AI Glasses and GPASS: Why Your Next Gadget Will Sit on Your Face, Not in Your Pocket
Chinese tech giants led by Ant Group (Alipay) have decided that pulling out a smartphone to pay for parking or find a guide is an unnecessary motion. The new GPASS architecture turns AI glasses into a proactive assistant. The device now understands on its own that you are standing at a parking barrier or looking at a landmark, and offers an action by voice. With the AI glasses market growing 370% a year, the industry is preparing for its "iPhone moment" as early as 2026. The key point: the focus has shifted from hardware to an ecosystem of services that literally follow your gaze.
AI-processed from 36Kr (36氪); edited by Hamidun News
Remember how we got used to paying with a phone instead of plastic? Soon the smartphone itself may start to feel like a clunky anachronism. In China, a new way of interacting with the world through AI glasses is being widely tested, and it is far more interesting than just a camera on the temple arm. It involves deep integration of the Alipay ecosystem into wearable devices through the GPASS (GPASS) technology framework. This is not merely "smart" glass, but an attempt to turn reality into an open game world where services are activated by your gaze.
Start with context that any resident of China will understand. The tradition of collecting "fortune" (Fu) characters for the New Year has gone from a family ritual to a tiresome quest of scanning bits of paper with a smartphone camera. This year Ant Group offered an alternative: owners of Rokid or Quark glasses can simply look at the symbols, and the system recognizes them and credits bonuses on its own. It seems minor, but it is a fundamental shift in interfaces. We are moving from a model where "a person looks for a service in an app" to one where "a service finds the person in the context of their life."
Take an ordinary situation in a parking lot. Usually it is a hassle: find your phone, open the right app, scan a code while leaning out of the window. With GPASS-based glasses, everything changes. The system understands that you have pulled into the lot and whispers in your ear: "The clock has started." When you leave, you only need to say "confirm" in response to a voice notification of the fee, and the barrier lifts. No hands, no extra motions.
The same happens in tourism. At a recent summit in Wuzhen, glasses with the digital guide "Tun Xiaowu" automatically told the history of the bridges as soon as the user simply looked at them.
Why does this work now, when there were attempts before? The secret is the combination of an agent architecture (Agent) and security. GPASS uses biometric identification by voice and gaze, which is critical for financial transactions. This is not a "dumb" gadget waiting for commands, but an active assistant. It analyzes the surroundings, anticipates a need and proposes a solution. If you walk into a pharmacy, the AI assistant "Ant Afu" is ready to advise on a medicine's composition without making you google it.
The numbers suggest we are on the threshold of a big boom. According to IDC, the AI glasses market grew 370% in the third quarter. Meanwhile, 80% of the global supply chain for such devices is concentrated in China. That means the path from prototype to mass-market product has shrunk to weeks. Rokid founder Zhu Mingming predicts that in three to five years glasses will become the main computing device, displacing smartphones. And 2026 looks like that very "iPhone moment" for the industry.
Of course, questions of privacy and habit remain. But when technology makes life seamless, when payment happens by a glance and navigation is drawn right on the asphalt before your eyes, resisting progress gets harder. We are entering an era where the world itself becomes the interface, and glasses are merely the key to it.
The key point: the success of AI glasses now depends not on lens resolution but on how useful the services built into them are. Are you ready to trust your wallet to your own gaze?
- Nov 2, 2026Karpathy builds a YouTube clone with 6454 videos from Hacker News recommendations in two hours
- Nov 2, 2026Karpathy used Claude to build HackerTube from 6,454 videos recommended on Hacker News
- Nov 2, 2026Arca gives AI agents finance tasks lasting up to 10 days, but leaves the trades to humans
- Nov 2, 2026DeepSeek open-sources Super-agent prototype: p95 task completion time of 15.7 minutes vs. 29.5
Test Frontier LLMs & Autonomous Agents in One Workspace
Run Claude Opus 5.5, GPT-6, Gemini 3.1, and DeepSeek side-by-side in Nomad AI (nomadnet.ai) — unified model routing, deep research, and multimodal workflows under one subscription.
The AI world, distilled — once a week
Seven stories that actually mattered, hand-picked. No noise, no reposts, no press releases.
Done! Check your inbox for a confirmation.