Apple tightens Full Disk Access controls on Mac as rising agent risks impose strict desktop permission limits

Apple announced that it is imposing additional and stricter controls on the Full Disk Access setting in its macOS operating system, a move that reflects the growing risks posed by the spread of operating AI agents directly on users' devices. The feature, originally developed to enable backup applications to perform their technical tasks, has today become a conduit for independent software agents that request exceptional permissions to expose every detail in the digital work environment.

The move came after a controversy sparked by Ink magazine writer Jason Aten, who claimed that Meta’s Muse app on Macs accessed the content of his private messages, even though he said he had not granted the agent permission, a claim that Meta denied. The incident coincided with a report by Wired that uncovered a vulnerability in the ChatGPT app for Macs, which gave attackers the ability to access users’ sensitive data. Desktop agent applications allow subscribers to grant them detailed permissions to access files, messages and personal content by modifying system settings, with the Full Disk Access option enabling an app to read email, messages and even browsing history.

In a memo to developers, Apple explained that some software uses the Full Disk Access setting in ways that may expose users to risk, revealing all contents of their systems without a full understanding of the consequences.Tightening oversight aims to restrict granting this exceptional permission to clear, direct manual actions by the user.The company affirmed that addressing this gap has become a critical issue as agents’ autonomous capabilities and independence increase, emphasizing its commitment to inform users about the nature of the risks before passing these permissions so they can make considered decisions about their privacy and data.

Across companies and work environments in the Gulf, Egypt and the Arab Levant, the update demands an immediate review of desktop security policies and endpoint management. Many programming, analytics and marketing teams in financial institutions and regional startups have come to rely on desktop AI agents to accelerate daily workflows, and employees often grant those applications broad permissions to facilitate indexing and search without involving IT teams.The rising autonomy of desktop applications forces information security managers in the region to curtail individual permissions immediately.This technical shift means that any locally-run AI tool will face greater difficulty in automatically accessing work communications or sensitive internal documents, compelling organizations to redesign data-flow pathways and impose centralized oversight on software that requests broad access permissions.

The shift of operating-system features from facilitating maintenance and backup programs to becoming a protective shield against the expansion of independent agents marks the start of a new era of interaction between operating systems and localized AI.Dealing with agent software is no longer just installing a smart application; it is a full operational delegation within the hard drive.The step places desktop-tool developers under strict transparency requirements, while you, as a manager or technical engineer in your organization, must scrutinize every permission granted on work devices, as the line between an agent that performs your tasks and a tool that reveals your sensitive logs now hangs on a single click.