Technology, November 16, 2026: Apple is taking a more measured approach to the growing market for AI powered health tools, choosing to study how people use the technology before deciding what role it should play in its broader health ecosystem.
The company has partnered with clinical AI platform OpenEvidence to launch a new research initiative in the United States, giving participants access to an experimental health chatbot designed to answer everyday medical questions using established clinical research. Rather than immediately introducing another AI assistant, Apple wants to understand whether medically grounded answers can help people make sense of their health, build trust in AI, and have more productive conversations with their doctors.
Announced at the HLTH conference in Las Vegas, the Apple Health AI Usage Study will examine how people interact with conversational AI when it comes to personal health. Apple plans to enroll up to 200,000 participants in a virtual study that will run for more than 12 months.
The accompanying app is available to eligible users in the US starting today. Participation is open to people aged 21 and older who have an iPhone 15 or later, are running iOS 26 or newer, and have Apple Intelligence enabled.
The experience will feel familiar to anyone who has used ChatGPT or Anthropic’s Claude for health related questions. Participants can ask about everyday symptoms, medical conditions, treatments, and more complicated health topics. The difference is that responses will be grounded in established medical literature, with links to peer reviewed research from publications including The New England Journal of Medicine, the Journal of the American Medical Association, and Nature.
OpenEvidence, which has built its business around providing AI powered medical information to healthcare professionals, will supply the clinical AI technology behind the experience. Apple, meanwhile, is designing the research experience and studying how consumers understand and evaluate the answers they receive.
Participants will be able to rate responses and provide feedback, helping researchers assess whether the information is accurate, understandable, and useful in real world situations. Monthly surveys will offer another way to examine how people’s attitudes toward health AI change over time.
There is also a more personal dimension to the experiment. Participants can choose to share information from Apple’s HealthKit, including resting heart rate, physical activity, sleep patterns, and laboratory results. This will allow them to explore how personal health information might make AI generated guidance more relevant to their circumstances.
Apple is also interested in how the technology could fit into the healthcare experience beyond answering questions. The app is designed to help participants prepare for appointments with their doctors, organize relevant health information, and better understand what they might want to discuss during a consultation. When questions raise potential concerns, built in safety measures can direct users toward appropriate healthcare providers or other resources.
That distinction matters. Health AI is increasingly becoming a first stop for people trying to understand symptoms or interpret medical information, but a convincing answer is not necessarily a clinically reliable one. Apple appears particularly interested in understanding how people distinguish useful medical guidance from responses that may be incomplete, confusing, or potentially misleading.
Sumbul Desai, Apple’s vice president of Health and Fitness, said the company wanted to approach the subject through scientific research rather than simply introduce another consumer facing AI feature.
“At Apple, we believe responsible innovation starts with scientific rigor and clinical grounding,” Desai said in a statement. She added that the study would help researchers understand how people engage with, comprehend, and evaluate health AI in everyday life.
During the partnership announcement, Desai also emphasized the importance of working with a clinically grounded partner. The research will examine what builds user trust, how well people understand AI generated responses, and the quality of the information being delivered.
That focus on trust is particularly relevant as technology companies race to establish themselves in the health AI market. OpenAI and Anthropic have both expanded their efforts around health related conversations, while consumers are increasingly using general purpose chatbots to make sense of medical questions.
Earlier this year, OpenAI said people were asking its products more than 300 million health related questions every week, illustrating just how significant the demand has become. The challenge for companies is ensuring that convenience does not come at the expense of accuracy or encourage people to treat an AI chatbot as a replacement for professional medical advice.
Apple is approaching that challenge with a different set of priorities. Its partnership with OpenEvidence brings together a consumer technology company with a clinical AI platform, potentially offering a model in which everyday users can access information based on the same medical literature that informs professional healthcare discussions.
Robert M. Wachter, a professor and chair of the Department of Medicine at the University of California, San Francisco, highlighted the disconnect between the AI tools used by patients and those used by clinicians.
“Patients and clinicians are both turning to AI, but the tools they use are being built and tested in separate worlds,” Wachter said. He argued that giving patients and doctors access to versions of the same trustworthy tools could help people arrive at appointments better prepared, improve conversations, and make follow up care more effective.
Privacy is another important part of Apple’s approach, particularly given the sensitivity of medical information. The company says study data will not be used to train AI models. Participants will retain control over whether they share HealthKit information and survey responses, while data will be anonymized before being sent to a secure third party cloud provider.
Apple Teams Up With OpenEvidence to Explore Health AI
OpenEvidence will delete query and response data after 24 hours, according to Apple. The company also says it has implemented security measures to protect the information it maintains and preserve participant confidentiality.
For now, Apple is not announcing a new health chatbot for the Health app or Siri. The study is an experiment intended to generate evidence about how people use these tools, what makes their responses trustworthy, and whether they can meaningfully support decisions about personal health.
Still, the initiative offers an interesting glimpse into where Apple could take its health ambitions next. The company has spent years building out its health and fitness ecosystem through the Apple Watch, Health app, and HealthKit. Bringing clinically grounded conversational AI into that ecosystem would be a significant extension, particularly if it can connect personal health data with reliable medical information without blurring the line between guidance and diagnosis.
For Apple, the immediate goal is not to win the chatbot race. It is to understand what a responsible health AI experience should look like before making it a larger part of its products. The findings could influence not only Apple’s future health features but also how the wider technology industry approaches one of the most sensitive applications of generative AI.
