California (Web Desk): US technology company Meta has temporarily suspended the chat-sharing feature in its artificial intelligence assistant agent, Muse, following the emergence of privacy concerns.
Cybersecurity researchers had warned that links to shared conversations could be discovered through public internet records, creating a risk that users’ sensitive personal and business information could fall into the hands of unrelated individuals.
According to the report, researchers at cybersecurity firm Claroc identified more than 6,800 shared links using passive DNS records. DNS records help track activity associated with domain names and can be used to identify certain web addresses.
The researchers said they accessed several hundred chats through these links. The conversations included company revenue assessments, loan applications, bank fraud alerts, partial account details and confidential notes concerning job interviews.
The report said the issue was related to Muse’s URL design, which included conversation titles and randomly generated identification codes. When these links were accessed, their web addresses could be stored in public databases, allowing other people to discover them even if the links had not been sent to them directly.
According to the researchers, accessing conversations through these links did not require users to sign in to a Meta account, further increasing the risk of private information being exposed.
Following the disclosure, Meta suspended the chat-sharing feature. The incident has raised important questions about data protection and user privacy in artificial intelligence-powered applications, particularly when conversations contain sensitive financial, personal or business information.