In March 2026, the wave of artificial intelligence (AI) is once again reshaping the hardware supply chain. This time, the spotlight falls on a seemingly inconspicuous yet essential component—the printed circuit board (PCB), often described as the “electronic backbone” of modern devices.
Following fluctuations in upstream raw material prices, the industry has received a significant signal. Japanese semiconductor materials giant Resonac Holdings Corporation announced that, starting March 1, it would raise prices for copper-clad laminates (CCL) and bonding sheets by 30%. The move is widely seen as the first domino in a potential industry-wide price increase that could quickly ripple through sectors such as MLCCs, HDI boards, IC substrates, and high-frequency, high-speed PCBs.
At the same time, a new variable—often described as a “super catalyst”—is emerging: the upcoming launch of NVIDIA’s LPU (Language Processing Unit) inference chip. When rising material costs meet technological upgrades, the PCB industry is entering a period of both volume and price growth. The ripple effects of this transformation may even extend to the global copper market.
A 30% Price Increase: Cost Pressure and Strategic Adjustment
Resonac’s decision to raise prices by 30% is far from routine and reflects two key underlying dynamics.
Tight supply-demand balance:
Demand for high-layer-count, low-loss high-speed CCL used in AI servers is growing exponentially, while capacity expansion for high-end materials has lagged far behind. Supply shortages are therefore providing strong support for higher prices.
Inevitable cost pass-through:
Upstream raw materials—including copper foil, resin, and fiberglass cloth—remain expensive. Combined with rising logistics costs linked to geopolitical factors, material suppliers are increasingly compelled to adjust prices to maintain profit margins.
For downstream PCB manufacturers, the price hike presents both a cost challenge and a catalyst for industry consolidation. Leading companies with advanced production capabilities and the ability to pass costs on to major clients such as Microsoft and NVIDIA are likely to strengthen their market positions, while smaller players may struggle under mounting cost pressures.
NVIDIA’s LPU: A Catalyst for PCB Value Reassessment
If Resonac’s price hike provides the “push,” NVIDIA’s upcoming LPU inference chip represents a powerful “pull.”
As AI development shifts from the training phase to large-scale inference deployment, demand for specialized inference chips is surging. The LPU architecture is optimized for large language model inference workloads and imposes unprecedented requirements on PCB technology:
· Higher layer counts: PCB structures may evolve from the traditional 20–30 layers to 40 layers or more to accommodate complex signal transmission.
· Advanced materials: Ultra-low-loss high-speed materials are becoming standard, significantly increasing value per square meter.
· More complex manufacturing: Technologies such as HDI (High-Density Interconnect) and SLP (Substrate-Like PCB) are raising both technical difficulty and product value.
Market analysts note that the adoption of LPU chips could drive simultaneous growth in PCB production volumes and pricing. More importantly, PCBs are occupying a growing share of the value within AI server systems. Rather than being simple connectors, PCBs are increasingly viewed as critical components that determine the efficiency of AI computing performance. This shift in perception could significantly raise the valuation ceiling of the entire industry.
A Hidden Chain Reaction: Could PCB Growth Boost Copper Prices?
Within the PCB value chain, one often-overlooked commodity stands out—copper.
The core substrate of PCBs is copper-clad laminate (CCL), and copper foil accounts for a substantial portion of CCL production costs. AI-driven upgrades in PCB design are increasing copper demand not only in quantity but also in quality:
· Higher consumption: Multi-layer PCBs require more stacked copper foil layers. A high-end AI server may use several times—or even up to ten times—more copper than a conventional server.
· Higher specifications: High-frequency and high-speed signal transmission demands copper foil with higher purity and lower surface roughness, which commands higher processing premiums.
· Macro demand alignment: Investors such as Stanley Druckenmiller have already highlighted the long-term potential of hard assets like copper. Sustained expansion in the PCB sector could provide a significant structural boost to copper demand.
Although copper prices are still primarily influenced by global macroeconomic conditions and mining supply, AI-driven computing infrastructure is emerging as a powerful new demand driver. If PCB capacity expansion accelerates, tight supply in the copper foil segment could eventually support stronger upstream copper prices.
Investment Insights: Finding Alpha Amid Certainty
For investors seeking opportunities in this trend, several areas stand out:
· Focus on industry leaders: PCB manufacturers with advanced HDI and IC substrate capabilities—and established positions within NVIDIA’s supply chain—may benefit the most from pricing power and demand growth.
· Track technological adoption: The rollout of LPU chips and other next-generation AI architectures will be key indicators of industry momentum.
· Watch upstream suppliers: High-end copper foil producers and copper resource companies could also benefit from structural demand growth.
Conclusion
From a price increase announcement by a Japanese materials giant to the blueprint of next-generation AI chips, the PCB industry in 2026 is undergoing a profound transformation. This is not merely a cyclical rebound but a structural shift driven by the AI revolution.
As the “foundation” of computing power, PCBs are being redefined in importance. And embedded within every advanced circuit board lies copper—potentially positioning the metal for its own moment in the spotlight as the AI era accelerates.
Source：Changjiang Nonferrous Metals Network