Apple is taking a closer look at how people use artificial intelligence for health information, launching a new research initiative called the Apple Health AI Usage Study in collaboration with clinical AI platform OpenEvidence. The study will examine how people interact with conversational health tools, whether they trust the information they receive, and how AI-generated answers influence their understanding of health-related topics and everyday decisions.
Announced on November 15, the virtual research study is designed to help Apple and the wider healthcare technology industry understand how conversational AI can deliver health information more responsibly. Rather than simply evaluating how accurately an AI tool answers questions, Apple wants to investigate whether people understand those answers, find them useful, and know when to seek professional medical advice.
Apple partners with OpenEvidence for clinically grounded AI
The study combines Apple’s research experience with medical AI technology developed by OpenEvidence, a clinical AI platform widely used by healthcare professionals.
OpenEvidence generates responses using medical information from established publications and research sources, including The New England Journal of Medicine, Journal of the American Medical Association (JAMA), and Nature. This approach is intended to give participants access to medically grounded information rather than relying solely on general-purpose AI responses.
Through the dedicated study app, participants will be able to ask everyday health questions, explore complicated medical topics explained in more approachable language, and review linked citations from peer-reviewed clinical literature supporting the responses.
Apple also wants to understand whether connecting personal health information to conversational AI can make its responses more relevant. Participants can voluntarily share selected HealthKit data, including resting heart rate, activity, sleep information, and laboratory results.
This will allow researchers to examine how access to personal health metrics affects the relevance and comprehension of AI-generated health information.
Studying trust, understanding, and real-world health decisions
One of the central questions behind the initiative is whether people can properly understand and evaluate health information generated by AI.
Although large language models are increasingly being used to answer medical questions, their responses can sometimes be incomplete, unclear, or difficult to interpret. In healthcare, that creates a particular challenge because misunderstanding an answer can influence decisions about symptoms, medications, treatments, and medical appointments.
Apple’s research will examine several aspects of these interactions, including user trust, response quality, comprehension, and everyday health actions.
The company will also investigate whether conversational AI can help people prepare more effectively for appointments with their doctors. For example, having an AI tool explain medical terminology or organize health-related questions could help patients have more informed conversations with healthcare professionals.
Robert M. Wachter, M.D., professor and chair of the Department of Medicine at the University of California, San Francisco (UCSF), highlighted the importance of connecting the AI tools used by patients and clinicians.
He noted that when patients and doctors can draw on similarly trustworthy tools, patients may arrive better prepared, conversations can improve, and follow-up steps can become more seamless.
Apple’s vice president of Health and Fitness, Sumbul Desai, M.D., said the research is intended to understand how people engage with, comprehend, and evaluate health AI in everyday life.
Privacy protections and medical safety guardrails
Given the sensitive nature of health information, Apple says the study has been designed with privacy protections, transparency, and clinical safeguards from the beginning.
Importantly, data collected through the study will not be used to train AI models. Participants will retain control over whether they share HealthKit metrics or survey responses, while Apple says it has implemented security measures to protect the research data it maintains and participants’ confidentiality.
The study app also includes safety guardrails designed to identify certain concerning questions and encourage participants to contact healthcare providers or consult other appropriate resources when necessary.
However, Apple is clear that the app is a research tool, not a medical service. It is not intended to diagnose or treat medical conditions or replace professional medical advice. Participants are encouraged to consult doctors or other qualified healthcare professionals about health concerns and to verify medical information.
Who can participate in Apple’s Health AI study?
Apple is opening the study to up to 200,000 eligible participants across the United States, with enrollment limited to adults aged 21 and older.
There are also specific device and software requirements. Participants must have an Apple Intelligence-compatible iPhone, which means an iPhone 15 Pro or later, running iOS 26 or a newer version.
Eligible users can download the dedicated Apple Health AI Usage Study app through the App Store to review the study information and choose whether to enroll.
For now, the research is limited to participants in the United States, and Apple has not announced plans to expand enrollment to other countries.
The initiative comes as conversational AI becomes an increasingly common way for people to search for health information. But healthcare is an area where a convincing-sounding answer is not necessarily a reliable one. By studying how people interpret AI-generated medical information, Apple and OpenEvidence hope to better understand what makes these tools genuinely useful, trustworthy, and safe in everyday life.
Discover more from GadgetBond
Subscribe to get the latest posts sent to your email.



