What Is Microsoft Copilot Health?
Tech companies have long argued that their AI tools improve with more data. Now, they want the most sensitive data of all — your health records. Microsoft recently launched Copilot Health, a dedicated feature within its Copilot AI assistant. The tool allows users in the United States to upload medical records from multiple health providers. It then connects that data with readings from wearable fitness devices, such as an Apple Watch or Fitbit.
Microsoft currently opens Copilot Health through a phased rollout. Interested users can join a waitlist to gain early access. Initially, the tool will be free. However, Microsoft plans to introduce a subscription fee in the future, though pricing details remain undisclosed.
How Does It Work?
Setting Up Your Health Profile
Users access Copilot Health through a dedicated “Health” tab inside the Copilot website or mobile app. First, they answer basic questions about age and sex to build a personal health profile. From there, they can optionally connect health records from over 50,000 U.S. hospitals and provider organizations through a platform called HealthEx.
Connecting Wearables and Lab Data
Additionally, Copilot Health supports more than 50 wearable devices, including Apple Health, Oura, and Fitbit. Users can also import comprehensive lab results through Function. Once connected, the chatbot analyzes all incoming data and surfaces personalized health insights.
What the Chatbot Can Tell You
Instead of offering a direct diagnosis, Copilot Health presents observations. For example, if a user reports poor sleep, the tool may highlight sleep trends recorded since a recent hospital visit. Furthermore, it can suggest questions to bring to a doctor’s appointment rather than prescribing a course of action. Microsoft says its AI can complete a full records review — a process that might take a doctor several hours — in just seconds.
Other Tech Giants Join the Race
Microsoft is not alone in this effort. Several major technology companies have launched similar tools in 2026. OpenAI rolled out ChatGPT Health in January, connecting its chatbot with users’ medical records and wellness apps. Amazon then expanded access to its Health AI assistant, developed through its primary care arm, One Medical. Shortly afterward, Anthropic introduced Claude for Healthcare for consumer use.
Consequently, 2026 has quickly become a defining year for consumer health AI. Microsoft alone handles over 50 million health-related questions per day across its AI products, including Bing and Copilot. Nearly one in five Copilot conversations, according to Microsoft’s own research, involves a personal symptom or health condition.
The Privacy Risk You Must Know
HIPAA Does Not Protect You Here
This is the most critical point for any user to understand. HIPAA — the Health Insurance Portability and Accountability Act — does not apply to tech companies offering AI chatbots. Traditional healthcare providers must protect patient privacy under strict federal law. Tech companies, however, face no such obligation. Therefore, these companies could theoretically use health data to train their AI models or serve targeted advertising based on a user’s medical conditions.
What Microsoft Promises
Microsoft states that Copilot Health data undergoes encryption both at rest and in transit. The company also pledges not to use personal health data for AI training or targeted ads. Moreover, it says it shares customer data with law enforcement only in response to valid legal requests. Users can disconnect their health data sources at any time. Copilot Health has also achieved ISO/IEC 42001 certification, an independent standard for AI management systems.
Nevertheless, independent researchers have not yet studied Copilot Health in a clinical setting. Users should weigh Microsoft’s assurances carefully against the absence of regulatory oversight.
What Doctors Say About AI Health Tools
Potential Benefits
Physicians acknowledge several genuine advantages. As healthcare costs rise and clinician shortages grow, an AI tool could help patients better understand their own health between appointments. Arjun Manrai, an assistant professor of biomedical informatics at Harvard Medical School, describes appointment preparation as “a spot-on target” for AI language models. He notes that AI is always available, while a doctor is not.
Real Concerns Remain
Despite those benefits, doctors also warn of serious risks. AI health chatbots have previously made headlines for contributing to users’ anxiety, unhealthy behaviors, and in some cases, psychological distress. Tools that flag health anomalies without context could easily trigger unnecessary doctor visits. As with earlier consumer health technologies like WebMD, overconsumption of AI-generated health insights may create more anxiety than it resolves.
What Copilot Health Will Not Do
Microsoft explicitly positions Copilot Health as a support tool, not a replacement for medical professionals. Its official release materials include a disclaimer stating the chatbot “is not intended to diagnose, treat, or prevent diseases.” Rather than confirming a condition, it may offer a list of possible diagnoses. Instead of recommending a specific medication, it will suggest questions a user can bring to their doctor.
Microsoft developed the tool alongside an internal clinical team and an external advisory panel of over 230 physicians from 24 countries. The company says it releases new features gradually, testing each stage with a small group of users to ensure safety and reliability.
Should You Share Your Health Records?
Before connecting your health data to any AI platform, consider these practical steps:
- Review the privacy policy carefully, not just the company’s public statements.
- Understand HIPAA’s limits — tech companies are not bound by it.
- Start with general questions before sharing detailed medical records.
- Use the tool as a supplement, not a substitute, for professional medical advice.
- Check data deletion options and ensure you can disconnect your records instantly.
AI health tools carry genuine promise — particularly for patients who lack affordable, consistent access to healthcare. However, meaningful consumer protections remain absent. Until regulators extend privacy rules to cover health AI platforms, users must proceed with informed caution.
