Could New US Export Rules Derail China's Tech Progress?
The next decisive battle between the United States and China may not be fought with ships, missiles, or soldiers. It may be fought through export licenses, semiconductor thresholds, manufacturing equipment, and the tiny components hidden inside artificial-intelligence data centers.
Washington has spent years tightening restrictions on China’s access to advanced computer chips and the machines required to manufacture them. The stated goal is not to stop ordinary commercial development, but to prevent sensitive American technology from supporting Chinese military modernization, intelligence operations, cyber capabilities, and advanced artificial intelligence.
Now, another round of regulatory action appears to be approaching.
A senior U.S. Commerce Department official said in July 2026 that further action involving artificial intelligence and semiconductors was coming, although the administration did not plan to recreate the previous global AI-chip distribution framework. At the same time, Washington has been moving to close loopholes that may allow restricted chips to reach Chinese companies through subsidiaries located outside mainland China.
The central question is enormous: Can tighter American export rules actually derail China’s technological rise?
The answer is complicated.
New restrictions could significantly increase China’s costs, delay access to cutting-edge computing power, and create serious obstacles for semiconductor manufacturers. But China has already shown that pressure can produce adaptation as well as damage. Chinese firms are developing domestic chips, improving older equipment, redesigning software, building alternative supply chains, and investing heavily in technological self-reliance.
Export controls may slow China down. Whether they can stop it is another matter.
Why Advanced Chips Matter So Much
Modern artificial intelligence depends on three major ingredients: data, software, and computing power.
Of those three, computing power is the easiest for governments to physically restrict.
Advanced AI systems require enormous clusters of graphics processing units and related accelerators. These chips perform the calculations needed to train large language models, analyze surveillance data, simulate weapons systems, discover new materials, and operate increasingly sophisticated autonomous machines.
The most advanced chips are extraordinarily difficult to design and manufacture.
American companies remain influential in chip design, software, networking equipment, and manufacturing tools. Meanwhile, essential parts of the production process involve companies located in Taiwan, the Netherlands, Japan, South Korea, and other allied economies.
This international concentration gives Washington substantial leverage.
A Chinese company may design an advanced processor, but it could still require foreign design software, American intellectual property, imported equipment, specialized memory, or manufacturing services from an overseas foundry.
Export controls target those technological chokepoints.
The Rules Have Been Tightening for Years
The modern U.S. semiconductor-control strategy accelerated in October 2022, when Washington introduced broad restrictions on advanced computing chips, supercomputing technology, and semiconductor-manufacturing equipment destined for China.
Those restrictions were revised and expanded in 2023 and later supplemented through additional entity listings and manufacturing-equipment controls. The Bureau of Industry and Security continues to maintain public guidance covering the advanced-computing and semiconductor rules imposed on China.
The policy was designed to prevent China from easily purchasing the most advanced American chips while also limiting its ability to manufacture comparable chips domestically.
Washington subsequently targeted more Chinese companies and adjusted technical thresholds as chipmakers created products designed to comply with the rules.
This created a regulatory competition.
The United States changed the limits.
American semiconductor companies redesigned chips.
Chinese buyers searched for alternatives.
Washington tightened the rules again.
Each round demonstrated both the power and the limitations of export controls.
A More Flexible Approach in 2026
American policy has not moved in only one direction.
In January 2026, the Commerce Department revised its license-review policy for certain advanced semiconductors exported to China. Applications involving chips such as Nvidia’s H200 and AMD’s MI325X could be reviewed individually, provided that specific security conditions were satisfied.
This case-by-case approach reflected an important debate inside Washington.
One side argues that denying China all access to advanced American chips protects national security and preserves the U.S. technological advantage.
The other side warns that excessively broad restrictions may drive Chinese customers away from American products, reduce revenue available for U.S. research, and accelerate China’s determination to build a completely independent technology ecosystem.
Allowing some controlled sales may preserve American commercial influence while still limiting access to the most sensitive capabilities.
