Eigen RadarAI
Analysis

Meta's Muse acts for users by connecting to their personal data

Meta launched Muse, a personal AI agent in the United States. It connects to email, calendars, payments, shopping and smart-home services and can keep working after the user leaves. Meta emphasizes a dedicated secure virtual machine and approval controls, but the product's usefulness arrives together with access to a person's most sensitive services.

Artificial Intelligence··Morning
In a bright home workspace, a seated person studies a face-free faceted light agent among a separate envelope, unbranded payment pouch and home-control device.

From conversation to carrying out tasks

Meta is not presenting Muse, launched in the United States, as merely a chatbot that answers questions. TechCrunch and Associated Press report that the agent can send email, manage calendars, shop and book travel on a user's behalf. It can also form a plan for longer goals and continue working on it. Muse is available through separate iOS and Android apps, muse.ai and a WhatsApp conversation, with the initial service limited to people aged 18 and over in the United States.[1], [2]

Connections expand usefulness and exposure together

Muse can connect not only to email and calendars but also to payments, health and fitness, smart-home, dining, shopping and music services. TechCrunch reports that when a service lacks a built-in connector but has a public API, Muse can connect using credentials supplied by the user; without an API it can use the browser. Checkout is completed through Link by Stripe. This design lets one instruction become action across several accounts. For the same reason, users must choose carefully which services Muse can reach and which actions require separate approval.[1], [2]

The security promise rests on a separate workspace

Meta says Muse and the user's data are housed in a dedicated virtual machine called Muse Secure VM. Associated Press places that security and privacy claim at the centre of the launch, while TechCrunch reports that the agent can keep working after the user leaves and learns from conversations what matters to them. Use begins free, with Power at 20 dollars a month and Maximum at 100 dollars as usage rises. Its everyday value will depend not only on the access granted, but also on how clearly activity records and approval moments are shown to the user.[1], [2]

References

  1. News sourceTechCrunchMeta's Muse agent connects to email, payments and the smart home in the US↩1↩2↩3
  2. News sourceAssociated PressMeta launches personal AI agent Muse with an emphasis on safety and privacy↩1↩2↩3