Meta's Muse AI agent asks for deep data access — and consumer trust

Meta is rolling out Muse, a personal AI agent designed to handle a wide range of tasks by tapping into users' email, calendars, payment systems, and even health services. It's the company's most ambitious consumer AI product to date, but it hinges on a question that has dogged Meta for years: will people hand over sensitive data to a company with a checkered privacy history? The launch, announced this week, positions Muse as a digital assistant that can schedule appointments, track spending, and coordinate with healthcare providers — but only if users grant it broad permissions.

Muse represents a strategic shift for Meta, which has mostly focused its AI efforts on chatbots and recommendation algorithms. With Muse, the company is moving into the realm of personal productivity tools, directly competing with the likes of Apple's Siri, Google's Gemini, and startups like OpenAI's ChatGPT. But unlike those rivals, Meta brings baggage: decades of privacy scandals, from the Cambridge Analytica fallout to repeated fines for mishandling user data. That history could make consumers hesitant to connect Muse to their most personal accounts.

The agent's capabilities are impressive on paper. Meta says Muse can draft emails, parse calendar conflicts, flag unusual charges, and even remind users about prescription refills. To do that, it needs access to third-party services like Gmail, Outlook, and banking apps. Meta insists that all data processing happens on-device or in encrypted environments, and that users retain control over what Muse can see and do. But the company has made similar promises before, and critics are already asking how Muse's data practices will differ from Meta's ad-targeting operations.