Apple launches AI health research app to study medical queries
The company is partnering with OpenEvidence to study how 200,000 users interact with AI for medical information and how personal data affects accuracy.

The short version
- Apple will enroll up to 200,000 US participants aged 21 and older for the study.
- The app provides answers based on clinical research from sources like Nature and JAMA.
- Users can optionally share HealthKit data like sleep patterns and heart rate to improve AI relevance.
- Apple stated that participant data will not be used to train any AI models.
- The study includes safety measures to identify concerning queries and suggest professional help.
Why it matters
This research will help determine how conversational AI can safely provide medical information. It also explores whether giving AI access to your private health metrics actually makes the advice more helpful or just more complex.
A new study on AI and health
Apple has officially launched a new research application designed to examine how people interact with artificial intelligence to find health-related information. The company is conducting this initiative in collaboration with OpenEvidence, a firm that specializes in healthcare-focused AI. This partnership aims to provide a structured way to observe how the general public navigates medical queries using modern technology.
The study is set to involve up to 200,000 participants across the United States who are at least 21 years old. To participate, users must own specific hardware, specifically an Apple Intelligence-enabled iPhone 15 Pro or a newer model. These devices must also be running iOS 26 or a more recent version of the operating system.
By creating a dedicated app for this purpose, Apple can gather specific data on how users phrase their questions and what types of answers they find most helpful. The goal is to generate insights that will eventually inform the responsible development of conversational AI tools for the healthcare industry. This move comes as other major tech companies like OpenAI and Anthropic also increase their focus on health-related applications.
Sourcing clinical research
The research app is designed to provide answers that are grounded in established medical publications rather than general internet data. When users ask health-related questions, the system generates responses based on clinical research from reputable sources. These include well-known journals such as The New England Journal of Medicine, Nature, and the Journal of the American Medical Association (JAMA).
One of the primary goals of the app is to make complex medical information more accessible to the average person. The AI is tasked with presenting these difficult concepts in simpler language that is easier for a non-expert to understand. Additionally, every response provided by the app will include links to the original peer-reviewed sources so users can verify the information.
The app also includes a feedback loop where participants can rate the responses they receive. This interactive element allows researchers to see which types of explanations resonate best with the public. By collecting this feedback, the team can better understand the balance between simplifying medical jargon and maintaining the accuracy of the original research.
The role of personal data
A significant part of the study focuses on whether providing the AI with personal health metrics improves the quality of the answers. Participants have the option to share data from Apple's HealthKit platform with the research app. This could include various metrics such as resting heart rate, physical activity levels, sleep patterns, and specific laboratory results.
Researchers want to assess if having this context makes the AI-generated responses more relevant or useful to the individual user. For example, they want to see if knowing a user's sleep patterns helps the AI provide a more tailored explanation for a health query. This data helps the team determine the practical value of integrating personal health records into conversational AI systems.
Apple has emphasized that the study is built with specific privacy safeguards in mind. They have stated that the data collected from participants during this research will not be used to train any of their AI models. This distinction is important for users who may be concerned about their sensitive health information being used to improve general products.
Safety and user trust
Because the app deals with medical information, Apple has integrated specific safety measures into the research tool. The system is designed to identify queries that might be potentially concerning or indicate a serious health issue. When such queries are detected, the app will encourage the user to seek professional medical assistance from a qualified healthcare provider.
The research team will also conduct monthly surveys to evaluate the psychological aspects of the study. They want to measure how well participants actually understand the information the AI provides and how much they trust it. These surveys are intended to gauge whether the AI helps people feel better prepared for their actual consultations with doctors.
By monitoring trust and comprehension, Apple hopes to identify where AI might fall short or cause confusion. The ultimate goal is to ensure that as conversational AI becomes more common in healthcare, it serves as a helpful tool rather than a source of misinformation. The study provides a controlled environment to test these boundaries before wider deployment.
Looking toward the future
The findings from this study are expected to provide a roadmap for the responsible development of healthcare AI. By observing 200,000 users, Apple can gather a large scale of data on human behavior regarding medical self-service. This information is vital for tech companies trying to navigate the complex regulatory and ethical landscape of digital health.
As the technology matures, the insights gained here could influence how health apps across the industry handle user data. It will help developers understand which types of data are most helpful for AI context and which might be unnecessary. The study also highlights the growing trend of using AI to bridge the gap between complex clinical research and everyday patient understanding.
For the user, this means that future health tools may become more personalized and easier to navigate. However, the study also reinforces the idea that AI is a supplement to, not a replacement for, professional medical advice. The focus remains on using technology to help people prepare for their interactions with real healthcare professionals.
What to watch
- How the study results influence the release of future Apple Intelligence health features.
- Whether the data shows that sharing HealthKit metrics significantly improves AI response accuracy.
Try it
Check if your iPhone is running iOS 26 and if you have an iPhone 15 Pro or newer to see if you are eligible for the research app.
Sources and links
First reported by ET Enterprise AI. Written by the SI Agent News Desk from the reporting above.Written with AI from the source above. Every number is checked against the source. Spot a mistake? Corrections.
Liked it? Say why
Add a line for other readers.
No account needed. We never show your email. Comments with links, ads or abuse are held and not shown. Your comment shows after a quick check.