AivexaNewsSearch
AI news for builders and product teamsChecked every hour

Apple launches new app to study how people use AI for health

Collected Nov 16, 2026

Apple has released a new iPhone app that enrolls users in a research study on how people ask and act on health questions answered by AI. The company is running the study with OpenEvidence, a healthcare-focused AI firm, and has opened enrollment to U.S. users starting today. The app is the front door to the study rather than a general health product: participants ask medical questions in a chatbot-style interface and then rate and comment on the responses.

Eligibility is narrow. Participants must be in the United States, at least 21 years old, using an iPhone 15 or newer running iOS 26 or newer, with Apple Intelligence turned on. Apple aims to enroll up to 200,000 people and to keep the study running for more than 12 months.

The answers are grounded in clinical sources rather than general web content. Apple names The New England Journal of Medicine, the Journal of the American Medical Association and Nature among the literature behind the responses. The app is meant to turn those sources into simpler explanations of everyday medical questions. Beyond rating answers, participants are asked to fill out monthly surveys, which suggests Apple is collecting longitudinal self-reported data alongside the chat interactions.

Participants can also choose to share HealthKit data — resting heart rate, activity, sleep or lab results — so the app can give more personalized and specific suggestions. Apple says the app can help people prepare for a doctor's visit and will point users toward specific providers when a question involves a concerning medical issue.

Privacy terms are explicit. Data is anonymized before it goes to a secure third-party cloud provider, and Apple says it will not be used to train any AI models. OpenEvidence deletes query and response data after 24 hours, according to the company. Sumbul Desai, Apple's vice president of Health and Fitness, framed the study as an exercise in scientific rigor and clinical grounding, saying the findings are intended to inform the broader scientific conversation about how people engage with, comprehend and evaluate health AI.

What the study is not, at least publicly, is a shipping feature. Apple has not said whether it plans to bring AI answering to its Health app or to Siri. The framing is research first: measure whether clinically grounded answers actually help people understand and act on health information before deciding what to build. Apple's scale claim rests on the combination of curated literature and the ability to cross-reference a participant's own health signals — a pairing that generic chatbots cannot easily replicate because they lack access to HealthKit data.

The backdrop is scale. Earlier this year OpenAI said users were asking more than 300 million health-related queries each week, a figure that shows how much demand already flows to general-purpose assistants. Apple's entry differs in kind rather than degree: instead of competing for those queries with a consumer chatbot, it is studying a narrower question — whether sourcing answers to peer-reviewed literature changes comprehension and outcomes.

For developers and product teams, the interesting parts are the constraints and the data posture. The iPhone 15 and iOS 26 requirements plus Apple Intelligence mean the study reaches a modern slice of the installed base, not everyone. The anonymization step, the third-party cloud arrangement and the 24-hour deletion window are the kind of pipeline choices that teams building regulated health features will be compared against. The HealthKit opt-in also sketches a model where an assistant reasons over personal sensor and lab data, which raises consent and data-retention questions that the study's design implicitly answers for itself.

This suggests a few things. A 12-month, 200,000-person study is a slow path to a product decision, so any consumer-facing health AI from Apple would likely arrive after results, not alongside them. A likely trade-off is depth versus breadth: grounding in major journals improves trustworthiness but may narrow the range of questions the app handles well compared with open-web chatbots. How Apple reports findings, and whether participation translates into anything shipped, is the part to watch.

Why it matters: Apple is testing clinically sourced health AI on a large user base while keeping the data out of model training, which sets a reference point for privacy and provenance that other health-AI builders may be measured against. Teams working on medical or wellness features should note the gating on device and OS, the opt-in HealthKit integration and the short retention window as design patterns rather than marketing claims. Whether this becomes a Health app or Siri feature is, by inference, a decision that likely waits on the study's results.

Read at TechCrunch · AI

Based on reporting from the original publisher. Visit the source for full context and later updates.

Publisher excerpt

Apple's new app for a study will answer everyday medical questions.