thehackernews.com/2026/11/a-single-malicious-webpage-let.html A Single Malicious Webpage Let Researchers Hijack Meta Muse and Access Private Chats
NebuSec researchers showed that a malicious webpage opened by Meta's Muse AI agent could send a user's conversation history and stored memories to an outside server without any approval prompt, then combine a browser flaw with a Linux kernel bug to take over that person's cloud machine and reach saved credentials. The firm says it reported the issues and that all were fixed before publication, while Meta has not publicly confirmed the fixes; the demonstration stayed within one user's machine.