High-end materials pose a time bomb for domestically produced AI glasses: In-depth analysis of Apple's "lock-in" of top-grade fiberglass cloth production capacity
In mid-month, several major global technology and financial media outlets reported that Apple had recently sent several senior engineers to Japan to attempt to secure Nittobo's high-end NE-grade fiberglass cloth production capacity for the next two years through government coordination, aiming to ensure the smooth mass production of its core products.
However, Nittobo, the main player in this situation and the monopolist of NE-grade fiberglass cloth (Low Dk/Low Df), has not been proactive. President Hiroyuki Tada repeatedly emphasized in front of the camera that there would be no expansion of production, nor would the interests of other small and medium-sized customers be sacrificed.
The reasons given sound noble, but the truth is: the drawing and stable yield of high-end fiberglass cloth requires long-term investment, and the delivery cycle of core equipment is extremely long. Expanding production rashly would carry immeasurable risks. It's worth noting that other giants, including Google, Nvidia, and Meta, also arrived in Japan earlier than Apple. Everyone is doing the same thing: scrambling for cloth.
In our view, Apple's move isn't new: in the rapidly developing AI era, upstream materials are scarce, and giants "preemptively positioning" themselves to compete is standard practice (the recent surge in memory prices is a prime example). However, when placed alongside keywords like "AI glasses," "visual intelligence," and "next-generation computing platforms," it takes on a deeper meaning.
According to industry insiders, even if Nittobo has new production capacity available in the future, it won't be released until the second half of 2027 at the earliest. This means that in 2026, the global technology sector, especially in AI, will face a "zero-sum game" regarding top-grade fiberglass cloth. Whoever acts first will secure the largest market share.
Against this backdrop, we share the same questions: Is this micrometer-thick "cloth" related to the rapidly developing AI smart glasses? Will domestic manufacturers be "held back" again? If not, are we prematurely projecting our anxieties from the mobile phone chip era onto the still-developing smart wearable category?
The conclusion in advance: Yes, it is related, and the consequences are severe.
A "Future War" Triggered by a Piece of Fabric
First, a bit of background: Fiberglass cloth is important not because it's "mysterious," but because it's so fundamental and crucial.
Specifically, it's a core reinforcing material in high-performance PCBs and advanced packaging substrate systems, determining a series of fundamental indicators such as dielectric loss, thermal expansion stability, and dimensional consistency. Over the past two years, AI technology has swept the world like a whirlwind, causing a surge in demand for servers, 5G communications, and advanced packaging. Basic materials have gradually transformed from optional to essential, making upstream supply shortages inevitable.
It's important to emphasize that this logic doesn't yet hold true for the current consumer-grade AI smart glasses market. Currently, most familiar products come from mature low-to-mid-range supply chains, and the quality of the product experience depends more on power consumption, heat dissipation, weight, and wearing comfort than on a "high-end electronic cloth."
What we're truly focused on is the future—a long-term perspective, a prediction aligned with the logic of industry giants. The battle between Apple and Nittobo has once again reminded us that when a product category is identified as the "next-generation terminal" (such as smart glasses), the supply of key materials becomes a matter of life and death. At this point, securing a foothold in advance and cutting off supply chain exits becomes one of the core factors determining success or failure.
The upward path for AI glasses may be "locked up"
The so-called "cutting off" or "securing a foothold" doesn't mean that AI smart glasses can't be made simply because top-grade fiberglass cloth isn't currently available. Rather, it means that when manufacturers push this product category to a certain engineering critical point, and high-specification PCBs appear inside the glasses (whether in the frame or temples), this cloth or other key materials are very likely to come to the forefront, becoming the foundation of high-end design.
Specifically, this tipping point comprises three key elements:
First, glasses are becoming increasingly independent, with a higher proportion of local AI inference, shifting from occasional low-level calls to continuous operation.
Second, high-speed connectivity for glasses is becoming the norm, including Wi-Fi 8, Bluetooth 6.X, and interconnection with other devices.
Third, the size of the glasses is severely limited, with the size of the temples and the temperature at contact with the face reaching their limits.
These three factors combined place immense pressure on PCB design, including dielectric loss, board layer stacking space constraints, and heat accumulation. Do these demanding conditions sound familiar? That's right, this is where high-end electronic fiberglass cloth comes in. At this point, high-quality electronic fiberglass cloth not only improves product performance but also allows for continued research and development.
