Meta reveals its AI agent that can shop, send emails and plan trips on your behalf

What Is Muse?

Meta has revealed its long-promised personal AI agent, called Muse, which is designed to complete a wide range of actions on users' behalf with "no technical experience" required to set up. The agent can perform tasks autonomously - including online shopping, drafting and sending emails, planning trips, and other productivity-focused tasks - if users trust it to do so.

CEO Mark Zuckerberg has previously stated his vision for Meta to create agents that work proactively to help people get things done, but in a more accessible format than other platforms like OpenClaw. Muse is more capable than the Meta AI assistant because it can handle complex tasks and work independently.

"Once a person shares a goal with Muse, it helps them develop a personalized plan and coordinate their time and resources, then advances the work on its own," the company explains in a blog post.

Muse has its own dedicated apps and website.

Capabilities and Integrations

Exactly what Muse can do depends on how much of a user's digital life they're willing to give it access to. Muse can connect to Meta's own apps, as well as a range of third-party services, including:

• Google Workspace
• Ticketmaster
• OpenTable
• Spotify
• Apple Health

For example, a user could ask Muse to look through all the recipe videos saved on Instagram and Facebook and organize them into a more readable format. Because it has shopping and browsing capabilities, Muse could then build grocery lists, compare prices, and order items on the user's behalf.

Muse is also able to set up its own connections to third-party services if a public API is available.

"Because it is a very capable AI product that can also build its own software, it isn't really limited in what it can do," Vishal Shah, Meta's VP of AI Products, tells Engadget. "From a complexity perspective, you just have to get more and more comfortable with giving it access to more, and also telling it what's important to you, so it can be the most helpful it can be."

Safety and Privacy

To actually do all that requires a fair bit of trust in Meta and the agent. The company says it has built in a range of safeguards to prevent its agents from going rogue and to protect users' data:

• Muse will check in for approval before doing things like sending an email or completing a purchase, though it can complete more "low-risk" tasks without users' signoff.
• Behind the scenes, Muse runs on a dedicated virtual machine meant to keep each person's agent - and the data it can access - separate from everyone else's.

Even so, agents can be unpredictable, and Meta says there will be some stumbles.

"Muse can and will still make mistakes, but we expect they'll be much less frequent and cause much less damage due to the safety systems we've built in," the company writes in a blog post.

On the privacy front, Meta is allowing users to opt out of having their interactions with Muse be used for further AI training. The company also won't share details from conversations with Muse to its advertising systems, though some interactions like shopping could "indirectly" influence the ads people see, according to Meta.

Availability and Future Plans

Meta is incorporating the agent into WhatsApp and its smart glasses to make on-the-go interactions easier. The company envisions a world in which users constantly have a Muse agent running in the background, plugging away at personal tasks and more "audacious" goals.

It's not yet clear at what point Muse could require a Meta AI subscription. The company says "most of what people need" will be free, but those who want to "do more" will need to pay.

The company has recently faced a privacy backlash over its AI glasses, and a controversial feature that allowed people to generate AI images from public Instagram posts (Meta rolled back the update after public outcry). Asking people to share so much of their lives with a Meta AI product could be a tough ask for some.