Meta has launched Muse, a personal AI agent that can be used through its app or directly on WhatsApp , to work towards larger goals and make suggestions on its own. Mark Zuckerberg unveils Muse Zuckerberg announced on X, “Introducing Muse, the personal agent that understands your goals and works 24/7 to get things done for you.” Muse can do more than respond to prompts. According to Meta, it can take over tasks, allowing users to give it instructions in a conversational format. Introducing Muse, the personal agent that understands your goals and works 24/7 to get things done for you. — Mark Zuckerberg (@finkd) September 8, 2026 Features of Muse The personal AI agent can handle tasks such as sending emails and booking travel, as well as take on larger goals. Users can tell Muse what they want to achieve, after which it can make a personalised plan, coordinate time and resources, and continue working towards the goal on its own. It can even open a browser, fill out forms and negotiate on a user’s behalf. It can continue working even after the user closes the app and return when something changes or when approval is required, such as before sending an email or making a purchase. Meta said Muse will help users get better results with less effort. For instance, selling a car for more, lowering a bill or modifying a plan, etc. Muse has been built around Muse Secure VM, a dedicated security system designed to house both the agent and a user’s data. Also, it runs on its own dedicated computer in the cloud, isolated from other agents. User data and credentials for connected services are stored within the system. “Muse is built from the ground up for privacy and security. Your data and credentials live on the Muse Secure VM -- an isolated linux computer with a browser, CPU, memory, and storage,” Zuckerberg said. A separate Sentinel agent runs on the same machine but is separated from Muse at the system level. Meta claims that Muse cannot access the internet unless Sentinel approves the action. It would ask the user for permission when required. The company also said Muse does not have visibility into users’ passwords or payment methods. Users can decide which apps Muse connects to and how much access it receives. Even a user can decide whether Muse can only read messages or also send emails on their behalf. Moreover, access can be changed, or a connected service can be disconnected at any time. Users can also opt out of having their interactions used to train Meta’s AI models. Also, it does not share users’ conversations or the data stored in their virtual machine with its advertising systems. Interestingly, the company claims that users can ask Muse to forget specific information it has learned. Also read: ‘AI could kill us all’: Anthropic researcher Jacob Coxon quits over fear of “out of control” AI systems
This is a summary. For the full story, refer to the original source. Public of India presents verified citizen-relevant reporting.