US shifts stance on China chip exports amid ongoing trade talks

In 2025, the United States will change its approach to chip exports to China, signaling a shift that could alter the global semiconductor landscape, especially in the context of china chips. The US controls nearly half of global chip design revenue and over 60% of semiconductor intellectual property. While China continues to invest heavily in its own china chips industry, trade restrictions have slowed its progress.

Ongoing trade talks between the two nations shape these policies. Business leaders and technology experts watch closely, as these changes impact supply chains and innovation worldwide.

Key Takeaways

• The US changed its chip export rules in 2025 to balance national security with trade interests, easing some restrictions while tightening others.

• New export controls limit China’s access to advanced AI chips and design software, pushing China to invest more in its own chip technology.

• US companies face higher compliance demands and costs but also find new growth in AI and supply chain diversification.

• Global supply chains shift as firms adopt the “China +1” strategy, moving some production to countries like Vietnam and India.

• Ongoing US-China trade talks influence chip policies, creating uncertainty but also opportunities for cooperation and market stability.

Policy Changes

2025 Timeline

The United States began shifting its chip export policy to China in early 2025. In May, the government rescinded the AI Diffusion Rule, which had set strict limits on advanced AI chip exports. This change came just days before the rule was set to take effect. By July 2025, the Department of Commerce planned to issue new executive orders. These orders focus on coordinating government agencies and setting a path forward, but they do not provide a detailed phased timeline for implementation. The new rules remain in drafting stages, with formal regulatory actions expected around mid-2025. Other countries, such as Malaysia, have responded by introducing their own permit requirements for U.S.-origin AI chip exports, showing how U.S. policy changes can influence global regulations.

Note: The lack of a clear, phased timeline means companies must stay alert for sudden regulatory changes throughout 2025.

Key Shifts

The 2025 policy changes mark a significant departure from previous years. The U.S. government now uses a mix of tightening and easing measures to balance national security with economic interests. Major changes include:

• The U.S. Commerce Department suspended some export licenses for technology sales to Chinese firms, such as COMAC, and paused exports of semiconductor design software from leading companies.

• In late May, the U.S. imposed new restrictions on electronic design automation (EDA) software exports to China, but lifted these restrictions in early July after trade talks. This move allowed major business deals to proceed and led to positive market reactions for EDA companies.

• The U.S. blocked sales of advanced AI chips, including Nvidia’s H20 chips, to China and threatened to revoke waivers for allied chip plants operating in China.

• The government introduced a tiered export control framework, classifying countries into three groups with different access levels to U.S. AI technology.

This framework extends beyond hardware, covering AI model weights and cloud computing services. The U.S. aims to prevent countries of concern, especially China, from accessing advanced AI capabilities.

• The U.S. also introduced a global license regime for frontier AI model weights, expanded licensing controls on advanced AI chips, and created new frameworks for validated end users.

• Earlier export controls focused mainly on restricting chipmaking equipment and end-user lists. The 2025 rules now include controls on AI model weights and cloud access, closing previous loopholes.

Policymakers argue that these measures protect national security and maintain U.S. leadership in AI and chip design. However, some restrictions, such as those on EDA software, have been used as bargaining chips in trade negotiations. Easing these controls can help stabilize global supply chains but may also encourage China to accelerate its own chip development.

The lifting of EDA software restrictions had immediate effects. Stock prices for major EDA firms rose, and business deals resumed. Yet, the episode highlighted China’s reliance on foreign EDA tools and spurred further investment in domestic alternatives. Many global semiconductor companies now follow a “China +1” strategy, diversifying manufacturing to reduce dependence on china chips and the Chinese market.

The U.S. approach in 2025 is more flexible and negotiation-driven than in previous years. The government balances pressure on China’s semiconductor industry with efforts to protect commercial interests and maintain global supply chain stability. This strategy reflects a broader trend of using export controls as both a security tool and a lever in trade talks.

China Chips Export Controls

AI and Semiconductor Rules

The United States has steadily increased export controls on advanced AI chips and semiconductor manufacturing equipment to China. These rules aim to limit China's ability to develop high-end semiconductors for artificial intelligence, high-performance computing, and military applications. Since October 2022, the U.S. has added more types of semiconductor equipment and software tools to its restricted list. By December 2024, the list included 24 types of manufacturing equipment and three key software tools. The U.S. also placed 140 Chinese entities on the Entity List, which means companies need special licenses to do business with them.

