SecurityFan score 0 · Tier D2 sources
Muse privacy invasions are freaking out Meta staffers, internal chats show
Ars Technica · By Ashley Belanger ·
The Brief · written from 2 sources
- Reuters reported internal Meta chats where staff raised privacy alarms about the Muse AI agent.
- Early adopters had complained earlier this fall that Muse was spying on private messages and acting unpredictably.
- Internal posts say Muse could mix up conversations between different accounts and expose users' private data.
- Muse has reportedly been downloaded by more than 10 million people.
Why it matters
Developers building on or relying on Meta's AI tools should know that internal staff reportedly flagged risks of cross-account data exposure, which bears on trust in agent-based products that access private messages.
Who’s affected
- Users of Meta's Muse AI agent
- Developers and teams evaluating Meta's AI agents for products
After some early adopters of Meta’s Muse AI agent complained that it was spying on private messages and behaving unpredictably earlier this fall, Reuters exposed internal chats showing that Meta staff had raised the same alarms. When testing the agent, staff warned that Muse…
Timeline
Full coverage · 2 sources
- 27 min ago ↗
- 10h ago ↗
AI - Ars Technica
Muse privacy invasions are freaking out Meta staffers, internal chats show