03/24/2026 | Press release | Distributed by Public on 03/24/2026 18:31
Photo: Maksym Yemelyanov/Adobe Stock
Commentary by Sujai Shivakumar, Charles Wessner, and Thomas Howell
Published March 24, 2026
U.S. and allied export controls on advanced semiconductor technologies, imposed beginning in 2022, were designed to constrain China's AI and high-end chip development. While progress on these goals has been limited, the controls have clearly accelerated Beijing's long-standing drive for semiconductor self-reliance. The initial round of controls restricted transfers of advanced logic chips used to train and operationally use AI models and semiconductor manufacturing equipment (SME) used to make those devices. The controls have subsequently been expanded to additional device, design, and SME categories. While the controls have disrupted Chinese access to leading-edge chips and equipment, their principal effect has been to accelerate the adoption and use of indigenous equipment and products, giving new impetus to coordinated state and industry efforts to localize semiconductor design and manufacturing. The Chinese government has reportedly directed domestic enterprises to acquire domestic devices and SME that are less advanced than foreign alternatives in order to advance national self-sufficiency.
The Chinese response of massive public investment and government procurement mandates, coupled with the perception of heightened supply chain risk, is driving substitution of foreign chips and equipment, expanding domestic production capacity, raising adoption of Chinese-made AI processors, and growing market share for local semiconductor equipment suppliers. At the same time, comprehensive national and provincial initiatives are fostering indigenous innovation in areas ranging from AI accelerators and RISC-V architectures to semiconductor materials and lithography.
Collectively, these developments suggest that export controls, while limiting China's short-term access to frontier technologies, are inadvertently catalyzing a more unified, urgent, and technologically ambitious push toward semiconductor autonomy.
In 2022, the Biden administration established controls on the export of semiconductor devices, equipment, and technology to China, controls which have subsequently been expanded and tightened. The stated purpose of the new controls was to forestall China's ability to develop an indigenous chip industry capable of competing with U.S. and allied chipmakers at the most advanced nodes, which support AI.
Official production data shows that China's integrated circuit output fell by 9.8 percent in 2022 following the imposition of export controls, which was the first decline in several years. In the near term, export controls on SME have effectively limited China's ability to produce chips beyond the 7-nanometer (nm) node.
However, it is now clear that U.S. and allied export controls are inadvertently adding impetus to the Chinese government's long-standing goal of localizing the domestic supply of semiconductors. More importantly than that, these controls are coordinating customer demand and utilization within China with the roadmaps and output of Chinese technology providers.
For decades, the Chinese government has pressured domestic chip-consuming firms to increase their sourcing of Chinese chips, with disappointing results and reluctant domestic customer adoption given the generally superior performance of foreign-made devices. Allied export controls, however, have added a new dimension to the government's industrial policy efforts-the prospect that Chinese chip-consuming enterprises will be cut off from access to high-end foreign-made chips, or will only be able to obtain them through irregular means such as smuggling.
In 2014 and 2015, the Chinese government launched its most ambitious effort to promote the expansion and international competitiveness of indigenous chipmakers, and to build out a domestic infrastructure of equipment and materials suppliers. Thus far, it is estimated that $150 billion in public funding has been channeled into this effort, which set a target of 70 percent localization of domestic chip supply by the end of 2025. By contrast, the United States' principal recent federal semiconductor initiative was the 2022 CHIPS and Science Act, which authorizes approximately $52.7 billion in manufacturing incentives, research, and workforce support, substantially smaller in scale than China's cumulative state-backed investment, and most of the R&D portion of the Chips Act is now on hold.
The effort, part of a broader national program, Made in China 2025, enabled a significant expansion of the Chinese semiconductor industry, but fell far short of its localization target, with domestically produced chips accounting for about 30 percent of domestic consumption in 2025. Furthermore, customer adoption for the more advanced chip and semiconductor manufacturing equipment products was relatively anemic.
The advent of allied chip export controls, however, has galvanized the central and regional governments and domestic industry into developing autonomous design, manufacturing, and infrastructure capabilities in its semiconductor ecosystem. While comprehensive data on the pace of China's localization effort in chips is not available, there are indicators of a major shift toward domestic sources by Chinese semiconductor consumers, which is attributable to the stimulus of the controls.
Since the 2022-2023 imposition of semiconductor export controls, Chinese end-users have increased their sourcing of domestically made chips, and this has been reflected in declining sales of foreign chips in China.
