Apple Launches Health AI Study With OpenEvidence for Up to 200,000 iPhone Users

Apple launched the Apple Health AI Usage Study with OpenEvidence, open to up to 200,000 US adults with an iPhone 15 Pro or later. Participants get cited answers to health questions and can share HealthKit data; Apple says study data will not train AI models.

Christopher Ort
By Christopher Ort · 4 min read
Share
A person seen from behind sitting on a sofa in a sunlit living room holding a smartphone with an unreadable softly glowing screen.
Image: AI-generated, i10X
On this page
  1. Quick Take
  2. In plain English
  3. What happened
  4. How it works
  5. Why it matters
  6. What's still open
  7. Sources
  8. i10x Perspective

Apple has launched a large research study to find out how people actually use AI chatbots for health questions. The Apple Health AI Usage Study, built with clinical AI company OpenEvidence, is open to up to 200,000 adults in the United States through a dedicated iPhone app. Participants can ask everyday health questions, get answers with linked peer-reviewed citations, and choose to share Apple Health data such as resting heart rate, sleep or lab results. Apple says study data will not be used to train AI models.

i10X Quick Take

Quick Take

  1. What launched: the Apple Health AI Usage Study and its app, in the United States, announced in Apple’s newsroom and presented at the HLTH conference.
  2. Partner: OpenEvidence, a clinical AI platform used by healthcare professionals that answers with information from sources such as NEJM, JAMA and Nature.
  3. Who can join: up to 200,000 adults aged 21 and older with an Apple Intelligence-enabled iPhone (iPhone 15 Pro or later) on iOS 26 or later.
  4. What Apple studies: trust, response quality, comprehension and whether people take everyday health actions.
  5. Privacy: sharing HealthKit data is optional, and Apple says study data will not train AI models.

In plain English

Apple wants to know whether people understand and trust AI answers about their health before it builds this kind of feature into its products. So it is running a big experiment: volunteers get an app where they can ask health questions, see answers backed by medical research, and optionally let the app use their own health data from the iPhone. Apple watches how people use it and what they do next. The app does not give medical advice and says so.

What happened

Apple announced the study “in the United States, a virtual research study and associated app designed to advance the understanding of how people interact with conversational health AI tools.” It says the goal is to “inform the responsible design of health AI experiences for Apple and the industry.”

The answers come from OpenEvidence’s technology, wrapped in an Apple-designed research experience. Participants can “ask the study app everyday health questions,” explore complex medical information “in approachable language,” and see “linked, peer-reviewed clinical citations that ground each response.” Apple also wants to learn whether health AI can help people prepare for visits with their own doctors.

Dr. Sumbul Desai, Apple’s vice president of health and fitness, said in a statement reported by MobiHealthNews: “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.”

How it works

  • Questions and answers: participants ask health questions in the app and get cited answers. The App Store listing suggests uses such as understanding symptoms, putting Health data in context, preparing for a medical visit and building exercise or meal plans.
  • Personal data: participants can choose to share HealthKit data types to see how personal context affects relevance and understanding.
  • Guardrails: the app is designed to detect certain concerning questions and suggest following up with a provider or other resources.
  • Limits: Apple acknowledges that LLM-based health tools “may at times offer responses that are incomplete or unclear.” Neither the app nor the study provides medical advice, diagnosis or treatment.

Why it matters

For Apple users, this is the clearest sign yet of where Apple’s assistant is heading in health. Apple has been rebuilding Siri into an agent, as we reported in Apple accelerating its Siri AI agent overhaul. Running a 200,000-person study first is a cautious path compared with rivals that shipped consumer health chat directly, such as ChatGPT Health.

For OpenEvidence, the study puts its clinician-focused platform in front of consumers at Apple scale. It lands as OpenEvidence expands inside health systems, including its role in NYU Langone’s Epic workflow.

For the industry, the outcome data could become a reference point. Apple says the findings will inform health AI design “for Apple and the industry,” and a large real-world dataset on trust and comprehension is something regulators and clinicians have been missing.

What’s still open

  • Whether and when Apple will publish results, and in what form.
  • How long the study runs and whether it expands beyond the United States.
  • Whether findings feed into Siri, the Health app or a future Apple health coaching product.
  • How the guardrails perform on urgent or mental health questions at scale.

Sources

  • Apple Newsroom, “Apple launches research study to explore the use of AI for health,” November 2026
  • App Store listing, Apple Health AI Study App
  • MobiHealthNews and Healthcare Dive, reporting on the announcement at HLTH, November 2026

i10x Perspective

Apple is doing what it usually does with risky categories: arrive late, with more evidence. A study app with citations, opt-in data and no model training is a careful way to learn how people react to AI health answers before Siri starts giving them. The risk is that the market moves faster than the study. The upside is that when Apple ships health AI, it may be the first assistant with real data on whether people understood and acted on what it said.

Christopher Ort

Christopher Ort

Christopher Ort handles SEO, AI agent workflows and PR & partnerships at i10x.ai, the all-in-one AI workspace built by i10X Global Pte. Ltd. in Singapore. He is a graduate of Friedrich-Alexander-Universität Erlangen-Nürnberg (FAU). He is the author of “The Style Penalty”, a study published on SSRN that analyses 1,576 evaluation decisions by four frontier AI models to show how AI-written CVs fare in automated hiring screens; the research has been covered by heise online and t3n.

Editor's picks

  1. CVE-2026-126681: Prompt injection flaw in AWS Labs Timestream for InfluxDB MCP Server leaks API tokens
  2. CVE-2026-115563: OAuth credential leak in Model Context Protocol Python SDK
  3. US and Nine Allies Sign San Francisco Declaration to Protect AI Supply Chains
  4. Apple Launches Health AI Study With OpenEvidence for Up to 200,000 iPhone Users
  5. AMD Completes $8.2 Billion World Labs Acquisition, Fei-Fei Li Becomes Chief Scientist