Amazon One Medical launches agentic Health AI assistant for 24/7 guidance and care coordination
Amazon One Medical has rolled out a Health AI assistant inside the One Medical app. It's an agentic tool-meaning it doesn't just answer questions, it takes action. It can book appointments, read labs in context, and help manage medications, all tied to each member's medical history.
Built with One Medical's clinical leadership, the assistant provides always-on guidance and knows when to hand off to a human clinician. It connects members to their care team by message, video, or an in-person visit, often same or next day.
What Health AI can do today
- Answer general and complex health questions, explain lab results, and interpret them based on the member's history and current meds.
- Offer 24/7 guidance on symptoms, conditions, potential treatments, and wellness questions.
- Recommend the right care path: virtual visit, in-person appointment, or urgent care.
- Streamline tasks like booking with a One Medical provider and renewing medications (with the option to fill via Amazon Pharmacy).
- Support parents and guardians with advice and treatment options for kids' issues such as pink eye, skin rashes, head lice, and asthma Rx renewals.
Clinician-first by design
Amazon positions the assistant as support for the patient-clinician relationship, not a replacement. Safety guardrails and clinical protocols are embedded throughout, including triggers for emergency and sensitive scenarios.
"The U.S. health care experience is fragmented, with each provider seeing only parts of your health puzzle," said Neil Lindsay, senior vice president of Amazon Health Services. "Health AI in the One Medical app brings together all the pieces of your personal health information to give you a more complete picture-helping you understand your health, and supporting you in getting the care you need to get and stay well."
Dr. Andrew Diamond, chief medical officer at One Medical, added, "Our Health AI enhances this relationship by helping members understand their health information and manage their routine health tasks, coaching them to stick to their health program, and quickly connecting them to their trusted providers when they need the care and expertise of a human clinician."
Privacy and safety
The assistant operates under HIPAA-compliant privacy and security standards used by Amazon Health Services. More on HIPAA is available from HHS: HIPAA overview.
It recognizes symptoms or queries that warrant clinical judgment and moves the member to appropriate care. If in-person evaluation is needed, it recommends the setting and can make the appointment-virtual or in-person-often the same or next day.
Availability and access
Health AI is live for One Medical members in the One Medical app. Members can opt out and use the standard app by tapping "Home" on the bottom nav.
One Medical membership is optional for scheduled in-person or remote visits. Prime members can add membership for $9 per month or $99 per year, or purchase directly from One Medical.
Under the hood
The assistant is powered by models on Amazon Bedrock. A beta has been running with select members since early 2025, with ongoing feedback loops to refine accuracy, usefulness, and new features.
What this means for care teams
- Expect more structured pre-visit context: symptoms, history, and patient intent captured upfront.
- Administrative load may shift from manual scheduling and renewals to reviewing AI-prepared context and finalizing decisions.
- Escalation logic matters: understand triggers that send patients to messaging, immediate video, or in-person care.
- Medication changes still sit with clinicians; the assistant can streamline requests and coordination.
Practical questions to align on
- What are the clinical escalation criteria, and how are edge cases handled?
- How are audit trails logged for AI-driven recommendations and actions?
- Which data sources feed the assistant, and how is data accuracy validated?
- How will teams monitor impact on response times, no-show rates, and refill safety?
For healthcare teams building AI fluency and workflows, explore role-based learning paths here: AI courses by job.
Your membership also unlocks:
