What Is Meta Muse? Meta’s New AI Agent Can Send Emails, Book Travel, and Shop for Users

Meta has launched Muse, a personal artificial intelligence agent designed to do more than answer questions by taking actions on behalf of users. The company says Muse can manage tasks and projects, create plans around long-term goals, work across everyday apps and continue operating after a user closes the app. It is initially rolling out in the US on iOS, Android and muse.ai.

The new agent is powered by Muse Spark, which Meta describes as its most capable AI model to date for real-world agentic work. Muse is built around a dedicated virtual machine called Muse Secure VM, which houses the agent and a user's data and is designed to give people greater control over what the system can access and do. Meta Tightens AI Glasses Rules After Reports of Secret Filming, Camera Shuts Off If LED Is Covered.

Meta Introduces Muse

How Muse Works

Meta says Muse is designed to work without requiring users to have technical expertise. People can tell the agent what they want accomplished, and it can develop a plan and carry out tasks on its own. The system can open a browser, fill out forms, send emails and book travel. For longer-running tasks, Muse can continue working after the user closes the app and return when something changes or when it needs approval.

For example, Meta says Muse could help sell a car, negotiate a lower bill or adjust a training plan as a person's circumstances change. The company also says users can communicate with Muse in a way similar to messaging another person, either through the Muse app or directly through WhatsApp. Meta Launches New AI-Powered Smart Glasses Lineup Without Ray-Ban Branding; Check Price, Features and Models.

Muse Can Remember User Preferences

A key part of the system is its ability to retain information about what matters to a user. Meta says Muse can use details mentioned in previous conversations to make suggestions and take action without requiring the user to repeat instructions.

For example, it can turn a recipe reel saved on Instagram into a grocery list, suggest a dinner-party menu and remember guests' dietary restrictions before invitations are sent. Users can also instruct Muse to forget specific information it has learned.

AI Agent Can Make Purchases

Muse is also designed to handle online purchases. Meta says the agent can check out using Link, the payment service built by Stripe. It is the first AI agent covered by Link's purchase protections, which include coverage for damaged or lost items, eligible price drops, no-fee returns and a return guarantee on eligible purchases.

Link's wallet for agents generates a one-time-use card, allowing users' actual card details to remain hidden during purchases. Meta says Shop Pay will be supported in the future, along with 1Password, which would allow Muse to use logins a person already has.

Muse Secure VM Focuses On Privacy

Meta says the architecture behind Muse was designed specifically for personal AI agents. Muse runs on its own dedicated computer in the cloud, where the agent and the user's connected data and credentials are stored. Meta says the environment is isolated so another user's agent cannot access it.

A separate Sentinel agent operates on the same machine but is isolated from Muse at the system level. According to Meta, Muse cannot access the internet unless Sentinel approves the activity, with user permission requested when necessary. The company says Muse also cannot see users' passwords or payment information. Credentials are stored securely so the agent can use them without being able to view them.

Users Approve Sensitive Actions

Muse is designed to require user approval before sensitive actions, including sending emails or making purchases. Meta says the system provides an audit trail showing users what Muse has already done and what it plans to do.

Users can decide which applications Muse connects to and determine how much access each service receives. For email, for example, they can choose whether Muse can only read messages or also send emails on their behalf.

Access can be changed or removed at any time. Users can also opt out of having their interactions used to train Meta's AI models, according to the company.

Meta says Muse conversations and information stored inside a user's Secure VM will not be shared with its advertising systems. The privacy architecture is intended to separate personal agent activity from the company's advertising infrastructure, an important distinction as AI assistants gain access to increasingly sensitive personal information. Meta says a further privacy feature is planned later this year.

Confidential VM Planned For Later This Year

Meta says it will introduce Muse Confidential VM later this year. Under that system, the entire virtual machine, including a person's data and conversations with Muse, will be encrypted using a key controlled only by the user.

Meta says this would mean that even the company itself would not be able to access the contents of the VM. The planned feature reflects the broader challenge facing personal AI agents: giving software enough access to act independently while limiting what the platform provider can see or control.

Muse Builds On Meta’s AI Push

Muse is the latest step in Meta's broader effort to develop personal AI systems capable of acting rather than simply generating responses. In July, Meta said Meta AI could use Muse Spark to make plans, connect with email and calendar applications, create slides and handle tasks on users' behalf.

Meta introduced Muse Spark earlier in 2026 as the first model in a new series developed by Meta Superintelligence Labs. The company said the model was designed to support reasoning and multimodal tasks and would form part of its push toward what it calls personal superintelligence.

Muse is rolling out in the United States on iOS, Android and muse.ai. Meta says the service is free for most users, with subscription plans available for people who want to do more. The company also plans to bring Muse to AI glasses, extending the agent beyond phones and computers.

Meta has increasingly positioned AI glasses as part of its personal AI strategy. The company has previously said its next generation of Meta Glasses will use Muse Spark-powered Meta AI from launch.

Meta says it believes personal superintelligence could become one of the most transformative technologies of a lifetime. Muse represents an early version of that vision: an AI system that can understand a person's goals, remember relevant preferences and carry out work rather than simply provide information. The company's approach also puts privacy and user control at the center of its pitch, particularly as agents gain the ability to access email, websites, payment systems and other personal services.

(The above story first appeared on LatestLY on Sep 09, 2026 09:38 AM IST. For more news and updates on politics, world, sports, entertainment and lifestyle, log on to our website latestly.com).