In January 2025, the U.S. introduced the AI Diffusion Rule. This rule required companies worldwide to get licenses before exporting advanced integrated circuits to China. However, the Trump administration later replaced this rule with targeted restrictions. Now, only specific chips, such as Nvidia’s H20 and AMD’s MI308X, can be exported to China, while more powerful chips like Nvidia’s H100 remain banned. The U.S. also imposed new controls on Electronic Design Automation (EDA) software, which is essential for designing advanced chips at 7nm, 5nm, and 3nm nodes. These controls make it harder for China to advance its own chip industry.

U.S. companies must comply with new export control classification numbers for advanced integrated circuits, including high-bandwidth memory chips. They need to conduct thorough due diligence and follow red flag guidance to prevent diversion of technology to unauthorized users. Exporters must also resolve any compliance issues before completing transactions, especially when dealing with foreign-produced items that contain controlled chips. These rules cover not only hardware but also AI model weights used for training AI systems in China.

EDA software has become a critical choke point in the global semiconductor supply chain. U.S. controls on this software have a major impact on both American and Chinese companies, as many Chinese customers rely on foreign EDA tools to design their chips.

China has responded by investing in domestic EDA firms and promoting self-sufficiency in chip design. However, technological gaps remain, especially for advanced nodes. The U.S. added leading Chinese EDA companies, such as Empyrean Technology, to its Entity List in December 2024, further limiting their access to global markets. The Chinese government has also enacted laws to counter foreign sanctions and protect its companies, while encouraging the use of domestic and open-source software.

License Reviews

The U.S. government reviews license applications for advanced chip exports to China under a strict "resumption of denial" policy. Most requests are denied by default, especially for items that could support china chips development in sensitive sectors. Only a few exceptions exist for end users based in the United States or allied countries, where the government reviews applications on a case-by-case basis.

The main criteria for license reviews include:

• The end use of the chips or equipment, such as whether they will be used in semiconductor fabrication or supercomputer manufacturing in China.

• The end user’s location and status, with special attention to companies on the Entity List.

• Technical thresholds, such as logic integrated circuits at 16/14 nm or less, NAND memory with 128 layers or more, and DRAM with 18 nm half-pitch or less.

• The exporter’s knowledge of how the items will be used and whether they could support military or high-performance computing applications.

No license exceptions are available for these controls. Companies must apply for individual licenses, and temporary general licenses do not override the denial policy. The U.S. government also faces challenges in conducting end-use checks and verifying compliance, which affects how it designates entities for additional restrictions.

U.S. companies face several compliance challenges under the new export control regime:

1. They must accurately classify goods and technologies and understand all licensing requirements.

2. Firms need to manage supply chain risks and ensure that suppliers and partners follow the rules.

3. Companies must dedicate resources to compliance, including workforce training and robust screening processes.

4. Enhanced due diligence is required for counterparties, especially in high-risk markets like china chips.

5. Businesses must navigate conflicting international sanctions, which adds complexity to multinational operations.

The strict enforcement of these rules has led to significant financial impacts for U.S. companies. For example, Nvidia reported a $5.5 billion loss due to new licensing requirements for its high-performance AI chips sold to China.

China continues to adapt by promoting domestic innovation and securing technology transfers before new restrictions take effect. The government encourages local companies to develop their own solutions and maintain access to critical software and hardware, even as U.S. controls tighten.

Trade Talks Impact

Tariffs and Truces

US-China trade talks have played a major role in shaping chip export policies. Negotiators from both countries have focused on reducing tariffs and restoring the flow of critical goods, such as Nvidia’s H20 AI chips and rare earth minerals. The US paused some technology export curbs to support these talks, which led to a temporary truce in the ongoing trade war. During this period, China resumed rare earth mineral exports, a move that helped US chipmakers maintain production.

Companies like Foxconn have responded by shifting manufacturing capacity from China to countries such as Vietnam and India. This strategy, known as "China +1," helps firms avoid the high 30% tariff threshold and manage geopolitical risks. Some electronics suppliers have also kept design centers in China while diversifying their supply chains. Although the US reduced some tariff rates for a short time, warnings of future increases have forced companies to rethink their sourcing and logistics plans.

China’s near-monopoly on rare earth magnets gives it leverage in negotiations, influencing US decisions on chip export restrictions. These trade talks have created a delicate balance between economic interests and national security concerns.

Future Outlook

Looking ahead, analysts expect the current tariff truce to continue, with tariffs on Chinese goods staying around 30%. The main challenge remains US technology export controls on chips critical for AI. Beijing insists on progress in this area, making negotiations complex. Both sides have made goodwill gestures, such as China suspending an antitrust probe against an American firm and the US lifting a ban on a key Nvidia AI chip sale.

