In a rare, high-profile televised appeal, two of China’s most influential semiconductor figures have called on domestic chipmakers to bridge the gap between prototype and high-volume manufacturing. During the May 17 broadcast of the CCTV program Dialogue, Semiconductor Manufacturing International Corporation (SMIC) founder Richard Chang and Advanced Micro-Fabrication Equipment Inc. (AMEC) chairman and CEO Dr. Gerald Yin urged fabs to accelerate the integration of locally produced equipment into active production lines.
This coordinated push comes as China’s semiconductor equipment sector reports record-breaking financial growth for 2025, even as firms grapple with eroding profit margins. As a former engineer, I see this as a pivotal moment: the industry is attempting to transition from a reliance on foreign tools—historically supplied by companies like Applied Materials and Lam Research—to a self-sustaining domestic ecosystem. However, the move is complicated by intense internal price wars and the persistent, high-stakes challenge of achieving yields that can compete with global leaders.
The Paradox of Growth and Margin Compression
The numbers from 2025 illustrate a sector experiencing a massive, state-backed expansion. AMEC reported full-year revenue of $1.74 billion USD, representing a 36.6% year-on-year increase, while Naura Technology generated $3.91 billion in revenue across the first three quarters alone. Similarly, Piotech and ACM Research saw significant revenue gains, reflecting the rapid pace of domestic fab construction.
Yet, these top-line figures mask a sobering reality for shareholders: gross margins are trending downward. AMEC’s gross margin fell to 39.2% for the year, with a notable 5.8 percentage point dip in the third quarter. ACM Research experienced a similar slide, dropping from 50.1% in 2024 to 44.4% in 2025. According to industry analysis, this margin erosion is primarily driven by fierce competition among domestic vendors as they battle for market share in a landscape increasingly defined by restricted access to international technology.
The following table outlines the reported revenue growth for key players in the Chinese semiconductor equipment space during 2025:
| Company | Reported Revenue | Growth Context |
|---|---|---|
| AMEC | $1.74 Billion | Full-year 2025 |
| Naura Technology | $3.91 Billion | First three quarters 2025 |
| Piotech | $617 Million | First nine months 2025 |
| ACM Research | $901.3 Million | Full-year 2025 |
The Qualification Hurdle
During the CCTV broadcast, both Richard Chang and Dr. Gerald Yin emphasized that the primary obstacle to further localization is not just technological capability, but the willingness of fabs to endure the “tuning” phase required for new equipment. In the semiconductor industry, qualifying a new tool on a leading-edge production line is a rigorous, 18-to-24-month process. It requires testing for particle contamination, process drift, and throughput stability under continuous, high-intensity operation.
Chang, whose foundational work in Chinese foundry operations gives him unique standing, argued that domestic equipment cannot reach the necessary maturity without exposure to real-world production runs. He suggested that fabs begin by testing these tools with small batches of 100 wafers to mitigate the risks associated with early-stage adoption. Yin echoed this sentiment, noting that even world-leading equipment providers typically require years of collaborative tuning when deploying new systems into a high-volume facility.
While AMEC has claimed that its etch technology is utilized in the supply chains of major foundries, including potential roles in 5nm and 3nm production nodes, these claims remain difficult to verify independently. SMIC itself has faced documented yield losses throughout 2025, which reports suggest were linked to maintenance and validation stalls—the very challenges Chang highlighted as a reason to prioritize domestic, serviciable equipment.
The Lithography Bottleneck
Despite the optimism surrounding etch and deposition tools, the most critical chokepoint remains lithography. Current data indicates that domestic lithography localization remains below 5%. While Shanghai Micro Electronics Equipment (SMEE) continues to produce 90nm-class ArF systems, a credible domestic solution for sub-10nm lithography remains a long-term goal, likely not arriving before 2030.
Industry observers are currently monitoring the “Mount Everest” project, a domestic immersion DUV scanner being tested at SMIC. The tool, reportedly linked to the Huawei-backed SiCarrier, is said to share design similarities with older ASML models. While this represents a significant engineering effort, it remains several generations behind the technology currently used for 7nm and 5nm production. As of Q3 2025, Chinese fabs still relied on ASML for 42% of their system sales by revenue, underscoring the continued necessity of foreign imports for advanced nodes.
Regulatory Pressures and Future Outlook
The urgency behind the Chang-Yin appeal is underscored by the U.S. Legislative landscape. The proposed MATCH Act, which has already passed the House Foreign Affairs Committee with a 36-8 vote, seeks to designate several major Chinese firms—including SMIC, YMTC, and AMEC—as “Covered Facilities.” If enacted, the bill would impose a country-wide prohibition on the export of DUV immersion lithography tools to China.
As this bill moves toward a potential Senate floor vote, the Chinese semiconductor industry finds itself in a race against time. The broadcast on May 17th served as a clear signal from Beijing: the next three to five years will determine whether China’s domestic toolmakers can evolve from prototyping to consistent, high-yield volume manufacturing. For now, the industry awaits the next legislative update from Washington and the subsequent impact on global supply chain access.
This report is for informational purposes and does not constitute financial or investment advice. We encourage our readers to share their perspectives on the shift toward domestic semiconductor manufacturing in the comments below.
