Live Feeds
● LIVE Updated 1h ago · 16 sources tracked

Meta Muse exposes users’ personal and business secrets when they share conversations

Meta has temporarily disabled the sharing feature for its Muse artificial intelligence assistant after security researchers discovered that private conversations were easily accessible to outsiders. When users shared chat logs with teammates, the system published them on public web pages that could be discovered without authorization. Security firm Claroc found that subdomain-discovery tools and passive-DNS tracking exposed sensitive data, including financial records, home-sale budgets, bank-fraud notes, and executive interview preparations. Meta announced the feature suspension following inquiries from Ars Technica.

🎙️

Listen to Live Briefing

Real-time synthesized voice briefing · Live Feeds Desk

⏱ ~3 min
Speed:
RSS Source map (16)
⚡ Key Developments & Real-Time Context
Text size:
  • ✓ Meta is suspending the Muse sharing feature while it investigates its privacy.
  • ✓ Claroc found confidential business records and personal financial details in a sample of shared conversations.
  • ✓ Conversation titles embedded in web addresses made the links easier to discover.
🛡️ Source Corroboration: 16 independent reporting domains (100% confidence) ⏱ Read time: ~2 min

What changed

Meta suspended the Muse conversation-sharing feature after independent researchers demonstrated that public links were easily discoverable through passive-DNS tools.

Live updates

  1. Meta Suspends Muse AI Sharing After Data Leaks

    Meta has temporarily disabled the sharing feature for its Muse artificial intelligence assistant after security researchers discovered that private conversations were easily accessible to outsiders. When users shared chat logs with teammates, the system published them on public web pages that could be discovered without authorization. Security firm Claroc found that subdomain-discovery tools and passive-DNS tracking exposed sensitive data, including financial records, home-sale budgets, bank-fraud notes, and executive interview preparations. Meta announced the feature suspension following inquiries from Ars Technica.

    Why it matters

    Mark Zuckerberg has positioned the Muse personal AI agent as a major driver for Meta to surpass traditional social media revenue and reach over one billion users. However, the rapid rollout of consumer AI assistants continues to expose tension between automated efficiency and basic user privacy. The discovery highlights how seemingly obscure technical implementation choices, such as embedding conversation titles directly into web addresses, can accidentally compromise confidential corporate documents and personal files.

    What is confirmed

    • Meta is suspending the Muse sharing feature while it investigates its privacy.
    • Claroc found confidential business records and personal financial details in a sample of shared conversations.
    • Conversation titles embedded in web addresses made the links easier to discover.

    What to watch next

    • Meta's formal security findings and remediation timeline for the Muse assistant
    • Whether shared conversation links already indexed by search engines are completely scrubbed
    Sources used for this update (17)
    1. www.meta.com — About Meta | Social Technology, VR, AR, and Innovation
    2. nypost.com — New York Post – Breaking News, Top Headlines, Photos & Videos
    3. docs.lovable.dev — Lovable changelog - Lovable Documentation
    4. arstechnica.com — Meta Muse exposes users’ personal and business secrets when they share conversations
    5. The Economist — AI-powered personal assistants could reshape the digital economy
    6. CBS News — Should you trust personal AI agents with your money and data? Here's what experts say.
    7. Vox — We all want to offload our most tedious tasks. But what do we lose in the process?
    8. Ars Technica — Meta Muse exposes users’ personal and business secrets when they share conversations
    9. Rest of World — Will developing nations be able to handle millions of AI agents?
    10. The San Francisco Standard — Everyone in San Francisco has an AI agent now. Privacy be damned
    11. note.com — The AI Giving Away 1 Billion Tokens. How Much Should We Entrust to It? | Vol. 5
    12. note.com — 10億トークンを配るAI。私たちはどこまで任せる？｜vol.5
    confidence 100%
📊

Community Sentiment: How do you assess this situation?

Voice your perspective · Real-time aggregated sentiment from the Live Feeds community