U.S. chip tariffs are creating new considerations for semiconductor foundry operators, chip designers, suppliers, and technology companies. The current policy framework includes a targeted duty on certain advanced computing chips, separate trade measures involving China and Taiwan, and incentives intended to expand domestic fabrication. Together, these developments are influencing how companies assess production locations, sourcing arrangements, and long-term supply-chain commitments.
Kings Research projects the global semiconductor foundry market size to increase from USD 88.93 billion in 2025 to USD 95.88 billion in 2026, reaching USD 162.30 billion by 2033 and registering a CAGR of 7.8% over the forecast period. This outlook provides a commercial backdrop to the changing trade environment, although the effects of tariffs and investment incentives vary according to product classifications, manufacturing locations, and customer requirements.
The immediate policy impact is specific. A targeted duty applies to certain advanced computing chips, while other trade actions cover broader categories of imported goods. Meanwhile, investment commitments and domestic fabrication projects are influencing where companies may establish or expand production. Understanding the scope of each measure is essential for businesses evaluating their exposure to trade policy changes.
What Is a Semiconductor Foundry?
A semiconductor foundry manufactures integrated circuits according to designs supplied by external customers. Its clients include fabless chip companies, which concentrate on chip design and outsource fabrication, as well as integrated device manufacturers (IDMs) that may use external production partners to supplement their own facilities. The Semiconductor Industry Association's industry primer explains the distinctions between these business models.
A pure-play foundry focuses on manufacturing chips for external customers rather than designing and selling its own branded semiconductor products. An IDM, by comparison, manages both chip design and manufacturing, although it may also outsource some production.
Foundries compete through process technology, production capacity, yield performance, manufacturing reliability, and their ability to meet customer requirements. Their operations also depend on access to specialized equipment, materials, skilled workers, and dependable supply chains.
The relationship between foundries and fabless semiconductor companies is particularly important. Fabless businesses can develop products without making the substantial capital commitments required to build and operate fabrication facilities. They rely on foundries to convert their chip designs into physical products and support production at the required scale.
Foundry customers may also work with multiple manufacturing partners to manage production requirements and supply-chain exposure. Their decisions depend on process capabilities, manufacturing costs, delivery schedules, product qualification requirements, and the geographic distribution of fabrication facilities.
How Are Foundry Capabilities and Production Priorities Evolving?
Foundry operations vary according to process technology, customer requirements, and the applications being served. Advanced nodes support products that require high computing performance and power efficiency, while mature nodes remain important for industrial equipment, automotive systems, consumer electronics, and other applications.
Advanced fabrication capabilities are particularly relevant to processors used in artificial intelligence (AI), high-performance computing, and data centers. These applications require chips that can handle demanding workloads while managing power consumption and thermal performance. As a result, chip designers developing AI processors increasingly depend on foundries with suitable process technologies and manufacturing capabilities.
However, advanced fabrication is only one part of the semiconductor supply chain. Mature processes continue to support a wide range of electronic products, and customers may prioritize production reliability, cost efficiency, and established manufacturing processes over the smallest available node.
Geographic concentration also influences production planning. A substantial share of global fabrication activity is located in Asia, while governments and technology companies are seeking to expand manufacturing capacity in other regions. This creates opportunities for suppliers and equipment providers, but establishing new fabrication facilities requires significant capital, specialized infrastructure, and skilled personnel.
The development of AI chip technology is one factor supporting demand for advanced fabrication. Nevertheless, foundries must balance the requirements of AI and high-performance computing customers with demand from established applications that rely on different process technologies.
What Chip Tariffs Has the United States Imposed?
U.S. trade policy includes a targeted duty on certain advanced computing chips and separate Section 301 actions involving products from different trading partners. These measures have distinct legal bases and scopes. Companies should therefore assess them separately rather than treating them as a single tariff framework.
The targeted semiconductor measure was introduced under Section 232 of the Trade Expansion Act of 1962. It applies to specified advanced computing chips and derivative products when their importation does not contribute to the development of the U.S. technology supply chain and domestic manufacturing capacity.
The Federal Register's January 2026 proclamation established a 25% duty on covered products, effective January 15, 2026. The White House fact sheet identifies NVIDIA H200 and AMD MI325X chips among the products covered by the measure.
The proclamation also specifies exemptions based on intended use. These include qualifying imports for U.S. data centers, repairs and replacements, research and development, startups, non-data-center consumer applications, civil industrial applications, public-sector uses, and other purposes determined to support the domestic technology supply chain or manufacturing capacity. Importers must therefore consider the product classification and its intended use when determining whether the duty applies.
The Congressional Research Service discusses the relationship between the targeted duty and Taiwan-made chips shipped onward to China. Its analysis also raises questions about the legal basis for the measure. This distinction is important because the tariff's application depends on the relevant product and transaction rather than on a broad assumption that every imported advanced chip receives the same treatment.
