Google is pushing deeper into the next phase of Android XR, the platform developed with Samsung and Qualcomm, where Gemini is increasingly becoming the core interface across headsets, glasses, and other wearable devices.

At Google I/O 2026, the company expanded its vision for intelligent eyewear: glasses designed to deliver assistance in real time without pulling you out of the moment. Two categories are emerging. Audio glasses, which provide spoken guidance directly in your ear, and display glasses, which surface visual information in your field of view exactly when you need it. Both are built around hands-free interaction with Gemini, activated simply by voice or a tap on the frame.

Audio glasses will arrive first, with availability expected later this fall. They represent the first consumer step into this new category.

The pricing has not been revealed so far.

Designed to be worn all day

A key focus is wearability. Google has partnered with Samsung and eyewear brands Gentle Monster and Warby Parker to ensure these glasses are not just functional, but also stylish enough for everyday use. Early designs were previewed as part of upcoming full collections from both fashion partners, signaling a push to make intelligent eyewear feel like regular eyewear first.

Gemini at the center of the experience

Once worn, the experience is built around Gemini. Users can say “Hey Google” or tap the frame to instantly activate it and interact naturally.

The capabilities go well beyond simple voice commands:

You can ask about what you are seeing in real time, such as identifying landmarks, decoding signs, or checking restaurant reviews as you walk past them. Navigation becomes context-aware, with turn-by-turn directions based on your exact position and orientation, plus the ability to adjust routes or find stops mid-journey.

Communication also becomes hands-free, with support for calls, messaging, and message summaries delivered directly through the glasses. Audio playback is designed to adapt to your environment, creating a more ambient listening experience.

The camera experience is also tightly integrated. You can capture photos and videos instantly, then use Gemini-powered tools to edit them. One standout capability is the integration of Nano Banana, which enables quick transformations like removing background distractions or adding playful edits such as generating themed overlays from a simple voice command.

Translation is another major pillar, covering both speech and written text in real time, while preserving tone and natural voice characteristics during conversation.

Beyond that, Gemini can handle multi-step tasks in the background. For example, placing a DoorDash order while you continue your day, only requiring final confirmation. The system is also designed to connect with third-party apps like Uber and language learning tools such as Mondly, extending functionality across both Android and iOS ecosystems.

A broader ecosystem play

This launch is not just about hardware. It is clearly positioned as an ecosystem shift. By integrating the full Google suite along with third-party app support like Uber, the glasses move closer to becoming a general-purpose interface layer for daily digital tasks rather than a standalone gadget.

That broader integration is one of the key differentiators being emphasized.

Positioned against Meta’s wearables push

This move also reads as a direct response to Meta’s growing presence in the category, particularly products like the Ray-Ban Meta smart glasses and its experimental display-based eyewear efforts. Where Meta has focused on early lifestyle-first smart glasses, Google is pushing a more deeply integrated AI system built around Gemini and the Android ecosystem.

The addition of Nano Banana-style generative editing, combined with tight app integration across services like Uber and other third-party platforms, strengthens Google’s argument for a more capable, action-oriented wearable assistant.

The open question: real-world performance

What remains to be seen is how well all of this works in practice. The promise is compelling: always-available AI, context-aware assistance, and seamless task execution without a phone in hand. But the real test will be output quality in real time, especially for translation accuracy, visual understanding, responsiveness, and how reliably these multi-step actions execute in everyday environments.