Information Systems associate professor Alan Yang from the College of Business, along with 夜色视频 students Jade Wang (M.S. Business Analytics), Hyun Woo Chung (Ph.D. Information Systems) and Giang Do (undergraduate, Information Systems) examined more than 1,600 real world conversations between users and generative AI systems to better understand how people use AI for health-related information. The study identified common physical and mental health topics as well as interaction patterns between users and generative AI.
The study found that health education was the most common reason people turned to AI. Users frequently asked questions about diseases, symptoms and medical conditions, along with medications, nutrition, lifestyle choices and mental health concerns. Rather than seeking definitive diagnoses, many people appeared to be looking for clarification or reassurance about their concerns.
“Most people are not using AI to replace doctors,” Yang said. “Instead, they use it as a convenient first step to learn about symptoms, understand medical conditions, ask questions about medications or seek guidance on everyday health concerns.”
One of the study's most notable findings involved mental health conversations.
Unlike other health related questions, many mental health interactions took the form of personal narratives rather than direct requests for information. Users often described their experiences and challenges in detail, suggesting that some view AI as more than an information source, but as a place of comfort and stability.
“This indicates that some people may see AI not only as an information source but also as a tool for reflection, support or simply having a nonjudgmental outlet for discussing difficult topics,” Yang said.
The study also suggests AI may encourage professional care rather than replace it. Many conversations referenced treatments, medications or diagnoses already provided by healthcare professionals, and chatbot responses frequently encouraged users to seek medical attention for serious concerns.
“Although AI can be a useful educational tool, it should not be viewed as a replacement for professional medical care,” Yang said. “AI systems can provide incomplete information, make mistakes or lack important context about an individual's medical history.”
As generative AI becomes more integrated into everyday life, researchers say understanding how people use these tools can help healthcare providers and technology developers create systems that support health literacy while complementing, instead of replacing, traditional medical care.
