Enterprise AI Analysis
How people use Copilot for Health
Authors: Beatriz Costa-Gomes*, Pavel Tolmachev*, Eloise Taysom, Viknesh Sounderajah, Hannah Richardson, Philipp Schoenegger, Xiaoxuan Liu, Matthew M Nour, Seth Spielman, Samuel F. Way, Yash Shah, Michael Bhaskar, Harsha Nori, Christopher Kelly, Peter Hames, Bay Gross, Mustafa Suleyman, Dominic King
Published: Microsoft AI, March 2026
This study analyzes over 500,000 de-identified health-related conversations with Microsoft Copilot to characterize how individuals engage with conversational AI for health inquiries. We reveal key insights into user intents, usage patterns, and the significant implications for responsible AI development and platform design in the healthcare domain.
Executive Impact: Key Findings for Enterprise AI
The research uncovers critical usage patterns that inform the strategic deployment of AI in health-related enterprise applications, from patient engagement to operational efficiency.
500,000+
Health Conversations Analyzed
20%
Conversations are Personal Health Intent
40%
Dominant General Information Category
14.3%
Personal Queries for Dependents
Deep Analysis & Enterprise Applications
Select a topic to dive deeper, then explore the specific findings from the research, rebuilt as interactive, enterprise-focused modules.
20%
of conversations involve personal symptom assessment or condition discussion.
Evening Surge in Personal Health Queries
The study reveals a marked increase in personal queries about symptoms and emotional health during evening and nighttime hours. This period often coincides with limited access to traditional healthcare services, positioning conversational AI as a critical after-hours resource. This temporal shift underscores the AI's role in addressing immediate personal health concerns when other options are scarce, aligning with population psychology trends of negative affect peaking in the evening.
14.3%
of personal health queries are made on behalf of dependents (children, parents, partners).
AI as a Caregiving Companion
A significant finding is that one in seven personal health queries are made on behalf of someone other than the user, such as a child, an aging parent, or a partner. This highlights conversational AI's potential as a caregiving tool, moving beyond individual self-care. Designing AI experiences to support caregivers requires different contextual cues and follow-up recommendations to ensure accuracy and completeness of information for vulnerable populations.
| Category | Mobile Usage | Desktop Usage |
|---|---|---|
| Primary Use |
|
|
| Timing |
|
|
| Task Complexity |
|
|
| Implications |
|
|
Contextual AI: Device as a Signal
The study reveals a sharp divergence in health AI usage patterns between mobile and desktop devices. Mobile users primarily engage with personal health intents, while desktop usage is dominated by research, academic support, and medical paperwork. This suggests that device choice signals fundamentally different modes of health engagement, with direct implications for platform-specific design and how AI responses are prioritized based on context.
High Prevalence
of queries related to finding providers, understanding insurance, and paperwork.
AI Addressing Healthcare System Friction
A significant portion of health AI queries focuses on navigating the complexities of healthcare systems, such as finding providers, understanding insurance, and completing paperwork. This indicates substantial friction in the delivery of existing healthcare services. Conversational AI is being used to simplify tasks that should, in principle, be straightforward, highlighting an unmet need for accessible support in navigating administrative healthcare challenges.
Privacy-Preserving Data Analysis Pipeline
Raw conversation transcripts
→
Automated PII scrubbing
→
LLM-generated privacy-preserving summary
→
LLM-driven topic clustering & analysis
Ethical AI Development: Privacy First
The methodology employs a two-stage, privacy-preserving pipeline to analyze de-identified health-related conversations. This involves automated scrubbing of Personally Identifiable Information (PII) followed by LLM-generated summaries that capture intent without reproducing original user words. This 'eyes-off' model ensures human researchers never access raw conversation content, underscoring a commitment to responsible and ethical AI development in sensitive domains like health.
Calculate Your Potential AI Impact
Estimate the efficiency gains and cost savings your organization could realize by strategically implementing AI solutions.
Estimated Annual Savings
$0
Annual Hours Reclaimed
0
Your AI Implementation Roadmap
Leverage the insights from this study to strategically plan your AI deployment in health, ensuring responsible and effective integration.
Phase 1: Longitudinal Tracking & Adaptation
Establish mechanisms to continuously monitor and track how AI intent distributions evolve within your organization. This will reveal new user applications and patterns as your conversational AI matures, allowing for adaptive strategies and feature development.
Phase 2: Geographical & Systemic Analysis
Conduct in-depth analysis of how health AI usage differs across various regions, healthcare systems, or departments within your enterprise. Identify unique needs in settings with varying primary care access to ensure responsible and globally relevant deployment strategies.
Phase 3: Outcome-Based Evaluation
Move beyond mere query characterization by linking AI intents to actual response quality and downstream outcomes. Evaluate whether the information and support provided truly help users improve their health decisions or operational efficiency.
Phase 4: Targeted Safety & Quality Investment
Concentrate investment in enhancing response quality and safety measures for high-stakes personal health intents, such as symptom assessment, condition management, and emotional wellbeing, where consequences of AI responses are highest.
Ready to Transform Healthcare with AI?
Understanding user needs is the first step. Let's discuss how these insights can shape a robust and responsible AI strategy for your enterprise.
Ready to Get Started?
Book Your Free Consultation.
Let's Discuss Your AI Strategy!
Lets Discuss Your Needs
Select a Date
Sun
Mon
Tue
Wed
Thu
Fri
Sat