To explain: Under the extreme design conditions of smart glasses, top-grade fiberglass cloth with low Dk/low loss (specifically referring to products from Dongfang) truly shines—it doesn't increase chip frequency, but rather ensures stability under high-frequency, high-speed, high-density, miniaturized, and easily heated conditions.
Conversely, if material loss is higher and stability is weaker, the engineering team can only mitigate the risk through more conservative wiring, larger spacing, thicker layers, and wider structural components. This compensation is the most extravagant cost in the form of eyeglasses: it directly squeezes temple thickness, thermal design space, and battery life margin.
Therefore, if future leading manufacturers push AI glasses towards higher integration, higher speed, and greater miniaturization, while simultaneously securing high-quality supplies of key materials through large-scale procurement, then even if small and medium-sized brands work hard on product innovation, they may be forced to choose conservative engineering architectures, unable to compete with large manufacturers who have a "stockpile" of high-end materials. Their products will become bulky and cumbersome, ultimately losing the ability to compete in the high-end market.
This is not a competition of "who is cheaper," but a competition of "who can sustain the product line."
Two Breakthrough Paths for Domestic AI Glasses
If we acknowledge that key materials may become a barrier to entry, then the response from domestic manufacturers and the media should not remain at the level of "emotional confrontation." There are two more realistic paths: one is upstream collaboration in materials and processes, and the other is inward collaboration in ecosystems and application scenarios.
Let's look at the first path. We believe manufacturers shouldn't wait until smart glasses reach (or are foreseeably about to reach) the limits of design before scrambling to buy "fabric" from Japan. Instead, they should proactively form joint R&D and verification matrices with upstream domestic material suppliers, such as copper-clad laminate (CCL), optical supply chains (e.g., silicon carbide substrates), and even PCB manufacturers, to ensure the usable material systems are tested and validated in advance.
In the high-end sector, domestic electronic fabric manufacturers have actually been making progress in recent years, especially in areas like ultra-fine yarn drawing, stable yield rates, and large-scale production lines. Once key indicators are successfully engineered and validated, both domestic and international manufacturers can secure a more proactive supply security during the production capacity release window at Tobo in Japan around 2027.
Now let's look at the second path. Even if the hardware requirements for smart glasses continue to rise uncontrollably in the future, the key to success for domestic consumer products shouldn't be the same as for smartphones—just piling on components and then running benchmarks. In our view, the biggest advantage of domestic eyewear manufacturers is clear: a mature software ecosystem and a localized user experience.
Specifically, given the nation's high emphasis on data security, they need to quickly integrate deeply with national-level applications, adapt their products to the daily lives of Chinese consumers through faster product iterations, and build a strong competitive advantage through pricing and distribution channels. These seemingly "low-level" strategies can allow eyewear manufacturers to continuously accumulate user experience and cultivate user loyalty during the phase where hardware development is slow (or even hopeless), buying them time and establishing an advantage before the crucial battle.
Final Conclusion
Fortunately, returning to reality, today's AI smart glasses are not yet "held back" by high-end NE-grade fiberglass cloth.
However, if AI glasses, as leading manufacturers claim, can reach the standard of integrating into the lives of end-users within the next two years, this "cloth" may change from an option to a necessity, or even a barrier to entry.
Apple's "lock-in" behavior today is insidious and underhanded, but it secures a certain future for itself. Behind this may lie a research and development progress that is difficult for outsiders to estimate or judge.
It's 2026 in the blink of an eye. After several years of fierce competition, the industry should shift its focus from "launch marketing," "celebrity endorsements," and "grand narratives" to a more fundamental level, emphasizing "materials," "processes," "systems engineering," and "software ecosystem." We must remain humble and focused on solid development.
High-end fiberglass cloth is merely a surface, an entry point. It points to a set of extremely difficult-to-break-through strategies, but essential for all manufacturers: combining real market demand with effective, down-to-earth R&D, and establishing a stable supply chain as the foundation for product iteration.
Whoever can establish a verifiable product roadmap in advance has the opportunity to build a robust materials system early, avoiding forced downgrading at engineering inflection points; whoever can establish a home-field advantage in the ecosystem can offset the slow pace of hardware and materials development with the speed of software and scenario iteration.