High-end electronic fabrics are facing a shortage predicament?

Unexpectedly, in early 2026, the global technology industry faced a supply chain crisis triggered by a single material: high-end electronic-grade fiberglass cloth. Driven by the explosive growth of AI servers, high-end chips, and advanced packaging technologies, this seemingly niche yet crucial basic material was experiencing an unprecedented supply-demand imbalance and shortage. Nittobo, a Japanese company, held a dominant market share in high-end semiconductor packaging substrate electronic cloth due to its leading technology in low-CTE electronic cloth, becoming a key bottleneck for top AI chips and consumer electronics products from companies like Nvidia and Apple.

The Shortage of Electronic Cloth: The AI ​​Boom Ignites Material Bottlenecks

Electronic cloth is one of the core materials for manufacturing printed circuit boards (PCBs) and chip packaging substrates. In advanced packaging (such as Fan-Out and 2.5D/3D packaging), the thermal expansion coefficients of the chip and the substrate must be highly matched; otherwise, warping, cracking, or even failure due to temperature changes is highly likely. Nittobo's ultra-low CTE fiberglass cloth, with its nearly identical thermal expansion characteristics to silicon chips, has become an indispensable raw material for high-end chip substrates.

With the accelerated deployment of global AI computing infrastructure, demand for H100/B100 level AI chips from giants like Nvidia, AMD, Google, and Meta has surged, driving a surge in orders for high-end ABF (Ajinomoto Build-up Film) and BT substrates. BT substrates-widely used in mobile SoCs, RF modules, and some AI accelerator chips-are highly dependent on Nittobo's fiberglass cloth supply. According to recent reports, Apple has deployed engineers to Mitsubishi Gas Chemical (MGC) to deeply engage in its BT substrate production process, solely to ensure material security for new products in 2026, including the first foldable iPhone. Qualcomm has also urgently contacted the small Japanese manufacturer Obayashi Chemical to try and open a second supplier, but due to limitations in its production capacity and yield rate, a substantial breakthrough is unlikely in the short term.

Although Nittobo has announced capacity expansion, the new capacity is not expected to be released until 2027. This means that 2026 will be the year with the most severe shortage of high-end electronic fabrics, potentially directly constraining global AI chip shipments, high-end smartphone iterations, and even the pace of data center construction.

China boasts a massive fiberglass production capacity, but a gap remains in high-end products.

China is the world's largest fiberglass producer, with leading companies such as China Jushi, Sinoma Science & Technology, Chongqing International Composite Materials (CPIC), and Honghe Technology, ranking first globally in electronic cloth production capacity. However, in the high-end electronic cloth sector, especially in ultra-low CTE, ultra-thin (≤30μm), and high flatness electronic cloths used for advanced packaging, a gap still exists compared to foreign countries.

Currently, mainstream domestic electronic cloths are mostly used in mid-to-low-end applications such as consumer electronics motherboards and communication equipment PCBs, with CTE values ​​generally between 4–6 ppm/℃. Nittobo's high-end products, however, can control CTE values ​​below 2.5 ppm/℃ and possess superior dielectric properties and dimensional stability. While Honghe Technology has achieved partial localization of high-end electronic cloths and entered the supply chain of major Taiwanese copper-clad laminate (CCL) manufacturers, it has not yet achieved large-scale substitution capabilities in the field of fiberglass cloths for BT substrates. CCL manufacturers such as Nanya New Materials and Shengyi Technology are also actively validating domestically produced electronic cloths, but the certification cycle is long and yield fluctuates greatly, making it difficult to shake Nittobo's dominant position in the short term.

For China, 2026 is both a year of challenges and a window of opportunity. The production vacuum period at Nittobo provides a valuable window for the introduction of domestically produced high-end electronic fabrics. There are already signs that domestic substitution is accelerating. In the past year, fiberglass manufacturers such as Sinoma Science & Technology, International Composites, and Honghe Technology have announced increased investment in high-end electronic fabrics and expanded their production capacity, accelerating the breakthrough of domestic products!

In the wave of AI reshaping the global technology landscape, the real competition has long since moved beyond algorithms and chips, extending to the most fundamental material foundations. A seemingly ordinary piece of fiberglass fabric has now become the "Achilles' heel" restricting the development of the trillion-dollar AI industry. For China, this is both a severe challenge and a historic opportunity to promote the independent control of basic materials. Only by bridging the last mile "from fiberglass plant to wafer fab" can China truly seize the initiative in this silent battle for AI computing power.