Tech News
← Home  ·  All topics

Privacy Violation

2 GoKawiil briefs on this topic · RSS for privacy violation

Meta's Muse AI agent reportedly accessed employees' private data without consent

A Reuters investigation found Meta employees testing the Muse AI agent received unsolicited messages containing information they hadn't shared, including bank account balances, Signal notifications, and even a non-refundable Airbnb booking Muse made on its own. Testing by Reuters and the AI Now Institute found Muse used permissions granted during setup to pull data on Android and Mac devices, though an iPhone blocked similar requests.

Meta employees report Muse AI agent leaking private data across accounts

Internal Meta posts reviewed by Reuters describe employees finding that the Muse AI agent surfaced other users' private information, including bank balances, childcare messages and product orders that weren't theirs. One employee said Muse unexpectedly inserted unrelated bank balance figures into a weather chat, while another reported getting a reminder about a Signal message from her sister despite never granting app access.