Summary
Google says its intelligent eyewear combines an outward-facing recording light, automatic face redaction, encryption and user-controlled app permissions. The company also says camera visuals used with Gemini are not saved in Gemini History or used for AI training or ad targeting.
Google has outlined privacy features for its intelligent eyewear, which uses Gemini to answer questions about what the wearer sees. In a post dated November 12, 2026, Google says the design combines controls for the wearer with signals and protections intended to address bystanders’ concerns.
The company says camera visuals used when asking Gemini about the surrounding world are not saved in Gemini History and are not used to train AI models or target ads. Google describes several hardware and software measures behind those commitments, while noting that a separate planned processing feature for the glasses is still to come.
How the eyewear handles images and data
When a wearer asks Gemini about something in view, such as a menu, the glasses are designed to redact bystanders’ faces before sending images to Gemini for processing. Google says the camera visuals are used to provide an answer, rather than being added to Gemini History. If a user asks the glasses to take a photo or video, however, that media is stored in the device’s gallery.
The distinction matters because Gemini conversations are handled separately: Google says chats are saved in Gemini History so users can return to them and receive more personalized help. Users can review or delete chats and control their history and personalization settings, including whether chat text is used to improve generative AI models.
Google says data stored on the glasses is encrypted. If data leaves the glasses for processing on a phone or in the cloud, it is encrypted in transit. The company says it plans to bring Private AI Compute to the glasses, describing it as a secure processing environment that would supplement the other safeguards.
Signals and controls for people nearby
An outward-facing LED lights up when the wearer takes a photo or video for the gallery, alerting people nearby. Google says sensors are designed to detect if the light is covered or tampered with and prevent photos or videos from being taken in that case.
The glasses also use directional microphones that Google says are designed to focus on the wearer’s voice by default, rather than other voices nearby. Apps must ask for permission before accessing the glasses’ camera, and users can grant or revoke access for individual apps.
Google’s rules add restrictions beyond the device itself. Users must agree not to record others without permission, record in sensitive places or cover the indicator light; violating these requirements can result in losing the ability to use the glasses. Google Play policies also prohibit apps from using the camera for social live-streaming or real-time facial identification, or from changing the indicator light’s behaviour.
The eyewear requires setup with a compatible Google Account for users aged 18 or older. Google lists availability in the United States and South Korea.