A Single Malicious Webpage Let Researchers Hijack Meta Muse and Access Private Chats
NebuSec researchers demonstrated that a malicious webpage opened by Meta’s Muse personal AI agent could extract the user’s private conversations and stored memories to an external server without any approval prompt, then chain a flaw in Muse’s built-in browser with a Linux kernel flaw to take full control of the user’s cloud machine, reaching its credential store and monitoring layer. The work was a laboratory demonstration confined to a single user’s machine, and the firm says Muse’s team has fixed every weakness described.