One Hidden Meta Muse Setting Could Let Attackers Turn the AI Assistant Into a Backdoor
Summary: Security researcher Patrick Wardle released a proof-of-concept on September 21 showing that malware already running on a Mac can change an undocumented Muse preference, endo_voyager_dictation_endpoint, so that dictated prompts go to an attacker-controlled address instead of Meta. The attack requires code already running as the logged-in user, and Wardle showed it can read dictated text, inject instructions Muse acts on, and capture the account token to control Muse on other signed-in devices, including an iPhone.
Key Takeaway: Muse was granted access to a user's files, email, messages, calendar, shopping and smart-home apps, and an attacker who steers it inherits all of that while the commands appear to come from a normal signed app that security software may not flag.
Read the original at The Hacker News ↗
- Topics
- Malware
- Related
- ARTEX and beyond: AI-enabled penetration testing platforms Cyber Risks · Nov 13, 2026Fake GTA 6 download goes after Steam logins and crypto wallet data Cyber Risks · Nov 13, 2026Operation Endgame Takes Down Gootloader, CastleRAT and StealC Servers, Seizes 6.4M Stolen Logins Cyber Risks · Nov 13, 2026