Instagram and WhatsApp's latest new feature Muse explained: How to use Meta’s personal AI agent to handle everyday tasks, shop and book tickets

Imagine telling an AI assistant to find a cheaper flight, organise your calendar, create a grocery list from an Instagram recipe and even complete an online purchase, then letting it continue working after you close the app. At the same time, there are many popular AI assistants taking over and tracking day to day tasks. Now Meta AI has rolled out personal AI assistants for WhatsApp and Instagram users, with a rather cute assistant which many would like to get their hands on. That is the idea behind Muse, Meta’s new personal AI agent. It goes beyond the question and answer model of traditional chatbots. Scroll on to find out how to add Muse to your smartphone devices

What is Muse? How is it different from ChatGPT and Google Gemini?

Unlike a conventional AI assistant, Muse is designed to take action on a user’s behalf. Meta says it can work through the Muse app and directly through WhatsApp. Muse is designed as a personal AI agent rather than simply a chatbot. Meta says users can ask it to complete multi-step tasks, browse the internet, fill out forms, make bookings, create documents, generate images and make purchases.

View this post on Instagram

A post shared by Muse (@muse)

The important difference is that Muse can continue working in the background. Meta describes it as an AI that can “take actions and get things done on your behalf”, rather than simply responding to individual questions. For example, a user could ask Muse to monitor flight prices, research options and help with a booking. It can also connect to services such as email and calendars.

How does Muse work on WhatsApp?

One of the biggest changes is that users do not necessarily have to open a separate AI application. Meta has designed Muse around messaging, letting people communicate with it through WhatsApp. Meta says users can message Muse like another person, tell it what they want accomplished and let it work through the necessary steps. This could make AI agents considerably more accessible because WhatsApp is already part of many people's daily communication habits.

Can Muse work with Instagram?

Muse can connect with Instagram and other apps, allowing it to use information from services that a user chooses to connect. One example given by Meta is turning a recipe saved on Instagram into a grocery list and using remembered information, such as guests' dietary restrictions, when helping organise an event. That integration is significant because it moves AI from generating suggestions to connecting those suggestions with real-world tasks.

View this post on Instagram

A post shared by Muse (@muse)

Is Muse allowed to make purchases on your behalf?

Muse can browse online shops and complete purchases, with Meta integrating Stripe's Link for payments. Meta says Link generates a one-time-use card so a user's actual card details are not exposed to the merchant. However, Muse is designed to ask for permission before sensitive actions such as purchases, sending messages or sharing information with connected services. Meta CEO Mark Zuckerberg has also indicated that commerce could eventually become an important part of Muse's business model.

View this post on Instagram

A post shared by Instagram (@instagram)

What about privacy and security while using Muse?

The convenience comes with an obvious question: how much access should an AI agent have to your digital life? According to Meta, Muse AI operates through a dedicated virtual machine and includes safeguards such as permission requests. An audit trail controls and allows users to disconnect services. Meta also said conversations and data in the Muse environment are not shared with its advertising systems.

Can Muse change how you use AI?

Meta's personalised AI assistant Muse shows a shift from “ask AI” to “delegate to AI”. Instead of asking a chatbot how to complete a task, users can increasingly expect the technology to perform parts of that task itself. That could make AI considerably more useful in everyday life, but it also means trust, permissions and oversight may become just as important as the technology's ability to get things done.