A decade after Google launched Google Glass to spectacular failure, the company is trying again with a new prototype that leverages its Gemini AI assistant. The difference this time may be that society is more receptive to smart glasses, thanks in part to the acceptance of similar devices from Meta. At Google I/O, I had the chance to test a prototype made by Samsung that combines elements from partnerships with Warby Parker and Gentle Monster. The glasses are not a final product but a vehicle to demonstrate what Google hopes to achieve in connected eyewear.
The original Google Glass, which debuted in 2012, was met with public derision. Wearers were often called 'Glassholes' due to the outward-facing camera that could record people without their consent. That stigma has faded significantly, as connected devices like the Meta Ray-Ban Stories have become more commonplace. Google’s new glasses hope to capitalize on this shift by offering a more refined, subtle design that looks like ordinary sunglasses. The camera is discreetly placed in a small circle on the front, and there is a button on the right frame for snapping photos. The right side also houses a touch-sensitive surface for swiping and tapping to activate Gemini or control volume through integrated speakers.
The first thing I noticed is how lightweight and comfortable the glasses are. There is no extra bulk or thickness, and they feel no different than a typical pair of shades. This is a major improvement over earlier smart glasses, which often felt clunky. The audio from the speakers was a bit faint in the noisy demo environment, but that is likely a prototype issue. The heads-up display (HUD) is minimal, appearing as a small window in the corner of the user's field of vision. It currently shows basic information like reminders, navigation maps, and AI responses. The map feature I tested showed an overall view of my surroundings rather than turn-by-turn directions, but Google indicated that this will evolve.
The core function of the glasses is to act as a hands-free interface for Google’s Gemini AI. Instead of pulling out a phone, you can ask Gemini to identify objects, play music, set reminders, or explain concepts. For example, you could look at a restaurant and ask for its hours, or glance at a poster to get ticket information. The AI processes visual data through the camera and provides audio responses through the speakers. This eliminates the friction of unlocking a phone, opening an app, and typing. The glasses connect to your phone via Bluetooth and rely on your phone’s cellular or Wi-Fi connection for data. Google confirmed that no facial recognition is being built into the prototype, addressing a major privacy concern.
Despite the convenience, there are limitations. Some AI features feel forced, such as applying anime filters to photos taken with the glasses. Battery life remains a question mark, as Google did not provide details. The HUD is currently basic and may not provide enough information for complex navigation. Privacy is still a concern, though the glasses require a tap or a wake word ('Hey Google') to start recording. Google says an LED indicator will be present in the final product to signal when the camera is active. Still, having a conversation with someone while quietly issuing voice commands to an AI assistant may feel awkward to many users.
In some scenarios, the always-on potential of AI glasses could be a boon. Imagine walking through a museum and having the AI spontaneously share details about a painting you are looking at, or getting real-time translation of signs and menus. These are possibilities that Gemini opens up. However, the current iteration is more of a tool for occasional queries rather than a constant companion.
The prototype I tried is a significant step forward from the original Glass. It feels less like a science experiment and more like a practical accessory. The design is unobtrusive, and the AI integration is promising. However, success will depend on whether Google can address battery life, refine the HUD, and ensure that the features genuinely enhance daily life without becoming intrusive. The smart glasses market is still nascent, and Google’s entry with Gemini could redefine how we interact with AI in the physical world. For now, the glasses remain a prototype, but they offer a glimpse of a future where AI is always just a glance away.
Google is also working on connected glasses with Warby Parker and Gentle Monster that will start with audio-only mode and later add a small heads-up display. The Samsung-made prototype blends these technologies into a single device for development and testing. The lack of official specifications or a release date suggests that Google is still iterating. The company wants to avoid the mistakes of the past, where hype outweighed the actual utility. The new glasses aim to be subtly helpful rather than flashy.
From a hardware perspective, the glasses include a camera, a small button for photos, a touchpad, and speakers. The camera captures what the user sees, and Gemini processes the images for context. The touchpad allows for swiping to scroll through notifications or volume control. The button can be used to take a photo or activate the AI. The speaker delivers audio prompts without blocking ambient sound, allowing for natural interaction. The lack of a screen in the main lens keeps the glasses lightweight, but the HUD in the corner could be expanded in future versions.
The success of Google’s AI glasses hinges on public acceptance and practical utility. Meta has shown that there is a market for smart glasses, albeit with controversy around recording. Google hopes that its privacy safeguards and seamless AI integration will win over consumers. The prototype I tested demonstrates that the technology can be both powerful and unobtrusive. Whether it will become a mainstream product remains to be seen, but the potential is undeniable. As AI continues to evolve, the glasses could become an essential tool for navigating the world, offering instant information without the need to pull out a phone.
Source: PCWorld News