However, implementation has remained cautious. By mid-July 2026, a senior administration official said that shipments of approved H200 chips to mainland China and Hong Kong remained very limited.
The result is a policy balancing act: Washington wants to protect security without unintentionally surrendering the Chinese market to domestic competitors.
Closing the Overseas-Subsidiary Loophole
One of the most difficult enforcement problems involves Chinese companies operating outside China.
A restricted firm might establish a subsidiary in Malaysia, Singapore, the Middle East, or another jurisdiction. That foreign entity could attempt to purchase chips, rent computing capacity, or access cloud infrastructure that the parent company could not legally obtain directly.
In May 2026, the United States took steps intended to stop advanced Nvidia chips from reaching Chinese companies through overseas subsidiaries. American officials were concerned that large numbers of chips might have entered networks connected to Chinese firms despite restrictions on direct shipments to mainland China.
Members of Congress have also called for tighter rules covering contract chip manufacturers. In June 2026, a bipartisan pair of senators urged the administration to prevent foreign foundries from manufacturing advanced AI chips for overseas subsidiaries of Chinese companies.
Such rules could be far more consequential than a simple ban on direct exports.
They would attempt to regulate not only where a chip is shipped, but also who ultimately controls it, who designed it, who benefits from it, and how it will be used.
That creates a much more complicated compliance system for semiconductor companies, banks, distributors, cloud providers, and overseas governments.
The Immediate Impact on China
Stricter enforcement could create several immediate problems for Chinese technology firms.
First, advanced computing capacity could become more expensive.
If Chinese AI developers cannot legally buy the most powerful processors, they may need to use larger quantities of less capable chips. That increases electricity consumption, cooling requirements, data-center space, engineering complexity, and training time.
Second, Chinese semiconductor manufacturers could struggle to upgrade their production lines.
The most advanced chips require highly sophisticated lithography, inspection, deposition, etching, and design technologies. Losing access to even one critical category of equipment can delay an entire manufacturing process.
Third, uncertainty could discourage investment.
Chinese companies may hesitate to design products around foreign components that could later become restricted. International suppliers may avoid Chinese customers because they fear legal penalties or reputational damage.
These effects can compound.
A delayed chip affects a server.
A delayed server affects a data center.
A delayed data center slows the development of an AI model.
A delayed model affects commercial applications, scientific research, and military capabilities.
That is the logic behind Washington’s strategy.
China’s Most Serious Weakness: Manufacturing Equipment
China has made rapid progress in chip design and manufacturing, but it remains constrained in some of the most advanced production processes.
Extreme ultraviolet lithography machines are particularly important. These machines help print extremely small circuits onto advanced chips and are produced by the Dutch company ASML.
China has been unable to freely acquire the most advanced EUV systems because of international export restrictions.
That limitation has not prevented Chinese manufacturers from producing capable chips. However, it can reduce efficiency, lower production yields, increase costs, and make mass production more difficult.
Recent reporting on China’s memory-chip industry illustrates this mixed picture. Chinese companies such as ChangXin Memory Technologies and Yangtze Memory Technologies have gained influence and expanded rapidly, but they continue to face restrictions on the most advanced foreign manufacturing equipment.
This may be where U.S. controls have their greatest effect.
Algorithms can be copied.
Software can be rewritten.
Engineers can find creative solutions.
A highly specialized manufacturing machine containing thousands of advanced components is much harder to reproduce quickly.
But China Is Not Standing Still
The greatest mistake Washington could make would be assuming that China will remain technologically frozen.
Chinese companies are responding aggressively.
Huawei has developed domestic AI processors and systems designed to combine large numbers of chips. Chinese engineers have used advanced packaging, software optimization, and three-dimensional stacking to improve performance even when individual chips remain behind the leading American products.
China has also launched a major national effort to reduce its dependence on foreign AI hardware. According to recent reporting, the share of foreign chips used in parts of China’s AI ecosystem has declined substantially, although significant hardware limitations remain.
State financing is another major factor.
