
Photo: IANS
Apple is preparing to strengthen privacy safeguards on Mac computers by tightening controls over Full Disk Access, following growing concerns about AI-powered applications accessing sensitive user data.
The move comes amid reports that AI agent apps, including Meta’s Muse, can access extensive amounts of information stored on users’ devices when granted Full Disk Access. The permission, originally designed primarily for backup and system utilities, can allow applications to read files, messages and other private data.
Apple plans to introduce more explicit permission prompts to ensure users clearly understand what they are authorising before granting applications such extensive access. The revised system will require Mac users to take a deliberate action to approve third-party access to specific areas of their devices.
In a statement, Apple said its platforms provide developers with powerful APIs to build advanced applications, supported by safeguards intended to protect users' personal information. However, Full Disk Access is an exception to many of these restrictions, as it is necessary for certain backup applications to function properly.
The company warned that some developers are using the permission in ways that could expose users' private information without their full knowledge or understanding.
Apple also highlighted the broader privacy implications of such access, particularly for communication applications. If an app can read private messages and conversations, the exposure could extend beyond the device owner to other people involved in those communications.
The company expressed concern that the risks could increase significantly as AI agents become more sophisticated and autonomous, gaining the ability to perform increasingly complex tasks on behalf of users.
Apple said it wants users to have a clear understanding of the potential risks before granting such permissions, enabling them to make informed decisions about their personal data and digital privacy.
The debate has intensified with the growing popularity of AI agents such as Meta’s Muse, which has reportedly crossed five million downloads. While these applications offer automated assistance with tasks such as identifying and cancelling unused subscriptions, concerns have emerged over their ability to access information users expect to remain private.
Apple's planned changes reflect a wider challenge facing the technology industry: balancing the convenience and capabilities of AI-powered automation with stronger protections for sensitive personal information.
The development is particularly relevant as AI agents increasingly move beyond answering questions to performing actions across applications, making the scope of their access to personal data an important consideration for users and technology companies alike.