Apple Wants 200,000 iPhone Owners to Grill a Health AI, and Siri Is Not Involved

Apple has opened a US research study app built with clinical AI firm OpenEvidence to learn whether people trust and understand AI health answers. It is a study with strict eligibility, not a new Siri or Health app feature.

Nov 16, 2026
Apple Wants 200,000 iPhone Owners to Grill a Health AI, and Siri Is Not Involved
A man's face lit only by his smartphone screen in a dark room

Apple wants an AI to answer your health questions. Just not in Siri, not in the Health app, and not for everyone. On Sunday the company opened enrollment for the Apple Health AI Usage Study, a virtual research study with its own iPhone app, built in collaboration with clinical AI company OpenEvidence. Apple wants up to 200,000 adults in the United States to ask questions, rate the answers, and show Apple what makes people trust a medical chatbot.

That framing matters. OpenAI and Anthropic have already pushed health features into their chatbots, and TechCrunch noted that OpenAI said earlier this year users were sending over 300 million health questions a week. Apple is moving slower. In its newsroom announcement, Apple says the study will examine "what drives user trust, the quality of responses, comprehension, and everyday health actions" to inform the responsible design of health AI "for Apple and the industry."

Apple is gathering evidence before building a product. TechCrunch says it is not clear whether Apple will ever bring AI health answers to Siri or the Health app.

"Health is deeply personal, and conducting a dedicated research study is the most thoughtful way to understand how people engage with, comprehend, and evaluate health AI in everyday life." Sumbul Desai, M.D., Apple vice president of Health and Fitness

The brains of the app come from OpenEvidence, which Apple describes as a clinical AI platform widely used by healthcare professionals. Apple says it generates answers using medical information from established sources such as The New England Journal of Medicine, the Journal of the American Medical Association and Nature. Participants get answers in plainer language with linked, peer reviewed citations. The Verge reports that users can rate responses on accuracy, detail and relevance, and TechCrunch says participants will also be asked to fill out monthly surveys.

Apple is blunt about why it needs the research. Its announcement concedes that consumer health tools powered by large language models "may at times offer responses that are incomplete or unclear."

Eligibility is tight. Apple says participants must be 21 or older, live in the United States, and own an Apple Intelligence enabled iPhone, meaning an iPhone 15 Pro or later, running iOS 26 or later. The odd wrinkle is that The Verge reports that the app itself does not use Apple Intelligence and is not integrated with Siri, relying instead on typed prompts with keyboard dictation as an option.

Participants can choose to share HealthKit data such as resting heart rate, activity, sleep or lab results, which Apple says will test how personal data affects relevance. Apple also says the study will explore whether health AI can help people prepare for visits with their own doctors. UCSF medicine chair Robert Wachter, quoted in the announcement, argued that patients and clinicians are using AI tools "being built and tested in separate worlds."

Trust Apple on privacy, for now

Every privacy promise here comes from Apple. The company says the study was designed with privacy protections, transparency and clinical guardrails, that study data will not be used to train AI models, and that participants control whether to share HealthKit metrics or survey responses. According to TechCrunch, Apple also said data will be anonymized before it goes to a secure third party cloud provider and that OpenEvidence deletes query and response data after 24 hours. TechCrunch adds that Apple aims to run the study for more than 12 months, though Apple's own post gives no length.

Apple says safety guardrails flag certain concerning questions and point people to a provider, and a footnote says the app is not meant to diagnose, treat or replace medical professionals. None of that has been independently audited yet. If Apple wants people to trust AI with their health, a study that publishes what it finds, including the bad answers, would be a strong start.

The study lands as money pours into consumer health AI, including Oura's 625 million dollar purchase of Counsel Health. Apple is betting that being late and careful beats being first.