At the end of 2023, startup company Humane released an AI device called AI Pin, which can be attached to clothes and connected to proprietary large models developed based on GPT-4, and interact through projection. AI Pin sounded the new horn of wearable device reform. Since 2024, many technology giants such as ByteDance have begun to enter the wearable field, and various new wearable products loaded with AI functions such as smart rings and smart pendants have been unveiled. Wearable devices, as widely used mobile terminal hardware today, are closest to the human body and most closely integrated with daily life. With the support of AI technology, they may replace hardware such as mobile phones and computers in some scenarios, realizing AI edge device reconstruction and becoming the most important carrier for large-scale model landing. Wearable devices are undergoing disruptive changes, and the market prospects are promising. With the continuous development of technology, wearable devices are facing disruptive changes. Firstly, in terms of product form, new wearable devices are showing a trend towards invisibility, lightweight, and flexibility. The continuous improvement of microelectronics technology and electronic chip integration has accelerated the miniaturization process of electronic devices in design, and wearable devices are becoming smaller and even invisible. At the same time, technology manufacturers prioritize lightweight design in order to increase user stickiness and frequency of use. In addition, in order to better integrate with the human body and improve the efficiency of information transmission, the research and development of flexible electronic materials and flexible interfaces for electronic devices has also become an important trend and direction in the design and development of wearable devices. Secondly, in terms of interaction mode, new wearable devices will completely overturn traditional human-computer interaction. The traditional interaction method of hardware terminals is through physical contact (buttons, touch, etc.), accompanied by more advanced language understanding and multimodal processing technology. New wearable devices can interact through voice, gestures, images, heart rate, and even thoughts, while processing both voice and visual information, fully understanding the user's needs and providing the most suitable assistance, achieving truly natural human-machine dialogue. Thirdly, on the technological path, new wearable devices are gradually separating from the mobile phone ecosystem and opening up independent terminal technology routes. Technology giants such as Apple and Samsung have always regarded wearable devices as an extension of mobile phones, and information push relies on communication between Bluetooth or networks and mobile phones. And a large number of startups have begun to develop and continuously launch wearable devices that no longer rely on smartphones, such as Humane releasing AI Pin, aiming to bring users a world without smartphones through AI technology and wearable computing; Rabbit R1 is positioned as an artificial intelligence assistant, without the need to connect to a mobile phone. On the day of its release, orders reached 10000 units; OPPO has partnered with WeChat to launch the WeChat Watch Edition, which allows for independent use of WeChat without the need for mobile phones, marking a milestone for smartwatches to move towards independence. Wearable devices are expected to become the core platform for AI application innovation. After the outbreak of generative AI, terminal manufacturers are constantly exploring ways to combine terminal products with AI in order to unleash greater product potential. From the dynamics of foreign technology giants, it seems that there is a consensus that AI landing requires new hardware platforms, and wearable devices are the most imaginative hardware series. Meta、Google、 Leading companies in the AI field such as Microsoft and OpenAI are trying to integrate their vision and language related AI technologies into smart glasses and other wearable devices with cameras. For example, the "RAY-BAN | Meta" series of glasses has launched AI functionality; Oura Ring launches a personal AI health consultant that can provide users with short-term and long-term health advice; Opuz Ring combines new sensing technology with artificial intelligence to monitor users' blood sugar at any time without the need for needles, skin patches, or finger punctures... Firstly, AI+wearable devices have a stronger data Matthew effect. The multimodal capability endows wearable devices with a unique value proposition compared to other devices. As the closest device to the human body, it can use ECG and other diverse sensors to collect a large amount of data about the user and the environment, which other devices cannot achieve. After integrating AI technology, it is no longer just about collecting and displaying data, but has the ability to reason and solve problems at the human level. For every additional sensor added to a wearable device, AI can create dozens of new application scenarios based on that sensor. The more user data obtained, the more understanding one has of the user, the more innovative application scenarios can be created, and the deeper the services that can be provided to the user. Secondly, wearable devices may become the most convenient entry point for AI. To achieve practical application of AI models, they need to be deeply integrated with devices. From the current trend, the relationship between AI and users mainly appears in the form of AI agents, which help make decisions by perceiving users and their surrounding environment. The integration and large-scale implementation of AI and wearable devices still face bottlenecks. From the current application situation, the three mainstream terminal products of TWS earphones, smartwatches, and AR smart glasses have all been equipped with AI technology. However, AI+wearable devices are still in the market exploration stage of various "small devices" growing freely, and there are still obstacles to scale. One is the overlapping functions of intelligent wearable devices, which belong to exploratory products. Bracelets, watches, rings, and other products all focus on social functions as well as health monitoring functions such as breathing, heart rate, and sleep. Although they come in different forms, their functional cores are not significantly different. It can be said that it is still unclear what form of device will become popular and profitable in the future. Manufacturers are constantly testing market reactions in the process of launching new forms of products. Secondly, the product positioning of smart wearable devices is lower than that of smartphones and PCs, lacking a dedicated main force for expanding territory. Wearable devices are mostly manufactured by mobile phone manufacturers. Data shows that the top five wearable device manufacturers in the world are Apple Imagine Marketing、 Xiaomi, Samsung, and Huawei have a combined market share of over 50%. These manufacturers mostly develop wearable devices based on the extension of mobile phone functions. In this situation, smart wearable devices lack their own "main force", and their research and development are inevitably subject to the squeeze of mobile phone manufacturers' main business, which affects the release of market potential. Thirdly, the computing power on the end side remains the biggest bottleneck for AI to be implemented in wearable devices. At present, the problem of insufficient computing power on the end side has become the main limiting factor for the development of end side AI. Even on mobile phones and PCs, the computing power of NPUs cannot fully meet the smooth operation requirements of end side large models, let alone wearable devices with higher requirements for size, power consumption, cost, and other aspects. Previously, so-called "native AI devices" including AI Pin and Rabbit R1 have encountered various problems such as inaccurate recognition, slow response, and overheating, proving that under current technological conditions, insufficient computing power on the end side will have a direct impact on user experience, leading to a significant decline in product market competitiveness. In recent years, the country has attached great importance to the development of the wearable smart device industry and formulated a series of policy measures to promote its research and application. Under the dual drive of policy and market, operators should actively layout and strengthen their role in the wearable device industry chain, actively seize and create new growth poles in the new track. One is to rely on the AI big model base and become a leading supplier of AI applications, providing innovative AI applications for wearable devices; The second is to actively layout and enter the hardware product service track of AI wearable through customization, investment, and cooperation, and to provide end-to-end service capabilities throughout wearable devices; The third is to leverage the ecological advantages of cloud services to provide AI suppliers of wearable devices with basic services such as computing power and storage for cloud generated AI applications; The fourth is to strengthen the shaping of a secure corporate image, solve the numerous sensitive data security issues faced in the widespread application of wearable smart devices, become the maker of wearable smart device data security standards, and provide highly reliable network security service solutions for AI suppliers and users of wearable devices. (New Society)