Fitbit AI Coach will read your medical records next month as Google pushes deeper into healthcare

Fitbit Integrates Medical Records with AI Coach

Google is expanding the capabilities of Fitbit's AI-powered health coach by allowing users to connect their medical records directly to the app starting next month 1

1

. The AI Coach will access lab results and medications, visit history, and other clinical data to provide personalized health guidance that considers both medical history and wearable data collected from Fitbit devices 2

2

. This feature, powered by Google's Gemini AI, aims to contextualize health questions with actual medical information rather than offering generic suggestions 4

4

.

The health data integration represents a significant shift in how consumer wearables interact with clinical information. Google is partnering with b.well Connected Health and Clear platforms to facilitate secure connections to healthcare providers 2

2

. Users will need to verify their identity using a selfie and valid ID to access their records through these partner platforms 3

3

. When users ask about specific health concerns like cholesterol levels, the AI Coach can summarize lab trends, highlight notable values, and provide wellness information based on both medical records and lifestyle patterns tracked by the device 1

1

.

Google's AI Plans Include Improved Sleep Tracking

Before the medical records feature arrives, Fitbit is rolling out significant improvements to sleep tracking with a reported 15% increase in sleep staging accuracy for Public Preview users 4

4

. The improved sleep tracking uses refined models trained on diverse datasets to better distinguish between attempting to sleep and actually sleeping 5

5

.

These models more accurately capture interruptions, naps, and transitions between sleep stages, aligning with clinical gold-standard measurements 5

5

.

The redesigned Sleep Score now evaluates specific aspects of rest, including how long it takes to fall asleep rather than focusing solely on total sleep time 4

4

. This level of detail enables the AI Coach to provide targeted coaching, helping users understand exactly where to focus their routine for better recovery 5

5

. The enhanced tracking is rolling out now to Pixel Watch and Fitbit devices, with the new Sleep Score experience expected in the coming weeks 2

2

.

Data Privacy Concerns and AI Limitations

The announcement raises questions about data privacy, particularly given Google's ownership of Fitbit since 2019 3

3

. Google states that medical records are securely stored within Fitbit and users maintain control over how their data is used, shared, or deleted 1

1

. The company emphasizes that medical records, like other health data in Fitbit, are not used for ads 2

2

. However, Google did not specify whether this data will be stored on-device or within Fitbit's servers 1

1

.

The integration of large language models with medical information presents risks, as chatbots have a tendency to hallucinate or generate inaccurate information 1

1

. Google acknowledges these limitations and says it invests heavily in validation processes to enhance model quality through ongoing evaluations and advanced architectures that rely on tools and self-critique to improve reliability 1

1

. When conversations venture too far into medical territory, the AI Coach will remind users to consult with healthcare professionals for medical needs, indicating it won't provide medical advice, diagnoses, or treatment plans 1

1

.

Continuous Glucose Monitor and Virtual Care Integration

Google is expanding health data integration beyond medical records. Starting next month in Public Preview, users will be able to connect a continuous glucose monitor through Health Connect and ask the AI Coach how specific meals or workouts impact glucose levels 1

1

. This feature allows users to query their wearable data alongside real-time glucose monitoring for more comprehensive health insights 4

4

.

Fitbit is also launching a "Get care now" research study in partnership with Included Health to assess how conversational AI can help users navigate virtual care during healthcare visits 5

5

. This research builds on Fitbit's pioneering study on predicting insulin resistance using wearable data, recently published in Nature, which explores how AI can better understand cardiometabolic health and enable early awareness 5

5

. The direction is clear: Fitbit is building a system that combines clinical data with wearable data to deliver more relevant personalized health guidance in daily decisions 4

4

.