Recently, many investors have reported that the semiconductor sector seems lively, but they are chasing GPU and optical module operations but have not made any money, and they are often trapped in high positions.
Most retail investors focus on popular tracks such as chip design and wafer manufacturing, but ignore the undervalued segment of advanced semiconductor packaging.
Industry professionals in the industry have reached a consensus that the two-year packaging and testing cycle of continuous adjustments has officially bottomed out, multiple benefits have been released, and the industry has reached an inflection point.
Electronic ceramic materials were originally ordinary, but now, under the triple logic of AI computing power, domestic substitution, and shortage of overseas supply, they have ushered in a golden development period with both volume and price rising.
Many people have misunderstandings about semiconductor packaging. They believe that packaging is just simple packaging, and ceramic substrates are ordinary consumables with no technical threshold.
However, with the iteration of large models, the power consumption of high-end computing chips has increased significantly, and the performance of traditional packaging materials cannot keep up. High-performance electronic ceramics have become the core materials in demand for AI servers, high-speed optical modules, and new energy vehicle power chips.
Let’s analyze in detail the core support and industry chain opportunities for this round of industry recovery.
1. Triple hard-core benefits have been implemented simultaneously, and semiconductor packaging and testing has completely emerged from the downward cycle
The turning point of this round of packaging and testing industry is the deterministic upward trend brought about by the improvement of the three major fundamentals of inventory cycle, industrial demand, and policy support.
1. The global chip inventory has been depleted and the industry has entered a replenishment cycle.
The semiconductor sector has weakened in the past two years due to weak end demand, high inventories of chip manufacturers, and upstream reductions in production capacity and price cuts.
In the second quarter of 2026, the end market has changed significantly. Overseas cloud manufacturers have increased their efforts in building AI computing power, projects such as domestic intelligent computing centers have been launched, and purchase orders for computing power chips have increased. The demand for consumer electronics and automotive semiconductors has recovered, and the price of memory chips has increased, driving the recovery of the industrial chain.
The inventory turnover days of global chip manufacturers have fallen, the order scheduling cycle of leading packaging and testing companies has lengthened, and downstream companies have taken the initiative to replenish their inventories, which has boosted the capacity utilization and profitability of packaging and testing companies.
2. Advanced packaging has become the mainstream development route of the industry, and domestic production lines have expanded on a large scale.
The dividend of chip manufacturing process shrinkage is close to the ceiling, and advanced packaging technology has become the core path to break through performance bottlenecks.
Domestic packaging and testing leaders continue to increase the construction of advanced packaging production lines. In 2026, multiple high-end packaging production lines will be put into production, and local advanced packaging production capacity will be doubled.
Advanced packaging has high requirements for supporting base materials, opening up the incremental demand space for high-end ceramic materials and achieving “both volume and price increases.”
3. Top-level policies continue to increase the emphasis on semiconductor materials, and domestic substitution is accelerated.
Semiconductor basic materials are the core direction of independent control of the industry. The policy includes high-end ceramic substrates and other products as encouraged development categories, and local governments provide subsidies and tax incentives.
For a long time, the high-end electronic ceramics market has been monopolized by overseas companies, and the localization rate is low.
With policy support, domestic enterprises have accelerated their research and development and customer verification, and import substitution has brought long-term stable growth.
2. Why will the humble ceramic material become the core line of this round of closed beta market?
The core reasons are the performance advantages brought by hardware upgrades in the AI computing power era, as well as the tight supply in overseas supply chains and the reversal of the supply and demand pattern.
1. The performance surpasses traditional packaging materials, and there is no alternative option for high-end chips.
There are three main types of packaging substrates: resin organic substrates, metal substrates, and ceramic substrates, with large performance gaps.
Ordinary resin substrates have poor thermal conductivity and high temperature resistance and are not suitable for high-end computing hardware; metal substrates have strong electrical conductivity and poor insulation effect, and cannot be adapted to high-speed optical modules; alumina and aluminum nitride ceramic substrates have the advantages of high thermal conductivity, strong insulation, high temperature resistance, and matching thermal expansion coefficient with silicon chips, and are indispensable materials for high-end AI servers and the like.
2. The supply of overseas leading players is limited, and the gap in the global supply chain continues to expand.
In the second quarter of this year, the global electronic ceramics supply chain was tense.
Affected by raw material export controls and slow expansion of production capacity, suppliers such as Japan’s Kyocera and Tosoh are unable to meet order demand.
