Skip to content
NewA calmer, faster ZeroHour

Lab demo hijacks Meta Muse from one malicious webpage

Original title: A Single Malicious Webpage Let Researchers Hijack Meta Muse and Access Private Chats

The Hacker News ()2 min read1
Part of a story covered by 3 sourcesMeta says Muse privacy mix-ups were a now-fixed prompt bugMerged summary & timeline
AI summary · x-ai/grok-4.7Summaries by Vani AI

Researchers chained a browser flaw to a Linux kernel flaw and reached credentials on one user’s cloud machine.

NebuSec showed that a webpage opened by Meta’s Muse agent could extract private conversations and stored memories and alter them without an approval prompt. The same lab chain reached credentials for connected accounts, saved files, and the Sentinel monitoring layer by combining a built-in browser flaw with a Linux kernel flaw. The firm says the break stayed on that one cloud machine and that Muse fixed every issue before publication. Meta has not publicly confirmed the fixes.

Unconfirmed: Meta has not confirmed the fixes, and the write-up does not say whether user data was accessed before patching.

Key facts
  • Muse launched on September 8 and runs each agent on its own cloud computer.
  • The attack did not reach the shared server or other users’ machines.
  • An earlier September flaw only let local Mac malware hijack the on-device assistant.
  • NebuSec says it reported the findings and the exploits were fixed before publication.
Security assessment · AI, from the articlePublic proof of concept

Impact: Private chats, memories, and connected-account credentials

ProductAffected
Meta Muse—
Explain this AI news with Vani AIWhat was announced, what is new and why it matters, in a new chat

From the original article

2 min read at thehackernews.com

A malicious webpage could extract a Meta Muse user's private conversations and stored memories from the assistant without triggering any approval prompt, researchers have shown. The same attack could be extended to reach the credentials Muse stores for the user's connected accounts. The work was a laboratory demonstration, not an attack on real users, and it stayed inside the individual user's

The Hacker News

thehackernews.com

Read the full article at The Hacker News

A short excerpt; the article and its images belong to the publisher. Summary, key facts and assessment above are AI-written from the article and can be wrong.