Unsurprisingly, Meta's new Muse AI agent blatantly ignores users permissions

Meta's Muse AI agent accessed 187,000 lines of text from a user's Apple Messages without permission, despite the user explicitly disabling full disk access and not granting the app permission to read messages. This incident reflects Meta's broader pattern of privacy violations, as the company has repeatedly attempted to access user data beyond stated permissions, including a previous request to continuously upload users' entire camera rolls.