Chinese semiconductor firms can receive government-backed funding, tax benefits, cheap credit, research support, and preferential purchasing arrangements. These resources do not guarantee technological success, but they allow companies to sustain costly development programs longer than ordinary market conditions might permit.
Export controls therefore create a race against time.
Can the United States preserve its advantage faster than China can replace restricted technologies?
That question matters more than whether controls inflict immediate pain.
Restrictions Could Accelerate Self-Reliance
Export controls contain a paradox.
The more China believes it cannot depend on American technology, the more urgently it will invest in alternatives.
Before restrictions, a Chinese company might prefer an Nvidia processor because it offered better performance and had a mature software ecosystem.
After restrictions, that company may have no choice but to buy a less capable Chinese chip. Its engineers then begin optimizing software for the domestic product. Revenue flows to the Chinese supplier, allowing it to hire more researchers and improve the next generation.
Over time, an inferior alternative can become a competitive one.
Recent research argues that U.S. restrictions increased the strategic value of open and locally adaptable AI systems inside China. Chinese developers increased their participation in open-source AI ecosystems as access to advanced hardware became more difficult.
This does not mean export controls are ineffective.
It means their impact changes over time.
In the short term, they create shortages and delays.
In the long term, they may encourage the development of competing technologies.
Software Efficiency Can Reduce the Hardware Gap
AI progress does not depend exclusively on owning the fastest chip.
Developers can improve model architecture, reduce numerical precision, compress data, optimize memory use, and distribute workloads more efficiently across multiple processors.
These techniques allow companies to produce stronger results using weaker hardware.
Researchers examining earlier restrictions found that Chinese firms could use legal, lower-performance processors and sophisticated software techniques to reduce the practical impact of hardware controls.
That is a critical challenge for American policymakers.
Export rules are based on measurable technical thresholds such as processing speed or data-transfer capacity. But the commercial value of a chip depends on how it is used.
A processor below the legal threshold might still become powerful when thousands are connected together.
A slower chip might deliver excellent results after software improvements.
A model may require less computing power because its developers discovered a more efficient training method.
Regulators are trying to control a rapidly moving technology using rules that may take months to write and revise.
By the time a restriction becomes effective, the industry may already have found a workaround.
Smuggling and Remote Access
Physical chips are small, expensive, and difficult to track after they enter international distribution networks.
That creates opportunities for diversion.
A chip legally shipped to one country could be resold through intermediaries. Documentation could conceal the true customer. Servers could be purchased by shell companies. Restricted firms could rent computing capacity abroad rather than importing hardware.
Cloud access creates an especially difficult problem.
A Chinese developer may not need to physically possess an advanced processor if it can remotely use a data center located in another country.
Restricting such activity would require rules governing identities, beneficial ownership, remote access, and computing services.
Those rules might strengthen enforcement, but they could also impose burdens on legitimate cloud customers and global technology companies.
The broader the controls become, the more difficult they are to administer without disrupting ordinary commerce.
The Cost to American Chipmakers
Export controls do not affect only China.
American semiconductor companies risk losing sales in one of the world’s largest technology markets.
Reduced revenue can affect research spending, manufacturing investment, and shareholder returns. Chinese customers may also redesign their software and equipment around non-American alternatives, making it difficult for U.S. companies to return even if restrictions are later relaxed.
This produces another strategic dilemma.
American policymakers want to prevent China from obtaining technology that could threaten national security.
But the long-term American advantage also depends on the financial strength of U.S. companies.
If restrictions reduce their global scale while providing Chinese competitors with a protected domestic market, the policy could eventually weaken the companies it was intended to defend.
Supporters of export controls respond that national security cannot be measured only through quarterly revenue. They argue that allowing commercial sales today could help China develop military capabilities that create much larger costs tomorrow.
Both concerns are legitimate.
The challenge is deciding where the line should be drawn.
America Needs Its Allies
U.S. export controls become much less effective when imposed alone.
Semiconductor supply chains include Dutch lithography systems, Japanese materials and equipment, Taiwanese manufacturing, South Korean memory chips, European components, and American design technology.
