Beim ersten Mal, dass man Hardware herstellt, sollte Alibaba erst einmal vorsichtig vorgehen.

A new player has entered the "Battle of a Hundred Smart Glasses."

On July 26th, at the 2025 World Artificial Intelligence Conference, Alibaba announced that it would launch its first self-developed AI glasses - Quark AI Glasses.

Screenshot from Weibo @Quark

In Alibaba's AI narrative, the launch of AI glasses is incorporated into its To C overall strategy. At the beginning of this year, Alibaba merged its AI application "Tongyi" into the Alibaba Smart Information Business Group and announced the integration of the Tmall Genie team and the Quark product team to explore new hardware directions such as AI glasses. Wu Jia, the CEO of the Alibaba Smart Information Business Group, is in charge of Alibaba's AI To C applications.

Just as Alibaba high - profilely enters the field, a price war has already begun. Xiaomi has lowered the price of its AI glasses to 1,999 yuan.

So, facing the "price butcher" Xiaomi in the front and the painful lesson of the failed Alibaba Cloud OS mobile phone in the back, is Alibaba's "hardcore" move this time a desperate gamble or a well - thought - out safe bet? It remains to be tested by the market.

Finally in the game

Alibaba, which has been making waves in the digital world, is finally going to enter the hardware market.

Looking back at Alibaba's development history, the last time it got so close to hardware was the ambitious Alibaba Cloud OS mobile phone project. At that time, Alibaba tried to use the operating system as a weapon to seize an entry point in the highly competitive mobile Internet market and cooperate with mobile phone manufacturers.

However, at that time, it only developed the mobile phone system and did not involve mobile phone hardware, but it was still the closest Alibaba had ever come to hardware.

With that ambition, Alibaba planned to achieve a shipment volume of tens of millions in 2013. It poured resources into the project, including the Taobao channel, support from one billion developers, and operators offering zero - cost mobile phones. The momentum was huge. But no one expected that this seemingly perfect layout would be suddenly stopped by Google.

In September 2012, Acer's mobile phone equipped with Alibaba Cloud OS was forced to be stopped by a ban from the Android giant before it could even be launched. That failure made Alibaba deeply understand how fatal the passivity and vulnerability caused by the lack of core control in the hardware ecosystem could be.

After that, Alibaba never had the intention to directly enter the hardware - related layout again. It only made corresponding hardware layouts as an investor, mainly focusing on hardcore technology fields such as chips and intelligent robots. Except for investing in Meitu in May this year, there have been few layouts in the C - end field.

After many years, Alibaba has stepped into the hardware territory again, and this time it is directly self - developing AI glasses. This is obviously not a simple "rekindling of an old dream."

This time, Alibaba is not just providing an operating system. It is comprehensively involved in R & D and production, from chip design, dual - system matching to function integration. According to Alibaba, the Quark AI Glasses adopt a dual - core parallel architecture of Qualcomm's AR1 flagship design chip and a low - power co - processor, supplemented by the dynamic resource scheduling of the Android and RTOS dual systems.

Actually, Alibaba has long made corresponding layouts for AI hardware. By combining the C - end shopping scenario with AI technology, Alibaba was the first to embrace Apple. Previously, the Taobao Vision Pro version was launched on the Apple App Store, becoming the first Chinese e - commerce shopping application on the Vision Pro.

However, judging from the sales situation and user experience after its launch, the Apple Vision Pro has not become a hit in the market. Guo Mingji, an analyst of Apple products at Tianfeng International, previously predicted that the shipment volume of the Apple Vision Pro in 2024 was expected to be about 500,000 units. In his view, Apple needs to solve the problems faced by the Vision Pro, such as the lack of key applications, high price, and poor comfort, without sacrificing user experience.

Having learned from the failure of the Alibaba Cloud mobile phone, instead of leaving its fate in the hands of others again, Alibaba might as well enter the market itself to shape the future.

Limited application space

Currently, the AI industry, especially in the C - end application field, is generally facing commercialization difficulties.

Even the once - famous Manus has now abandoned the domestic market, laid off employees, and moved overseas. In March this year, Manus became the darling of the tech circle overnight with a demonstration video of "an intelligent agent autonomously completing tasks." Within a week of its release, two million users rushed to make reservations, and the internal test code was even sold at a sky - high price of 100,000 yuan.

However, this frenzy only lasted for 130 days. When the first - batch users actually experienced the product, their beautiful expectations were shattered. Eventually, it quickly cooled down due to problems such as premature commercialization with immature technology, high costs, and lack of a clear product positioning.

In the context of the lack of a clear profit path and users' payment habits, the survival space for pure C - end AI applications is indeed relatively limited.

Wu Yongming, the CEO of Alibaba Group, hit the nail on the head: "The greatest imagination of AI is not to create one or two new super apps on the mobile phone screen, but to take over the digital world and change the physical world."

