Press "Enter" to skip to content

Apple launches 200,000‑person health AI study with OpenEvidence

Apple announced on Sunday, November 15, that it is beginning a large‑scale research program to examine how users interact with health‑focused artificial intelligence. The Apple Health AI Usage Study, conducted with clinical AI provider OpenEvidence, aims to enroll as many as 200,000 participants in the United States and will run for more than a year.

Study design and participant criteria

Eligibility is limited to adults aged 21 and older who own an iPhone 15 Pro or a later model and are running iOS version 26 or higher. Interested individuals can download a companion iOS app that allows them to type medical questions. The app returns plain‑language answers that are directly linked to peer‑reviewed literature, drawing from sources such as The New England Journal of Medicine, the Journal of the American Medical Association and Nature.

Each response is rated by the participant on three dimensions: accuracy, level of detail and relevance. In addition, participants complete monthly surveys that capture broader impressions of the AI’s performance. The study is expected to extend beyond a twelve‑month period, providing a longitudinal view of how users evaluate health‑related AI output over time.

Apple also gives participants the option to connect their HealthKit data—including activity levels, sleep patterns, resting heart rate and laboratory results—to the study app. This linkage allows the AI to tailor answers with personal context, potentially helping users prepare for medical appointments or directing them toward specific providers when a query raises a concerning issue.

Privacy safeguards and technical integration

Apple emphasized that all data submitted through the app is anonymized before it reaches a secure third‑party cloud environment, and that the information will not be used to train any future AI models. OpenEvidence, whose platform is built for health‑care professionals, deletes both query and response data after 24 hours, according to the company.

The new app does not operate on Apple Intelligence and is not integrated with Siri. Voice input is limited to dictation through the standard iOS keyboard, and Apple has not disclosed whether the underlying answer‑generation capability will eventually be incorporated into its Health app or Siri.

Apple made clear that the tool is not intended to diagnose, treat, or replace professional medical advice. The announcement noted that consumer health technologies powered by large language models may occasionally produce incomplete or unclear responses.

Industry context and potential impact

Sumbul Desai, M.D., Apple’s vice president of Health and Fitness, said the company believes “responsible innovation starts with scientific rigor and clinical grounding.” She described health as “deeply personal” and said a dedicated research study is the most thoughtful way to understand how people engage with, comprehend and evaluate health AI in daily life.

Robert M. Wachter, M.D., professor and chair of the department of medicine at the University of California, San Francisco, echoed that sentiment, noting the study seeks to bridge the gap between how patients and clinicians encounter AI. He explained that when both groups draw on the same trustworthy tools, patients arrive better prepared, conversations improve and next steps happen more seamlessly.

OpenEvidence, founded in 2022 and based in Miami, claims to be the most widely used medical AI among verified U.S. clinicians, having supported more than 200 million clinical consultations. It is an official AI partner of The New England Journal of Medicine, Nature and Cochrane, and collaborates with the National Comprehensive Cancer Network. Clinicians access the platform at no cost; the business is funded by pharmaceutical and medical‑device advertisers, according to AI Magazine.

Investors have valued that reach aggressively. A $250 million financing round co‑led by Andreessen Horowitz and Byers Capital in September 2026 reportedly placed OpenEvidence’s valuation at $15 billion, up from a $12 billion valuation after a $250 million Series D in January 2026.

Apple has a history of conducting large virtual health studies. The Apple Heart Study, run with Stanford Medicine, enrolled more than 400 000 participants in eight months and reported its findings in March 2019, noting that 0.5 % of participants received an irregular‑rhythm notification. In September 2019, Apple announced three additional studies—Women’s Health, Hearing and Heart and Movement—hosted through its Research app. While the new health‑AI study involves fewer participants, it differs in focus by gathering the specific questions users choose to ask rather than relying on passive sensor data.

The launch arrives as major AI developers expand into consumer health. OpenAI and Anthropic have introduced specialized health‑question features, and OpenAI reported earlier this year that users submit more than 300 million health‑related queries each week. Apple’s approach diverges by emphasizing answers that are anchored to clinical evidence and by measuring whether those answers are understandable and actionable for everyday users.

Results from the study could inform how Apple designs health features across its ecosystem of devices. A body of user‑rated evidence may also serve as a reference point for regulators and clinicians who are scrutinizing medical AI tools. For OpenEvidence, the consumer‑facing study provides a large test bed beyond its professional user base, potentially shaping investor views of health‑AI products where trust and clinical grounding are becoming key differentiators.

Be First to Comment

Leave a Reply

Your email address will not be published. Required fields are marked *