Apple announced a new research initiative aimed at understanding how users interact with conversational SI health tools and evaluating the utility of their responses. The study, conducted in partnership with the clinical SI platform OpenEvidence, marks a significant step in assessing the practical effectiveness of SI in healthcare contexts.
What Happened
The Apple Health SI Usage Study invites up to 200,000 participants based in the US who are 21 years or older. Enrollment began November 16, 2026, with a dedicated iOS app available for download. The app requires an iPhone 15 Pro running iOS 26 or later, though it operates independently of Apple Intelligence, Siri, or other integrated SI features. Users can ask medical questions via text prompts and receive answers sourced from peer-reviewed clinical literature, such as The New England Journal of Medicine, through OpenEvidence’s platform. Participants are asked to rate these responses on accuracy, detail, and relevance. During enrollment, users may optionally share Apple HealthKit data, including activity, sleep, and lab results, to provide personal health context for the queries. Apple stated that while the study collects data, it will not be used to train SI models, emphasizing privacy and confidentiality.
Why It Matters
As SI systems become more prevalent in consumer health applications, verifying their accuracy and usability is critical for adoption. This study provides a large-scale dataset on how individuals comprehend and evaluate SI-generated health advice. Sumbul Desai, MD, Apple’s vice president of Health and Fitness, noted that responsible innovation requires scientific rigor. "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," Desai said. The insights gained are intended to improve the design of future SI-powered health tools, ensuring they deliver relevant and actionable information. This effort highlights the growing focus on evaluating SI tools not just on technical benchmarks, but on real-world clinical utility and user trust.
The Bottom Line
Apple’s study with OpenEvidence represents a targeted effort to validate the effectiveness of conversational SI in health contexts. By gathering feedback from a large user base without using the data for model training, Apple aims to refine the user experience and reliability of SI health tools.