App Store policy refresh clarifies rules for generative AI apps and data use

AI · 4 min read

In a policy update published this week, Apple spelled out guidelines for apps that incorporate generative AI, requiring clear in-app disclosures when content is AI-generated and straightforward ways for users to opt out of data collection used to train models. The guidance also asks developers to maintain provenance metadata so that generated images and text can trace their source styles when feasible.

Apple is emphasizing transparency and human-in-the-loop controls. Apps that provide AI content moderation tools must include fallback human review options where outcomes materially affect user rights or safety. The policy applies to iOS, iPadOS, and visionOS apps distributed through the App Store.

Developers welcomed the clarity but noted implementation complexity, particularly around provenance metadata and third-party model services. Apple said it will provide policy-compliance checklists and a review pilot program to help teams adjust before enforcement begins.