The global production capacity gap for high-end aluminum nitride ceramic substrates is nearly 30%. The delivery cycle of overseas manufacturers has lengthened. Downstream companies have increased the proportion of domestic ceramic manufacturers purchasing, and local companies’ orders and quotations have both increased.
3. The three major downstream tracks are expanding simultaneously, opening up the long-term growth ceiling.
The demand for ceramic materials comes from the three high-prosperity industries of artificial intelligence, new energy vehicles, and commercial aerospace.
In the AI computing power center track, the growth in demand for AI servers has driven the demand for ceramic substrates and other consumables; in the new energy vehicle track, the demand for automotive electronic ceramics has grown steadily; in the satellite and communications track, there is a huge demand for ceramic tubes and casings, which is a long-term stable incremental market.
3. The entire semiconductor ceramic industry chain is dismantled, and the core logic of the upstream, midstream and downstream is clear
The electronic ceramic packaging industry chain is divided into three layers: upstream ceramic powder raw materials, midstream ceramic substrate/ceramic tube and casing manufacturing, and downstream semiconductor packaging and testing applications.
(1) Upstream: high-purity ceramic powder, the core technical barrier of the entire industry chain.
High-end aluminum nitride and high-purity alumina powder have been monopolized by Japanese companies for a long time, and it is difficult for domestic companies to break through.
Enterprises that realize self-development and self-production of powders can control costs, have strong independent controllable logic in the industrial chain, cover multiple scenarios in business, and drive performance through resonance in multiple tracks.
(2) Midstream: Ceramic substrates and ceramic tubes and shells are the links with the greatest profit elasticity in the industry.
The midstream segment is the core track with the highest added value and the strongest performance flexibility, and is divided into two major products: high-speed optical module ceramic shells and computing power chip aluminum nitride ceramic substrates.
The high-speed optical module ceramic shell is the core supporting component of the optical module. The domestic leading product has a high yield rate, a large market share, and strong order certainty; the aluminum nitride ceramic substrate of the computing power chip has a high process difficulty, a long customer certification cycle, and stable supply after entering the supply chain.
(3) Downstream: Semiconductor packaging and testing manufacturers, the demand is directly fulfilled.
Downstream packaging and testing companies are direct purchasers of ceramic materials, and the expansion of advanced packaging production capacity has driven an increase in the purchase of ceramic consumables.
Domestic leading packaging and testing companies have introduced domestic ceramic substrates to reduce costs, avoid risks, and drive orders from local ceramic companies.
4. Sort out the core targets of the industrial chain (only an objective introduction to the company’s business, and does not constitute investment advice)
Combined with factors such as industrial chain stratification, sort out local companies with core competitiveness in each link, such as China Ceramics Electronics, China Ceramics Materials, Xuguang Electronics, Yishitong, Wuhan Fangu, etc.
5. Ordinary investors must avoid these 4 common misunderstandings when planning this track
Misunderstanding 1: Confusing low-end ceramic processing companies with high-end computing power ceramics.
Priority will be given to screening targets to see whether they have production capacity for aluminum nitride products and whether they have entered the leading supply chain.
Misunderstanding 2: Chasing high prices in the short term and ignoring the customer certification cycle.
The customer certification cycle for high-end ceramic substrates is long, and companies have a time lag from research and development to supply, so do not simply chase the price increase.
Misunderstanding 3: Ignoring the cost pressure caused by fluctuations in overseas raw materials.
Give priority to companies that are vertically integrated and produce their own powders, as they have stronger cost stability.
Misunderstanding 4: Betting solely on short-term themes and ignoring the medium- and long-term rhythm of the industry.
The market for ceramic materials is a mid- to long-term boom track, which is suitable for adjustments to the layout in batches and to avoid frequent short-term operations.
6. Objective risks in the industry, and a rational view of track opportunities
The industry has potential risks such as AI computing power construction progress being slower than expected, domestic customer verification progress delays, overseas companies releasing production capacity, and overall market sentiment fluctuations. Investors need to make rational judgments.
Interactive topics:
1. When you usually lay out the semiconductor track, do you prefer front-end chip design, or low-level subdivision tracks such as packaging and testing and materials?
2. In the ceramic packaging industry chain, which link in the upstream powder or midstream substrate do you think has greater performance flexibility?
3. How long do you think the current boom cycle of the semiconductor packaging and testing industry can last?
Welcome to leave your opinions in the comment area and discuss together.
