BabaCare: Testing a Care Copilot
Johns Hopkins University
Summary
Baba is a novel AI "care copilot" designed to improve emotional well-being, cognitive stimulation, and social connection through emotionally present conversations, proactive engagement, and personalized communication. Unlike many existing solutions, Baba operates through SMS text and voice calls, requiring no apps, Wi-Fi, or new devices, making it particularly accessible to older adults.
Description
The purpose of this study is to evaluate the feasibility, usability, and preliminary impact of the Baba platform among older adults. This study will examine how older adults interact with Baba, how it integrates into the older adult's daily lives, and whether it contributes to improvements in well-being, cognitive engagement, and self-management. Findings will inform the feasibility of AI companions in supporting independent aging.
Eligibility
- Age range
- 65+ years
- Sex
- All
- Healthy volunteers
- Yes
Inclusion Criteria: * Lives alone * Responds "occasionally" or "most" to the exhaustion questionnaire: 1. "How often in the last week did you feel that everything you did was an effort?" 2. "How often in the last week did you feel that you could not get going?" * Willingness and ability to regularly engage with Baba for 12 weeks Exclusion Criteria: * Significant cognitive impairment that prevents consent or participation * Inability to communicate via SMS or phone * Severe hearing or speech impairment precluding use of voice interaction
Interventions
- OtherAI phone companion
Baba uses a fine-tuned large language model to hold conversations with older adults and analyze tone, sentiment shifts, and emotional cues. The system logs call summaries and risk flags for caregiver review and tracks changes over time.
Location
- Johns Hopkins Bayview CampusBaltimore, Maryland