Smart glasses are becoming more capable, but putting cameras and AI assistants on someone’s face comes with an obvious concern: what happens to the privacy of everyone around the wearer? Google is addressing that question ahead of the launch of its new Intelligent Eyewear, outlining a combination of hardware and software protections designed to safeguard both wearers and bystanders.

In a blog post published on November 12, Shahram Izadi, Google’s VP and General Manager of XR, explained how the company is approaching privacy as it prepares to bring Gemini-powered eyewear to consumers. The protections cover everything from encrypted camera data and automatic face redaction to physical recording indicators that cannot simply be covered up.

Google says its goal is to make AI-powered glasses useful without making people uncomfortable about being around someone wearing them.

Google wants people to know when they’re being recorded

One of the most visible privacy features is an outward-facing LED indicator built into every pair of Intelligent Eyewear. When someone takes a photo or records a video for their gallery, the light illuminates to notify people nearby.

More importantly, Google says it has designed the indicator to resist tampering. Sensors can detect when someone covers or interferes with the recording light, preventing the glasses from capturing photos or videos in that situation.

This is an important distinction because a recording indicator is only useful if wearers cannot easily disable it while continuing to use the camera.

Google is also introducing responsible-use requirements during device setup. Users must agree not to record people without permission, record in sensitive locations, or cover the indicator light. Violating these requirements could result in the user’s ability to use the glasses being disabled under Google’s Terms of Service.

Gemini automatically redacts bystanders’ faces

Recording indicators are only one part of the privacy equation. Google’s glasses can also use their cameras to understand the world around the wearer, which means images may contain people who never agreed to interact with an AI assistant.

To address this, Google says the glasses are designed to automatically redact bystanders’ faces before camera visuals are processed by Gemini.

The company also says Gemini is designed to focus on the wearer’s voice by default rather than picking up surrounding conversations. Directional microphones built into the glasses help prioritize the wearer’s speech.

These protections are particularly relevant when someone asks Gemini to translate a menu, read a sign, identify an object, or provide information about something in their surroundings.

Camera visuals aren’t saved in Gemini History

Google is also making specific commitments about how camera data is handled.

When users point their glasses at something and ask Gemini a question, the camera visuals used to generate the response are not saved in Gemini History. Google also says these visuals are not used to train its AI models or target advertisements.

Data stored directly on the glasses is encrypted. Information transmitted to a connected phone or Google’s cloud infrastructure is also encrypted while in transit.

Google additionally plans to bring its Private AI Compute technology to Intelligent Eyewear, providing a secure environment for processing data alongside the existing protections.

However, there is an important distinction between camera visuals and conversations. Text-based Gemini chats are saved in Gemini History, allowing users to revisit previous interactions and receive more personalized assistance. Users can review or delete individual conversations, clear their entire history, and control whether chat text can be used to improve generative AI models.

Google is also giving users control over which third-party applications can access the glasses’ cameras. Apps must request permission before accessing the camera for the first time, and users can revoke that permission whenever they want.

For situations where someone simply wants to wear the frames without their smart features, every pair includes a physical power switch that completely turns off the technology.

Google is restricting apps that could misuse the cameras

Google’s privacy approach extends beyond the glasses themselves. The company is introducing restrictions across its services and platforms to discourage developers and users from bypassing the safeguards.

Google Play policies prohibit applications from using the glasses’ cameras for social livestreaming or real-time facial identification. Apps are also prohibited from changing how the external recording indicator works.

Google Shopping and Google Ads policies will not allow the promotion of products or services designed to disable or hide recording indicator lights on smart glasses.

YouTube’s existing content policies also apply to footage captured with the eyewear. Videos depicting harassment or unwanted sexualization can be removed, regardless of whether they were recorded using a smartphone, conventional camera, or smart glasses.

Google is also restricting device setup to compatible Google Accounts belonging to users aged 18 or older.

Privacy will be a bigger challenge as AI glasses evolve

Google’s announcement comes as the company prepares to introduce its new Intelligent Eyewear collections developed with Warby Parker and Gentle Monster. The glasses combine Gemini assistance with cameras, open-ear speakers, and other connected features, making privacy an important part of the product experience rather than an afterthought.

The company has clearly considered two different perspectives: the wearer who wants convenient AI assistance and the bystander who doesn’t necessarily want to appear in someone’s recordings or have their surroundings analyzed by an AI model.

Still, these protections will need to prove themselves in everyday use. Automatic face redaction, directional microphones, and tamper-resistant recording indicators are promising measures, but they cannot eliminate every privacy concern associated with camera-equipped eyewear.

For Google, the bigger challenge will be convincing people that having Gemini on their glasses doesn’t mean giving up their privacy. And that includes people who aren’t wearing the glasses at all.