The trend has been most observable in mature node and commodity devices, in which Chinese capacity increased four times faster than global demand in the years 2014-2025, now accounting for about half of global capacity. Chinese consuming firms using such "legacy foundational" or "workhorse" chips are reportedly turning to import substitution procurement practices as a hedge against potential supply chain disruptions from new allied export controls, as a response to government mandates and prodding, and as a reflection of the growing competitiveness of domestically made mature node devices.
But the localization trend is observable in more advanced nodes as well. TrendForce projects that in 2026, the domestic share of China's AI chip market will increase to 50 percent.
The Trump administration recently approved the export of Nvidia's second-most powerful AI graphics processing unit (GPU) to China, the H200, as well as AMD's GPU, Instinct MI308, but the Chinese government is encouraging a boycott of U.S. devices, "unless absolutely necessary," and is reportedly planning new rules placing a cap on the total number of advanced AI chips Chinese firms are allowed to import. "After being tied up by U.S. export controls and sanctions, it has become apparent that escaping this chokehold is more important in the long run than making any temporary advances."
Prior to the tightening of allied export controls, Chinese-made semiconductor manufacturing equipment accounted for only about 10-15 percent of China's domestic market. Despite sustained state support beginning in 2014, domestic penetration remained limited through 2021 due to low customer demand and adoption. Conversely, in the years following the imposition of export controls, that share has surged from 25 percent to 35 percent of the market between 2024 and 2025, exceeding the Made in China 2025 target of 30 percent. Progress has been particularly dramatic in etching and thin film deposition, where local firms now supply 40 percent of the domestic market. These gains probably reflect sales of equipment serving the mature nodes-Naura reportedly accounts for 60 percent of the oxidation and diffusion furnaces serving Semiconductor Manufacturing International Corporation's 28 nm node chip production.
In 2024, China's Ministry of Industry and Information Technology reportedly told Chinese telecom operators that they must remove all foreign semiconductors from their networks by 2027. Huawei, the largest smartphone maker in China and a major producer of IT equipment, is making rapid strides toward its goal of phasing out all foreign-made chips from its products.
Paralleling and reinforcing the localization drive, the central and regional governments and Chinese industry are undertaking an unprecedented effort to develop domestic alternatives to foreign semiconductors, chipmaking equipment, and materials.
A comprehensive 2025 CSIS study of the impact of export controls in SME reported that Chinese chipmakers were designing out US equipment. The report stated that "many U.S. toolmakers are still seeing increasing sales to China due to surging industry growth. However, companies told CSIS that this growth is significantly below what it would otherwise be in the absence of [Chinese] design-out practices."
In the short run, semiconductor export controls directly constrained China's access to the most advanced AI accelerators, high-performance logic chips, and cutting-edge manufacturing equipment, particularly GPUs and EUV-related tools. This has had a near-term effect of slowing the deployment of large-scale AI models, advanced military simulations, and supercomputing workloads that rely on thousands of leading-edge chips.
U.S. policymakers who began tightening controls on chip technology exports to China in 2022 presumably recognized that their actions would prompt China to try harder to achieve self-sufficiency in semiconductors, but what is not clear is whether they foresaw the intensity and speed of the Chinese response. The expansion of domestic production capacity for legacy chips and SME may have been an unforeseen consequence of U.S. actions. China is already rapidly supplanting foreign chips in its market with domestic ones and is embarking on an innovation drive that is already producing impressive results.
As a result of these efforts, any short-term advantage gained from these controls may be just that. Ironically, U.S. measures are proving more successful at catalyzing true Chinese chip innovation, as opposed to reverse engineering-manifested in recent achievements in advanced packaging, alternative forms of lithography, and other SME, and novel systems architecture. Thus, the controls may be achieving a goal that decades of promotional measures by Chinese industrial policy planners could not, with long-term implications for national security and technological competitiveness.
Sujai Shivakumar is the director and senior fellow of Renewing American Innovation at the Center for Strategic and International Studies (CSIS) in Washington, D.C. Charles Wessner is a senior adviser (non-resident) for Renewing American Innovation at CSIS. Thomas Howell is an international trade attorney specializing in the semiconductor industry and a consultant with the CSIS Renewing American Innovation.
Commentary is produced by the Center for Strategic and International Studies (CSIS), a private, tax-exempt institution focusing on international public policy issues. Its research is nonpartisan and nonproprietary. CSIS does not take specific policy positions. Accordingly, all views, positions, and conclusions expressed in this publication should be understood to be solely those of the author(s).
© 2026 by the Center for Strategic and International Studies. All rights reserved.
Commentary by Sujai Shivakumar, Charles Wessner, and Thomas Howell - April 14, 2025
Report by Charles Wessner and Shruti Sharma - November 6, 2025
Blog Post by Sujai Shivakumar and Thomas Howell - October 21, 2025