Meta Platforms has seen rapid adoption of its new personal artificial intelligence assistant, Muse, which has been downloaded more than 5 million times from Apple’s App Store since its recent launch. Positioned as an AI tool that can manage everyday tasks including reading and responding to emails, organizing calendars, and making online purchases, Muse is currently available on Apple and Android devices.
Meta promotes Muse as a convenient assistant capable of handling routine activities with less effort from users. Advertising campaigns depict the AI prioritizing important emails, managing family schedules, ordering supplies, and controlling home devices. However, some of these capabilities have drawn skepticism, with Meta offering limited explanations of how Muse achieves outcomes such as negotiating higher resale prices of cars or reducing bills. A notable feature is that Muse continues operating even after users close the app, which Meta cites as an advantage despite widespread concerns about AI systems running without active oversight and the associated security risks.
Users must grant Muse access to their personal data, including email and retail accounts, to enable its functionality. Privacy advocates have voiced strong concerns about sharing sensitive information with the service due to Meta’s history of privacy violations and data mismanagement. The company has faced significant legal repercussions, including a $5 billion Federal Trade Commission fine in 2019 for deceiving users about privacy controls and a $17 billion settlement with California and other states over allegations of exposing children on Facebook and Instagram to abuse and predatory behavior. Experts warn that these past issues call into question Meta’s reliability as a custodian of personal data.
Meta has responded by emphasizing that Muse incorporates advanced privacy, safety, and security protections that are unique to the product. The company maintains that Muse cannot view user passwords or payment methods, although it can use them for transactions. Despite these assurances, early users and reviewers have reported privacy-related glitches. In one case, Muse revealed a seller’s home address to a potential buyer without consent, attributed by Meta to account settings confusion. In another, the AI accessed and repeated personal messages despite permission restrictions, highlighting potential complexities in managing Muse’s settings.
Questions remain about the necessity of a tool like Muse for typical users. Reviewers have noted that while Muse can perform tasks like scheduling, ordering items, and managing calls, these functions are already achievable with existing smartphone and computer features. Some users find that Muse’s benefits do not justify the risks, particularly the possibility of errors such as sending incorrect messages, deleting important data, or making unintended purchases. Additionally, Amazon has restricted Muse’s ability to make purchases on its platform, although the AI can still transact on some other retail sites like Walmart and Expedia.
As AI assistants become more embedded in daily life, users must weigh the convenience offered by services like Muse against potential privacy vulnerabilities and operational risks. Meta’s introduction of a friendly, approachable AI character masks broader concerns about security and user control, underscoring the ongoing debate over trust in technology providers handling sensitive personal information.