Meta suspends Muse Ai sharing feature over privacy concerns
The discovery has raised serious questions about Meta Muse AI and how shared conversations are protected.
The findings also highlight growing Muse AI privacy concerns as more people use artificial intelligence tools for personal and professional tasks.
Researchers Discover Thousands of Shared Chats
Security researchers at Claroc identified the issue over the weekend. According to their findings, Muse AI generated conversation links using a predictable format that included the title of each chat.
These links also appeared in public Domain Name System (DNS) records, which help connect website names with internet addresses.
Researchers used archived DNS information to identify approximately 6,800 shared conversations.
Some of the chats reportedly contained sensitive details, including meetings involving a major AI laboratory, a small-business loan application, a bank fraud alert with partial account information, a company vacation calendar and internal business documents.
The findings did not mean that every Muse AI conversation was exposed. The reported issue involved shared conversations whose links researchers were able to discover.
The discovery intensified Muse AI privacy concerns because users may not have expected their shared links to be so easily found.
Why Did Meta Suspend the Feature?
Meta confirmed that it had paused the sharing function while reviewing the findings and considering changes to the system.
The company explained that shared conversations created webpages accessible to anyone who had the link.
However, researchers found that public DNS records could make some of these links easier to discover. Meta suspends sharing feature access while it assesses the issue and reviews possible improvements.
Read Also: Kamala Harris Urges US And China To Cooperate On AI Safety
Meta also said its bug bounty programme allows security researchers to report weaknesses and help improve its products.
Following the suspension, Claroc reported that researchers could no longer access a sample of the previously available links.
“We are glad to see it closed this quickly,” Claroc said.
The response may reduce further access through the reported method, although users should not assume that all previously shared information has automatically been deleted.
Previous Privacy Concerns
The latest incident follows another dispute involving Muse AI in September.
Technology columnist Jason Aten alleged that Muse accessed private messages on his Mac without the system’s “Full Disk Access” permission being enabled.
Meta spokesperson Andy Stone rejected the claim, saying Muse could access Apple’s Messages app only when users explicitly opted in.
These separate incidents have brought Meta Muse AI privacy practices under greater scrutiny. They have also added to wider concerns about how AI products handle personal information and user permissions.
Meta’s Wider AI Plans
Meta is investing heavily in artificial intelligence and sees Muse as an important part of its future services.
The company aims to use AI across platforms such as Facebook and Instagram, with potential revenue from subscriptions and other services.
However, continuing Muse AI privacy concerns could affect public confidence if users become uncertain about how their conversations are stored and shared.
The decision to pause the feature shows that security reports can lead to changes in an AI product.
Still, Meta suspends sharing feature access only as an immediate response; further improvements may be needed to strengthen privacy protections.
