The global semiconductor industry ushered in a landmark policy adjustment at the dawn of 2026. On January 1, TSMC officially announced that it had obtained a 2026 annual export license issued by the U.S. government. This license allows the company to continuously ship U.S.-made chip manufacturing equipment to its Nanjing fab in mainland China without the need for case-by-case approval, ensuring the uninterrupted operation of the fab and the delivery of its products.

Coming hot on the heels of similar annual licenses granted earlier to Samsung Electronics and SK hynix, this decision marks a phased shift in the U.S. export control policy on chip equipment toward China – evolving from a blanket ban to an annual review mechanism. It injects a new variable into the global chip supply chain amid complex strategic games.

The backdrop to this policy adjustment stems from a major revision of the U.S. semiconductor export control rules implemented previously. In August 2025, the U.S. Department of Commerce’s Bureau of Industry and Security (BIS) announced the revocation of the indefinite Validated End-User (VEU) exemptions enjoyed by Samsung, SK hynix and TSMC. The original policy expired on December 31, 2024, nullifying the equipment export privileges that once required no individual approval.

This move sent shockwaves through the industry: Samsung and SK hynix’s fabs in China account for nearly half of the world’s DRAM production capacity and the output of core memory chips. Meanwhile, TSMC’s Nanjing fab focuses on manufacturing mature-process chips at the 16nm node and above, which are widely used in automotive, industrial and consumer electronics sectors. Any disruption to equipment supply would have directly exacerbated global supply chain chaos.

Compared with the previous VEU exemption system, the new annual licensing mechanism features a dual nature of "control and relaxation". Under the new rules, Samsung, SK hynix and TSMC are required to submit an annual list of equipment and material requirements in advance. After review and approval by the U.S. government, they receive a fixed-quota export license valid for one year.

The U.S. side has explicitly restricted enterprises from using the license to expand production capacity or upgrade advanced technologies. The core objective remains to curb the transfer of high-end chip technology to China and maintain its technological generation advantage. Industry analysts point out that this arrangement not only eliminates the administrative redundancy caused by case-by-case applications – South Korean companies alone would have had to handle an additional 1,000 license applications each year – but also enables the U.S. to continuously monitor the flow of technology through annual reviews, forming a control model of "limited openness and dynamic supervision".

From a deeper perspective, this U.S. policy shift is essentially a strategic trade-off under multiple pressures. On one hand, as the world’s largest semiconductor consumer market, China absorbs nearly half of the global demand for wafer fabrication equipment. A forced complete cutoff would lead to huge losses for relevant enterprises, and ultimately backfire on the interests of U.S. domestic equipment suppliers. Take NVIDIA as an example: its chips previously customized for the Chinese market were once banned from sale, resulting in a staggering $4.5 billion inventory loss for the company and a market value evaporation of $160 billion at one point.

On the other hand, China’s breakthroughs in semiconductor localization have diminished the marginal effectiveness of the U.S. blockade policy. From the control of strategic materials such as gallium and germanium to the steady progress in the performance of domestic chips, China has gained a certain level of substitution capability. The U.S. can no longer slow down China’s industrial upgrading process through unilateral blockade measures.

For the global semiconductor industry, this decision will play a "supply chain stabilizing" role in the short term. TSMC’s Nanjing fab can maintain stable production capacity of mature-process chips, and the equipment maintenance and production continuity of Samsung and SK hynix’s fabs in China are guaranteed, which helps alleviate the chip supply pressure in downstream industries such as automotive and consumer electronics.

However, in the long run, the core contradiction of the technological game has not been resolved. The U.S. still aims to contain China’s semiconductor industry from advancing toward high-end segments through measures such as annual reviews and technological restrictions, while China’s independent innovation drive has not slowed down due to this short-term relaxation. Data shows that from January to August 2025, China’s integrated circuit export value reached $126.512 billion, a year-on-year increase of 22.1%, with mature-process chips emerging as the mainstay of exports, demonstrating strong industrial resilience.

Industry insiders generally agree that the 2026 annual license is not the end of the China-U.S. chip game, but rather a new starting point. It not only exposes the reality of interdependence in the global semiconductor supply chain, but also serves as a reminder of the importance of independent control over core technologies. In the coming year, the global semiconductor landscape will continue to be characterized by "parallel systems and deepening games", and ultimately, it is the hard power of technological innovation and the resilience of the industrial chain that will determine the future direction of the industry.