Getty Images
Microsoft joins the AI chatbot market with Copilot Health
Microsoft is framing Copilot Health as a complement to patient–provider relationships, but the company's own data suggest it's filling in where care access falters.
Microsoft is getting in on the consumer-facing health AI chatbot game, unveiling its Copilot Health product that the company says can give personalized health insights to patients.
The tool, which Microsoft said is secure but did not say was HIPAA-compliant, can take in an individual's medical records, wearable data and health history to provide contextualized input about the user's health.
At launch, Copilot Health can integrate information from more than 50 wearable devices, including Apple Health, Oura and Fitbit. Copilot Health can also integrate health records from over 50,000 U.S. hospitals and provider organizations through HealthEx and can integrate lab results from Function.
Microsoft said its tool will be able to take all that personal health data and make sense of it using "increasingly sophisticated AI," giving users more actionable insights. Information communicated via Copilot Health will include clear citations and links to source material.
To address medical misinformation concerns, the company said it will fortify the quality and reliability of its AI chatbot's answers by elevating health information from "credible health organizations." That information will be verified by a clinical team following National Academy of Medicine principles.
Microsoft is pitching Copilot Health as a tool that works alongside the patient–provider relationships many users already have.
"Copilot Health doesn't replace your doctor," the company wrote in its press announcement. "It makes every minute you have with them count more. You arrive prepared, with the right questions, the right context and the confidence that comes from better understanding your own body."
Still, many technology companies, Microsoft included, are touting AI chatbots as the key to filling patient engagement gaps.
Data has shown that patients are accessing AI chatbots for their healthcare because they face barriers to traditional healthcare. AI chatbots have stepped in when patients can't navigate an opaque system, can't afford a doctor, face long travel times or need advice outside of typical clinic hours.
Is Copilot Health poised to fill medicine's patient engagement lapses?
The Copilot Health announcement comes on the heels of a Microsoft report examining how people use Copilot in healthcare. The report argued that consumer-facing AI chatbots are stepping in where the traditional healthcare system can't meet patient demands.
The report, based on a sample of 500,000 conversations people had with Copilot, builds on Microsoft's 2025 Copilot Usage Report, which looked at trends in how people use Copilot across all use cases. Most notably, the report found that health and fitness searches were the top use case for Copilot throughout the year.
In this most recent report, Microsoft researchers analyzed healthcare-related conversations on Copilot, finding that people are using AI to engage with their health when information is otherwise not accessible.
What do people ask Copilot?
By and large, people are querying Copilot for health information and to learn more about healthcare or a specific symptom, with around 40% of chats falling into this category. The report also flagged a 10.9% share of messages related to symptoms or health concerns that users explicitly said they experience.
"Questions framed in general terms may well reflect a user's own health concern rather than casual curiosity, and the true share of personal health questions may be higher," Microsoft added as a caveat.
Another 9% of users chatted with Copilot about lifestyle and fitness coaching, while 5.8% used the tool for healthcare navigation and explanations of insurance benefits.
These trends suggest that health information or advice isn't often readily available for patients. Meeting with a doctor upon the onset of new symptoms can be challenging, as the average appointment wait time in the U.S. is around 31 days. When an individual or their loved one experiences new symptoms, it might be easier to chat with an AI bot to get some peace of mind.
Similarly, in an opaque healthcare system, AI chatbots might be better equipped to navigate insurance benefit design and healthcare access than patients are. This comes as care coordination has become increasingly laborious for healthcare consumers.
Patients seek AI health advice outside clinic hours
Microsoft's analysis added that patients rely on chatbots because they don't have access to their doctor.
For example, people were more likely to ask questions about symptoms in the evening and at night, potentially because doctors' offices are closed during those hours. A similar trend emerged for queries related to healthcare coverage and healthcare navigation, likewise indicating a lack of after-hours access to human support.
On the flip side, queries related to academic research were more likely during the day, suggesting that these types of messages are for educational or professional purposes.
This hypothesis is supported by trends in device usage. People chatting with Copilot for research or academic support were more likely to be using a desktop computer.
Conversely, messages seeking health information, advice about symptoms, fitness or lifestyle coaching and emotional well-being were more likely to happen on a mobile device.
AI chatbots provide support where the health system can't
Microsoft argued that these trends have emerged because the traditional healthcare system cannot keep up with patient needs and demands.
"Until recently, many people relied on internet search for navigating health questions," Microsoft said. "The problem is this can offer limited help in distinguishing between simple explanations and alarming possibilities. With growing pressure on healthcare services, we believe people need better tools to make sense of health information when access is challenging."
Despite the promise of AI chatbots, healthcare professionals have their concerns.
For one thing, chatbots aren't always good at detecting when a patient needs to meet with a clinician. AI is fallible, experts have stressed, and they might not be able to provide information within the correct context for an individual patient.
And that's not to mention the privacy and security concerns that come with patients leveraging a health tool that might not have the right safeguards in place.
Microsoft said it's taking the first steps in addressing those concerns, foremost by forming consumer health teams to support access to credible health information and connection to real-time provider directories.
Still, it remains to be seen how AI chatbots affect patient experiences and outcomes. As these tools continue to proliferate, industry experts must keep an eye on patient use habits and the impact utilization might have.
Sara Heath has reported news related to patient engagement and health equity since 2015.
