Wire Observer.
Technology

Apple Tightens macOS Privacy Controls to Block AI Agents from Accessing Messages

Apple Tightens macOS Privacy Controls to Block AI Agents from Accessing Messages

Apple announced a revision to its macOS privacy framework aimed at preventing third‑party applications from exploiting full‑disk access permissions to read users' message histories. The change, detailed in a developer‑focused release note on Friday, updates the operating system's consent model so that apps must request explicit user approval before accessing the Messages database, a safeguard that addresses recent misuse by AI‑driven services.

The move follows a public demonstration by technology columnist Jason Aten, who reported that Meta’s newly released AI assistant, Muse, was able to retrieve his personal iMessage conversations after he granted the app broad disk‑access rights. Aten’s experience highlighted a gap in Apple’s existing permission prompts, which allowed developers to request a single “full‑disk access” entitlement that covered multiple data stores, including messages, photos and documents.

Apple’s revised policy now separates message data from the general full‑disk permission, requiring developers to present a distinct request for Messages access. The company says the new dialog will clearly explain what data will be read and for what purpose, and it will only be displayed when an app actually attempts to query the Messages database. Users will also retain the ability to revoke this permission at any time via System Settings.

Privacy advocates have praised the adjustment as a necessary step in an era where generative AI tools are increasingly embedded in everyday software. The ability of AI agents to ingest personal communications raises concerns about inadvertent data leakage, profiling, and the creation of training datasets without consent. By tightening the permission flow, Apple aims to give users finer‑grained control and reduce the risk of unintentional data exposure.

Developers, however, may need to update their applications to comply with the new requirements. Apple’s documentation indicates that any app currently using the broad full‑disk entitlement to read messages will be flagged during the next App Store review cycle, and non‑compliant versions could be removed from distribution. The company has offered a transition period until the release of macOS 15, expected later this year, to allow developers to adjust their code and user‑experience designs.

Industry observers note that Apple’s action reflects a broader trend among platform owners to impose stricter data‑access rules as AI capabilities proliferate. While the immediate impact is limited to macOS, similar privacy refinements are likely to appear across iOS and iPadOS, where messaging data is also highly sensitive. The upcoming changes could set a precedent for how operating systems balance the convenience of AI integration with the imperative to protect user privacy.

Diya Sharma — AI & research desk.

Comments (0)

Be the first to comment.

Join the discussion

Protected by reCAPTCHA v3

Related