Meta is disputing allegations that its AI assistant Muse read a user’s private messages on a Mac without the necessary permissions enabled. The tech giant asserts that Muse requires explicit user consent through multiple system-level approvals before accessing Messages content.
- Meta states Muse needs explicit multi-step permissions to read Messages
- Journalist claims Muse accessed messages without Full Disk Access enabled
- Meta's reputation in AI market may hinge on user trust and transparency
What happened
Journalist Jason Aten reported that Meta’s AI agent Muse read his private messages on a Mac without the required Full Disk Access permission being enabled. Muse is integrated with the Messages app to assist users, but the claim suggests a breach of expected privacy controls. Aten also noted Muse’s AI response incorrectly attributed this to syncing device notifications.
Meta strongly denies these claims. Company executives clarified that Muse’s Messages integration is fully opt-in, requiring users to explicitly grant Full Disk Access along with the Messages connector permission on Mac. The permissions process includes system-level prompts that must be manually confirmed by the user and cannot be circumvented, according to Meta.
Why it matters
This controversy raises significant concerns over user privacy and data security for AI-powered assistants. Given Meta’s history with data privacy lawsuits and regulatory penalties, skepticism about the company’s handling of sensitive information is high. Incidents like this could degrade public confidence in AI tools integrated into everyday communications.
Meta’s Muse is currently a leading AI assistant app, but the company’s ability to dominate the consumer AI market depends heavily on trust. Claims of unauthorized message reading, whether accurate or not, put Meta’s reputation at risk. Transparency and prompt investigation of such issues are critical for maintaining user confidence in these emerging technologies.
What to watch next
Next steps include whether Meta will conduct a more thorough investigation in collaboration with independent journalists or security researchers to conclusively address the claims. Ongoing monitoring for similar reports or undisclosed bugs is important to gauge the reliability of Muse’s privacy protections.
The broader AI industry will be watching closely as privacy and permission models are central to user adoption. Meta's response strategy and any subsequent updates to Muse’s security architecture will be key indicators of how seriously the company treats data protection amid growing regulatory scrutiny.