Meta recently introduced Muse, a new AI-powered personal assistant aimed at helping users with everyday activities such as managing emails, booking travel, and making purchases. Available initially in the U.S. via web, iOS, Android, and WhatsApp, Muse connects to users' apps and services, including calendars, payments, health, and smart home applications. The AI agent leverages Meta's Muse Spark model and integrates with payment platforms like Link by Stripe to facilitate secure transactions.
Users can selectively connect services to Muse, which operates with an opt-in approach and offers a free tier alongside two subscription plans priced at $20 and $100 per month. The assistant is designed to learn user preferences over time and continue functioning outside the app to provide proactive suggestions.
Muse represents Meta's significant move beyond traditional AI chatbots, focusing on performing tasks rather than just answering questions. The company highlights security measures, including running Muse on a dedicated virtual machine that isolates sensitive data such as passwords and payment information from Meta's advertising systems.
However, Meta's history of privacy issues, including past FTC settlements and data breaches, casts doubt on whether consumers will readily trust the new AI agent. Previous incidents like the Cambridge Analytica scandal and multiple regulatory penalties have affected public perception of Meta's data handling practices.
To address these concerns, Meta offers detailed technical documentation on Muse's security features and allows users to personalize the assistant's appearance and interaction style. Despite these efforts, the success of Muse will depend on whether users feel comfortable entrusting Meta with deeper access to their personal information in exchange for the convenience the AI agent promises.