Wink Pings

Apple Designs a Grading Challenge for Its AI Health Tools: Users Will Be the Raters

Apple has partnered with clinical AI platform OpenEvidence to launch the Apple Health AI Usage Study. Participants will ask health-related questions via a new iOS app, then score the generated answers based on accuracy, level of detail and relevance.

Apple Health AI Usage Study 应用界面

Apple has announced a new research project to clarify how people use conversational AI health tools, and how useful the responses these tools deliver really are. The research, named the Apple Health AI Usage Study, is co-launched with OpenEvidence — a clinical AI platform built for medical practitioners, whose outputs are generated from medical information sourced from authoritative publications including the New England Journal of Medicine.

Participants will use a new dedicated iOS app to submit health questions, then rate the answers against metrics including accuracy, detail sufficiency and relevance. The app is already available for download, and enrollment for the study is now open.

Accuracy sits at the top of all grading dimensions. Different from general chatbot use cases, errors in health Q&A come with far higher costs, and the level of detail and relevance of responses also directly determine whether they can be put into real use.

The full original story from The Verge is available here: Read the full story at The Verge.

Photo: Apple

发布时间: 2026-11-16 09:26