Meta’s AI glasses are about to become considerably more useful — and potentially a lot more demanding.

At Meta Connect 2026, Meta announced that its AI glasses will be powered by Muse, the company’s new personal AI agent designed to go beyond answering questions and actually get things done. The idea is simple: instead of reaching for your phone whenever you need something, you can ask your glasses to handle it for you.

Meta says it is working to bring Muse directly to its AI glasses so the agent can help users stay on top of tasks, manage projects and turn longer-term goals into actionable plans, all without requiring a phone or computer.

That changes the role of the glasses quite a bit. They are no longer just a camera, microphone and AI interface sitting on your face. With Muse, Meta is positioning them as a kind of always-available personal agent.

And, yes, that means your glasses could potentially be the thing you ask to do the boring stuff.

Muse wants to do more than answer questions

Meta introduced Muse earlier this month as a personal AI agent that can work across your phone, computer and the web. Unlike a conventional chatbot that waits for a prompt and responds, Muse is designed to carry out multi-step tasks on a user’s behalf.

On AI glasses, that agent becomes hands-free.

Meta says users will eventually be able to ask Muse to guide a personalized workout, log meals, provide nutrition information and even help purchase products they see. You could also ask it to book an appointment, check flight prices or find a replacement part while driving, cooking or exercising.

The particularly interesting part is that Muse does not necessarily have to finish the task while you are standing there waiting for it.

Meta says Muse can work in the background and check back when it has completed the request. Users can also activate it simply by saying its name and telling it what they need.

That makes the concept feel much closer to an actual assistant than the traditional voice-command model. You give it a job, get on with whatever you were doing, and let the agent handle the work.

Muse itself is designed to browse the web, complete tasks, make purchases, generate content and connect with external apps and services. Meta also says it can continue working in the background after the user closes the app.

The glasses are getting smarter in other ways, too

Muse is only part of Meta’s broader push to make its AI glasses more capable.

The company is also preparing a collection of new features focused on capturing and understanding the world around the wearer.

One particularly notable addition is Dolby Atmos Capture. Meta says its AI glasses will become the first consumer device capable of capturing video with Dolby Atmos, recording 360-degree spatial sound so compatible playback can provide a more immersive audio experience.

The glasses are also getting Dynamic Photo, which automatically captures multiple frames when taking a picture. Meta AI can then suggest the best frame, while users can still choose their preferred shot.

There is also a Landscape Crop Mode that lets users capture photos and videos horizontally instead of being locked into the usual portrait orientation. The option applies to photos, Dynamic Photos, videos, hyperlapse and slow-motion capture.

Meta is also bringing Private Processing to its AI glasses. The company says the feature is designed to enable richer AI experiences while preventing anyone, including Meta, from seeing the user’s data during that processing.

Meta is thinking about accessibility, too

Some of the upcoming features are less flashy but potentially more meaningful.

Meta says it is bringing an FDA-cleared hearing enhancement software feature to supported AI glasses for adults with perceived mild to moderate hearing loss. The company says it can be configured at home without a clinic visit or prescription.

The feature is scheduled to launch in the US later this year at $149.99, with HSA/FSA eligibility, or as part of a Meta One subscription.

Meta is also improving the glasses’ ability to understand and describe what users are looking at. The company says its work with blind and low-vision users has helped inform improvements to Meta AI’s visual understanding, with the goal of providing more detailed and context-rich responses.

Other additions include audio balance controls, mono audio and customizable temple controls that can be assigned to actions such as placing a call or opening an app.

The bigger shift is from assistant to agent

The most interesting part of Meta’s announcement isn’t necessarily another AI feature for glasses. It is the shift in what Meta expects these glasses to actually do.

The first generation of AI glasses largely revolved around questions and commands: ask what you’re looking at, get information, take a picture, make a call or play something.

Muse introduces another layer. The glasses can become the interface through which you delegate work.

That distinction matters because an agent doesn’t just tell you how to book an appointment — it can potentially go and do it. It doesn’t merely identify a product you’re looking at — Meta says it can help buy it. And rather than requiring you to stand there talking to your glasses until a task is finished, Muse can work in the background.

Meta’s broader Muse platform is built around exactly this agentic approach, including a secure virtual machine and browser that allow Muse to navigate websites and complete tasks.

For AI glasses, that could make the technology considerably more practical. The glasses already have something phones don’t: a first-person view of the world and microphones that are always positioned where the user is looking and speaking.

Combine that with an AI agent capable of taking action, and the glasses start looking less like a wearable version of a chatbot and more like an interface to a personal computer that you don’t have to hold.

That is the direction Meta is taking its AI glasses — and Muse is a pretty big piece of that puzzle.