Looking around, the intelligent hardware field shows a broader commercialization imagination space. Compared with pure software applications, intelligent hardware can provide a more intuitive, immersive, and convenient user experience. And initial commercialization can be achieved through hardware sales, and the profit model can be expanded through content subscriptions, value - added services, etc. later.

Meanwhile, the strategic value of the "portable entry point" is self - evident: As a head - worn device, AI glasses can capture more than 80% of the human body's sensory input and have a strong ability to penetrate different scenarios, covering various scenarios such as office, life, and mobile use. It is expected to become the most important personal mobile entry point after the smartphone.

According to relevant data from IDC, the global shipment volume of smart glasses in the first quarter of 2025 was 1.487 million units, a year - on - year increase of 82.3%. It is predicted that the shipment volume of smart glasses in China in 2025 is expected to reach 2.907 million units, a year - on - year increase of 121.1%.

Image from IDC

More realistically, the AI learning machine has already proven the viability of the AI education application scenario. Although there are still suspicions of it being a "IQ tax" in public opinion, it represents AI hardware and has successfully penetrated the C - end market ahead of the application side. According to public data, the penetration rate of AI learning machines is expected to exceed 40% in 2025, becoming an important carrier for AI + education.

Playing it safe

Compared with focusing solely on C - end applications, AI glasses are indeed a safe bet.

Since the end of last year, Alibaba has carried out overall adjustments and accelerated the integration of its AI To C business. The AI application "Tongyi," originally belonging to Alibaba Cloud, was split and merged into the Alibaba Smart Information Business Group where Quark is located.

At the same time, the Tmall Genie hardware team and the Quark product team started to integrate their work to jointly explore new hardware directions such as AI glasses. At the beginning of this year, Quark received more attention and resource tilt within the group. The appearance of Alibaba founder Jack Ma in the Quark office area and the large - scale recruitment of hundreds of positions for AI technology and product R & D all confirm Alibaba's extraordinary emphasis on the AI To C business.

At the organizational structure level, Wu Jia, a "post - 85s" executive, was entrusted with an important task. By overseeing the Smart Information Business Group and the Smart Interconnection Business Group, he has become the actual operator of Alibaba's AI To C strategy, further ensuring the resource investment and strategic implementation of the AI glasses project.

With such a large - scale resource tilt, it is particularly important to take the first step "steadily, accurately, and forcefully."

After years of market cultivation, the AI glasses market has reached the critical point of explosion. Xiaomi, which is always good at seizing opportunities, has just entered the market. On June 26th, Lei Jun, the founder of Xiaomi, officially launched the first AI glasses at a new product launch event. Goldman Sachs predicts that Xiaomi's AI glasses are expected to occupy 10% of the Chinese smart glasses market share in 2025.

According to the forecast of TrendForce, the global market scale of AI smart glasses will reach 5.7 million units in 2025, a year - on - year increase of 110%, and is expected to exceed 13.6 million units by 2030. In 2025, the AI glasses market has shifted from the technology verification stage to the commercial stratification stage. The consumer - grade and high - end markets show different characteristics, and the price - performance battle is dominated by Chinese manufacturers.

Although there are already heavyweight players like Xiaomi in the industry, Alibaba still has differential advantages.

In Alibaba's view, the biggest selling point of its AI glasses lies in the high - level integration of Alibaba's ecosystem. It will be equipped with the Tongyi Qianwen large - scale model and Quark's latest AI capabilities and will deeply support high - frequency Alibaba - affiliated applications such as Gaode Navigation, Alipay's "Take a Look" payment, Taobao price comparison, and Fliggy travel reminders. The closed - loop ecosystem of "hardware + services" can provide users with a "one - stop" convenient experience, forming a unique differential competitiveness.

Screenshot from the promotional long - picture of Quark AI Glasses

Another market variable worth noting is that the founders of several AI glasses companies have "Alibaba - related" backgrounds. In addition to rational factors such as financial data and business models, the human factor should also be fully considered. These human resources scattered in various links of the industrial chain have, to a certain extent, reduced Alibaba's cost of trial - and - error.

Zhu Mingming, the founder of Rokid, known as the "Seventh Dragon of Hangzhou," once worked at Alibaba, where he was responsible for deep learning, smart devices, and other businesses in the M studio and accumulated rich experience in the system aspect. Ru Yi, the founder and CEO of Liweike Technology, another AI glasses company born in Hangzhou, was once the general manager of the Tmall Genie product. From her resume, she has served as the general manager of the hardware terminal of Alibaba's Artificial Intelligence Laboratory, the general manager of the Tmall Genie product, a partner, and the vice - president of Xiaomi TV.

As a giant entering the market, Alibaba has stronger trial - and - error capabilities and resource investment. It may have the opportunity to integrate these resources with Alibaba - related backgrounds to compete with heavyweight players like Xiaomi.

In short, with Alibaba's entry, the current "Battle of a Hundred Smart Glasses" is becoming more and more interesting.

This article is from the WeChat official account "Source Byte." Author: Source Byte. Republished by 36Kr with permission.