Microsoft has launched Copilot Health, a dedicated secure environment within its Copilot AI assistant that aggregates electronic health records, wearable device data from platforms including Apple Watch, Oura, and Fitbit, and lab results to generate personalized health insights. Announced by Mustafa Suleyman, CEO of Microsoft AI, the product enters a market that Microsoft's own internal research suggests is already substantial: nearly one in five Copilot conversations already involves assessment of a personal symptom or condition. The company frames Copilot Health as a wellness and appointment-preparation tool, not a clinical decision support system, with the launch post carrying an explicit disclaimer that the product "is not intended to diagnose, treat, or prevent diseases or other conditions and is not a substitute for professional medical advice."
The launch places Microsoft in direct competition with OpenAI, which announced ChatGPT Health in January 2026, and Anthropic, which followed shortly after with Claude for Healthcare. The near-simultaneous market entries by all three major AI players appear tied to a structural regulatory shift: the FDA revised its wearable device oversight rules at the start of 2026, creating a broader "non-device CDS" category that allows AI systems reasoning over wearable data to be offered to consumers without premarket FDA review, provided they avoid explicit diagnostic or treatment claims. Law firm Arnold & Porter noted in January that the revised policy "likely means that more AI-enabled CDS can be made available as non-device CDS, i.e., without FDA review." The legal carve-out effectively de-risked the precise product architecture all three companies are now shipping.
Suleyman pitched the service as reaching "billions of people around the world who struggle to access reliable medical advice" — a goal that sits uneasily beside the product's non-medical-advice disclaimers. The Register, covering the launch, cited a recent UK study finding that chatbots give poor medical advice, a finding that carries more weight given the scale at which these tools now operate: OpenAI has reported over 40 million people globally consulting ChatGPT for healthcare advice daily. Microsoft has pledged data isolation, encryption at rest and in transit, strict access controls, and a firm policy against using health data for model training. Whether <a href="/news/2026-03-14-microsoft-copilot-health-centralizes-personal-medical-records-outside-hipaa">those pledges hold</a> will be watched closely: the company spent much of 2024 defending its Recall feature, an AI screenshot logger that privacy advocates forced Microsoft to substantially retool before it shipped — a controversy that has yet to fully recede from public memory.