AD/ADRD showed that users found the system trustworthy, competent, and likable, and were able to complete the evaluated coordination tasks through conversation. We describe the design goals, system architecture, safety controls, and findings from this formative evaluation.