If Washington restricts an item but a foreign company can supply an equivalent product, China may simply change suppliers.
That is why diplomatic coordination is essential.
The United States has worked with allied governments to restrict some advanced semiconductor-manufacturing equipment. But allies do not always share the same economic interests or threat perceptions.
A European or Asian government may agree that military technology should be protected while resisting restrictions that appear designed to benefit American companies.
Overly aggressive pressure could create resentment and reduce cooperation.
Effective controls must therefore be narrow enough to preserve allied unity while strong enough to affect Chinese capabilities.
China Can Retaliate
Beijing also possesses economic leverage.
China plays a major role in processing critical minerals used in electronics, batteries, telecommunications equipment, renewable-energy systems, and defense products.
It can impose licensing requirements, slow customs approvals, investigate foreign firms, or restrict access to strategically important materials.
China is also considering tighter controls on exports of its own AI technologies, semiconductor designs, model data, and other sensitive assets. Those proposals remain under discussion, but they demonstrate how technological restrictions can spread in both directions.
A cycle of retaliation could divide the global technology industry.
American companies would lose access to Chinese customers and materials.
Chinese companies would lose access to American chips and software.
Allies and developing nations would face pressure to align with one system.
Innovation could slow as researchers, capital, data, and components become trapped inside competing geopolitical blocs.
Could the Rules Actually Derail China?
The word “derail” suggests that China’s technological progress could be knocked permanently off course.
That is unlikely.
China possesses an enormous domestic market, world-class engineers, major universities, powerful technology companies, substantial government funding, and strong manufacturing capabilities.
Those advantages cannot be eliminated through export licensing.
However, new rules could delay specific objectives.
They could make it harder for China to build the largest AI training clusters.
They could slow the development of advanced semiconductor factories.
They could raise the cost of achieving the same computing performance.
They could limit the capabilities available to military and intelligence organizations.
In a fast-moving technological competition, a delay of several years can matter enormously.
A country that reaches a breakthrough first may establish global standards, attract talent, dominate markets, and gain military advantages.
Export controls do not need to stop all Chinese progress to serve Washington’s objectives.
They need to preserve enough time for the United States and its allies to move faster.
The Bigger Risk for Washington
The greatest danger is believing that restrictions can substitute for domestic investment.
America cannot maintain technological leadership only by denying tools to competitors.
It must also fund research, train engineers, build power infrastructure, expand advanced manufacturing, secure mineral supplies, attract global talent, and create conditions in which innovative companies can grow.
Controls can slow an opponent.
They cannot produce American innovation.
Without a strong domestic strategy, export restrictions may simply create a temporary advantage that gradually disappears.
The most effective policy would combine carefully targeted controls with massive investment in the American technology ecosystem.
Conclusion
New U.S. export rules could disrupt China’s technological progress, particularly in advanced semiconductors, artificial intelligence computing, and chip-manufacturing equipment.
They could raise costs, delay projects, restrict access to critical hardware, and make it harder for Chinese firms to compete at the technological frontier.
But they are unlikely to permanently derail China’s rise.
China is adapting through domestic chip development, state investment, open-source software, engineering workarounds, alternative supply chains, and increasingly sophisticated manufacturing.
Some restrictions may even accelerate Beijing’s campaign for technological independence.
The real contest is therefore not a simple battle between prohibition and access.
It is a race between American control of today’s technological chokepoints and China’s ability to eliminate those chokepoints tomorrow.
If Washington targets restrictions carefully, coordinates with allies, closes genuine loopholes, and invests aggressively at home, export controls may preserve a meaningful U.S. advantage.
If the rules become overly broad, unpredictable, or primarily commercial, they could damage American companies, alienate partners, encourage smuggling, and help create a separate Chinese technology system beyond Washington’s influence.
America can make China’s path harder.
It can make progress more expensive.
It can delay access to the world’s most advanced tools.
May you like
But no export rule can erase China’s engineers, factories, capital, or determination.
The final outcome will depend not only on what the United States prevents China from buying, but on which country proves more capable of inventing what comes next.
