The idea of being able to complete everyday tasks simply by wearing glasses is shifting from aspiration to reality.

In early 2025, dozens of new AI glasses were unveiled at the CES show in Las Vegas, igniting renewed excitement in the industry.

With an average of at least three new products released each month, over 20 AI glasses were released in the first half of 2025.

Although the penetration rate in this industry is less than 1%, it hasn't stopped major players from pouring money into it.

When it comes to smart glasses, many people think of Google Glass, launched in 2012. That product, despite massive investment, flopped miserably. But the situation is completely different now.

In the first half of 2025, Ray-Ban Meta smart glasses saw sales more than triple compared to the same period last year, a figure that offers hope for the entire industry.

Meta, a collaboration with Ray-Ban, has a starting price of $299 and weighs less than 50 grams. Since its launch in October 2023, over 2 million pairs have been sold.

Even more impressively, their partner, EssilorLuxottica, plans to increase annual production capacity to 10 million pairs by the end of 2026. Behind this ambition lies a strong optimism about the market.

Domestic manufacturers are also busy. On June 26th of this year, Xiaomi officially released its first AI glasses. Sales reached nearly 50,000 pairs in the first three days, setting a record for the fastest sales of AI glasses in China. This speed shocked the industry and made other players nervous.

In the first quarter of 2025, Thunderbird Innovations led the Chinese online AR/AI photography glasses market with a 50% market share, firmly securing first place.

As a leading domestic brand, Thunderbird launched two new products at CES in January. The AI photography glasses V3 weigh only 39 grams and are equipped with the first-generation Qualcomm Snapdragon AR1 platform and a dedicated NPU. The average AI response speed is increased to 1.3 seconds, with a recognition accuracy rate of up to 98%.

Behind these figures lies a larger story. In the first quarter of 2025, global sales of AI smart glasses reached 600,000 units, a year-on-year increase of 216%. In the Chinese market, 71,000 units of AI smart glasses were sold across all channels, a year-on-year surge of 193%. With such astonishing growth, it's no wonder that major players are jumping into the market.

Even Huawei can't hold back. It will hold a new product launch on May 19th, and among the upcoming releases will be smart glasses. Alibaba has partnered closely with Thunderbird to launch AI glasses powered by the Tongyi big model. Baidu and ByteDance are also developing their own products.

The changes in Shenzhen's Huaqiangbei district are even more telling. AI glasses have become the fourth most popular product in the area, after headphones, watches, and drones.

Approximately 80% of watch stalls now feature AI glasses, and there are even two or three dedicated glasses stalls. By 2025, monthly sales of AI glasses in Huaqiangbei will exceed 40,000 units, with nearly 40% exported, and global market share exceeding 10%.

The core driver of this boom is the maturity of AI big models. Previously, smart glasses could only perform simple functions, but now, when integrated with big models, they can perform complex tasks such as real-time translation, intelligent object recognition, and voice interaction.

Meta's products allow users to activate an AI assistant by simply saying "Hey Meta," while Xiaomi Glasses use Xiao Ai for instant responses.

A mature supply chain is also a key factor. Manufacturers such as Goertek, Luxshare, Sunny Optical, AAC Technologies, Jiahe, and Emdoor all offer mature OEM solutions comparable to Ray-Ban Meta, virtually eliminating manufacturing barriers.

This has led to a rapid decline in product prices, with mainstream models now priced between $300 and $400, and domestic products even reaching prices between 1,500 and 2,000 yuan.

However, the industry's boom also presents challenges. Global sales of smart glasses are expected to reach 2.983 million units in 2024, and by 2025, this number is expected to more than quadruple.

However, global smartglasses shipments are projected to exceed 14.518 million units in 2025, with shipments in China projected to reach 2.907 million units. This volume pales in comparison to smartphone shipments, which often exceed hundreds of millions.

Product homogeneity is another issue. Currently, AI glasses on the market offer similar functionality, focusing primarily on scenarios like question-and-answering, translation, and object recognition. Companies compete primarily on basic metrics like large-scale model response speed, battery life, and image quality. A true killer app has yet to emerge.

Battery life is also a significant drawback. Existing products typically have a battery life of four to six hours, far less than a smartphone. Coupled with chip computing power limitations, many complex tasks still rely on the phone or the cloud for processing, lacking independence.

Despite this, investors and tech giants remain bullish on this sector. The underlying logic is simple: AI requires a platform, and glasses, being closest to the human eye and unlike smartphones, requiring no physical manipulation, offer an ideal entry point for interaction.

Meta founder Mark Zuckerberg believes that AI glasses will be a significant future trend, and predicts that AI glasses priced around $300 will become a hit.

Looking at the application ecosystem, there will be only 5,000 to 8,000 smart glasses apps globally in 2025, far fewer than mobile phone apps. Third-party predictions suggest the number will exceed 50,000 within the next three years. This presents a wealth of entrepreneurial opportunities waiting to be tapped, and those who can first establish an ecosystem advantage will likely seize the initiative.

AI glasses are poised for a breakout moment. While penetration remains at less than 1%, the growth rate speaks for itself. The industry giants aren't blindly following trends, but are betting big on the next generation of computing platforms.

This battle of the hundreds of glasses has just begun. The ultimate winner will depend on who can truly address user pain points and create irreplaceable use cases.