Wearable AI coach

Google Health Coach (Gemini)

Google Health Coach describes adjustable goals, optional research participation, and controls for disabling the coach and deleting conversations. Enabling it grants broad account-data access. Disconnecting a feature and deleting information already used in conversations are separate actions.

AI-assisted draft Read full report Open directory

Published September 8, 2026 as an AI-assisted draft. The public report separates documented facts, agency judgments and unresolved questions.

66 /100 toward patient-directed

Agency posture Mixed, patient-chosen but ecosystem-gated

The question we ask Who does Google Health Coach (Gemini) serve in this deployment?

Control Consumer Google Health Coach in the Google Health app, powered by Gemini and requiring Premium. This is separate from the Gemini apps.

Agency read Mixed, patient-chosen but ecosystem-gated

Vendor: Google
Who it serves: Patient-directed consumer wellness AI, vendor-controlled and gated behind a device-plus-subscription ecosystem
Primary User: Consumers with Fitbit and Pixel Watch devices who subscribe to Google Health Premium, including people syncing U.S. medical records
Control Model: Vendor-hosted public service. Controls and downstream processors depend on the assessed deployment.
Patient Impact: Conversation and record context can help users prepare questions and interpret trends. Public pages do not establish consistent source citations for individual suggestions or correction of underlying medical records.
Profile Status: AI-assisted draft
Last assessed: Sep 8, 2026
Review Confidence: Moderate for documented controls and positioning. No account or coach interaction was tested. (AI-assisted draft)
AI / Model: OpenAI Codex / GPT-6
Human Review: Hugo Campos authorized publication of these AI-assisted draft reassessments on September 8, 2026. This does not claim comprehensive human verification of every finding.

Summary judgment · 66 out of 100 toward patient-directed

Consumer Google Health Coach in the Google Health app, powered by Gemini and requiring Premium. This is separate from the Gemini apps.

Mixed, patient-chosen but ecosystem-gated

Patient agency

How this tool changes agency

Expands agency when

Conversation and record context can help users prepare questions and interpret trends. Public pages do not establish consistent source citations for individual suggestions or correction of underlying medical records.

Limits agency when

Actual recommendation provenance, correction behavior, and the usability of separate deletion controls remain unverified.

Patient agency assessment

Who sets and changes the goal?

Patient authority

Users choose goals and can adjust plans or disable the coach. Broad account access limits selective permission once enabled. No current evidence establishes a compulsory treatment or adherence path.

What can the patient understand, question, or do?

Critical capacity

Can the patient evaluate the conditions of use?

Informed control

The help page explains data categories and deletion limits unusually specifically. Feature revocation and conversation deletion are separate actions. Optional research and feedback review should not be described as default training or unrestricted human review.

Text findings

Conditions of use

Published controls and their limits

What remains unknown?

Not tested or not established

Actual recommendation provenance, correction behavior, and the usability of separate deletion controls remain unverified.

Who evaluated this?

AI-assisted public-source draft

Vendor statements describe published conditions, not independently verified behavior. No clinical, security, accessibility, or legal validation is claimed. Earlier evidence remains dated in the report and history.

Sources checked

Source-specific findings and retrieval limitations are recorded in the full report.

Review provenance

Criteria

CAIHL-derived HugoScore framework and September 7 qualitative review priorities. Draft v1.2 numerical anchors remain unadopted.

Reviewer

AI-assisted public-source reassessment prepared in OpenAI Codex.

AI / model

OpenAI Codex / GPT-6

Human review

Hugo Campos authorized publication of these AI-assisted draft reassessments on September 8, 2026. This does not claim comprehensive human verification of every finding.

Review date

2026-09-08

Limitations

Actual recommendation provenance, correction behavior, and the usability of separate deletion controls remain unverified. No live product use, patient-data upload, account creation, code audit, clinical evaluation, or independent implementation validation.

Review method

Focused public-source reassessment using CAIHL: patient authority, critical capacity, and informed control. Existing evidence plus one focused primary-source pass and at most one targeted follow-up. No live product testing. Numerical scores remain provisional editorial placements, not a new calculation.

AI-assisted draft · Moderate for documented controls and positioning. No account or coach interaction was tested. (AI-assisted draft)