• The US Commerce Department has started a Section 232 investigation to assess national security risks from chip imports.

• Semiconductors and related equipment may face new tariffs or export controls.

• China uses its control over strategic minerals to pressure the US on export controls.

• The relationship remains fragile, with possible flashpoints that could disrupt progress.

Many experts believe that the talks could lead to a meeting between President Trump and President Xi later in the year. However, a final resolution on chip export controls may take longer. Companies dealing with china chips must stay alert for sudden changes in policy as trade talks continue.

Industry Effects

US and China Companies

US chip companies have responded to the 2025 policy changes with bold strategies. Many firms now focus on AI infrastructure, which drives most of their revenue growth. Gartner predicts the semiconductor market will grow by about 11.8% in 2025 and 11.2% in 2026. Companies like Intel and Micron have increased capital spending, with some forecasting up to 75% growth in investments. These firms also invest in workforce development, working with schools and local governments to train new talent. AI now leads as the top revenue source, pushing companies to expand research and development in high-bandwidth memory and advanced packaging.

Strategic partnerships have become common. For example, Apollo invested $11 billion in Intel’s Fab 34, and Vanguard International Semiconductor joined with NXP to share risks. US companies also adapt by reshaping supply chains, moving some operations closer to home or to friendly countries. Internally, many firms use generative AI to improve IT, research, and even marketing.

Chinese semiconductor companies face different challenges. US export controls have limited their access to advanced technology. In response, Chinese firms have formed new partnerships within China and invested heavily in research. Companies like Huawei and SMIC have developed advanced chips, such as 7nm technology, despite restrictions. The government encourages local firms to use Chinese-made semiconductors, which has changed supply chains and increased the focus on self-reliance. This shift has created a more divided global market, making collaboration harder but pushing innovation in china chips.

Global Supply Chain

The 2025 US chip export policy has reshaped the global semiconductor supply chain. Tariffs on Chinese semiconductor imports have reached up to 145%, causing a sharp drop in exports to the US. Companies like AMD expect large revenue losses, with projections of $1.5 billion lost in 2025 due to export restrictions. About 30% of businesses now take a cautious approach, waiting to see how policies evolve.

Many firms have diversified their supply chains. They invest more in domestic manufacturing and look to countries like India, which benefits from lower tariffs. The US has tripled its semiconductor fab capacity, aiming to boost its share of global production from 10% to 14% by 2032. Investments from the CHIPS Act have reached nearly $450 billion, supporting new factories and research centers.

Other countries have also responded. Malaysia now requires special permits for high-performance US-origin AI chips, and India has strengthened its export control programs. The US works closely with partners in the Indo-Pacific to prevent chip diversion to China. These actions show a global trend toward stronger export controls and more reliable supply chains.

The global semiconductor market remains strong, but the industry faces new risks. Companies must adapt quickly to changing rules and invest in compliance to avoid disruptions. The push for self-reliance and new partnerships continues to shape the future of the semiconductor industry.

The US continues to reshape chip export policy, driving global supply chain changes and new compliance demands. Companies face rising costs, shifting alliances, and rapid regulatory updates.

Experts predict ongoing rivalry and innovation as both countries invest in technology and seek supply chain resilience.
Key trends to watch include expanding export controls, evolving compliance rules, and the impact of new alliances. Leaders must stay alert to shifting risks and opportunities in the semiconductor industry.

FAQ

What is the main goal of US chip export controls on China?

The US aims to protect national security and keep its lead in advanced technology. These controls limit China’s access to high-end chips and software used for artificial intelligence and military applications.

How do the new rules affect US and Chinese companies?

US companies face stricter compliance checks and higher costs. Chinese firms must find new suppliers or develop their own technology. Both sides adjust supply chains and business strategies to manage risks.

Why did the US lift some EDA software restrictions in 2025?

The US lifted EDA software restrictions to support trade talks and help global supply chains. This move allowed important business deals to continue and eased pressure on American technology firms.

What is the “China +1” strategy?

Many companies use the “China +1” strategy to reduce risk. They keep some operations in China but add factories or suppliers in other countries like Vietnam or India.

Will chip export policies change again soon?

Policymakers review export controls often. Companies should watch for updates, as new rules or changes can happen quickly during ongoing trade talks.

See Also

Advancements Shaping Local Memory Chip Production In 2025

New Developments In Automotive Chip Technology Trends

Proven Approaches To Overcome Electronic Part Shortages

Ways Analog Devices Maintains A Strong Chip Supply Chain

Examining How Chip Shortages Affect Avionics Systems