AI & Tech··1 sources verified
Meta's Internal AI Agent Muse Reportedly Mixes Up User Conversations and Exposes Private Data
Summarised by Relevant News AI · Read time: 3 min
According to a report, employees said Meta's AI agent Muse surfaced messages about childcare requests, career advice, product orders, and bank account balances that belonged to other people or were never requested. The incidents suggest the agent is confusing conversation context and leaking data across users. The report relies on employee accounts, so the scope of the problem is not yet clear.
Why it matters: Cross-user data leakage from an AI agent is a concrete failure mode that teams deploying agents with access to personal or financial data need to test for and guard against.
All sources