Separate Section 301 measures also affect international trade. In July 2026, the U.S. Trade Representative (USTR) took action under Section 301 of the Trade Act of 1974 following investigations into 60 economies concerning their failure to impose and effectively enforce prohibitions on imports of goods produced with forced labor. The USTR's July 2026 notice sets out the actions, country-specific duty treatment, and product exclusions. These duties arise from the forced-labor investigations and should be distinguished from the targeted Section 232 semiconductor measure. The relevant statutory authority is set out in Section 301 of the Trade Act of 1974.
A Commerce and USTR notice issued in May 2026 also clarified the treatment of existing trade measures following the Supreme Court's ruling concerning tariffs imposed under the International Emergency Economic Powers Act (IEEPA). The notice explains that the ruling addressed the President's authority to impose additional duties under IEEPA. It did not affect other duties imposed under Section 232 or Section 301.
Will Broader Chip Tariffs Follow?
The United States has also considered further action on semiconductor imports under Section 232. The January 2026 proclamation directed the Secretary of Commerce and the U.S. Trade Representative to pursue negotiations with foreign jurisdictions concerning semiconductor imports, manufacturing equipment, and derivative products. It also required a review of imports and an update concerning chips used in U.S. data centers.
The proclamation established a targeted duty rather than implementing a general tariff on all imported semiconductors. It also provided for continued monitoring and consideration of further action if circumstances indicated that additional measures were necessary.
Through November 6, 2026, a review of the relevant Federal Register publications identified no separate notice implementing broader Section 232 duties across semiconductor imports. Any wider measure should therefore be treated as a potential policy development rather than an existing tariff.
This distinction matters for foundries, chip designers, and buyers. Companies can incorporate duties already in force into sourcing and cost assessments. Proposed measures, by contrast, create uncertainty and require monitoring of official announcements, implementation details, and product coverage.
For businesses with international production networks, the eventual scope of any additional duties would be particularly important. The treatment of imported chips, manufacturing equipment, and derivative products could affect procurement decisions and the economics of operating across multiple jurisdictions.
What Do Section 301 Tariffs on Chinese Semiconductors Cover?
A separate U.S. trade action specifically addresses China's practices concerning its semiconductor industry. The measure maintains an existing duty on specified Chinese semiconductor products while establishing a later increase under Section 301 of the Trade Act of 1974.
The USTR's December 2025 notice introduced an additional tariff rate of 0% on covered Chinese semiconductor products, effective December 23, 2025. The notice schedules an increase for June 23, 2027, with the applicable rate to be announced at least 30 days before that date. The additional rate is applied on top of the existing 50% Section 301 duty.
The action covers specified tariff classifications, including integrated circuits and semiconductor wafers. Importers should consult the classifications in the official notice to determine whether a particular product falls within its scope. Broad product descriptions alone may be insufficient to establish the applicable tariff treatment.
The measure is relevant to companies that source semiconductor products or fabrication inputs from China. Its implications may extend to procurement planning, supplier selection, customs procedures, and the evaluation of alternative sourcing arrangements.
Semiconductor fabrication also depends on specialized equipment used in wafer processing, deposition, etching, inspection, and other production stages. Companies involved in semiconductor manufacturing equipment form an important part of this supply chain. Changes in trade treatment can therefore influence purchasing decisions beyond finished integrated circuits.
For foundry operators, the practical implications depend on product classification, country of origin, and the specific trade measure involved. Businesses with cross-border production networks need to assess these factors at the product level and monitor changes to applicable tariff schedules.
Is Trade Policy Pulling Foundry Capacity Into the United States?
The U.S. government is seeking to strengthen domestic semiconductor fabrication through incentives, investment commitments, and trade measures. This effort reflects the importance of reliable chip supplies for computing, communications, automotive systems, industrial equipment, and national security.
The United States accounted for 10% of global semiconductor wafer fabrication capacity in 2024, according to the Federal Register's August 2026 publication. Expanding domestic production is therefore a central policy objective, particularly as advanced computing requirements increase and companies review the geographic distribution of their supply chains.
In January 2026, the American Institute in Taiwan (AIT) and the Taipei Economic and Cultural Representative Office in the United States (TECRO) signed a memorandum of understanding (MOU) outlining investment commitments by Taiwanese semiconductor and technology companies. Under the arrangement, these companies committed at least USD 250 billion in new direct investments in the United States. Taiwan also committed to providing at least USD 250 billion in credit guarantees to facilitate additional investments by Taiwanese enterprises.
The U.S. Department of Commerce's fact sheet on the Taiwan investment arrangement describes the intended expansion of semiconductor, energy, and AI production and innovation capacity. These commitments are intended to support the development of domestic production capabilities and strengthen the U.S. technology supply chain.
The MOU should be distinguished from the separate Agreement on Reciprocal Trade (ART). The Commerce and USTR notice published in May 2026 confirms that the agencies were implementing the MOU but were not implementing the ART at that time because it had not yet entered into force. The two arrangements therefore had different implementation statuses as of the notice date.
TSMC's Arizona operations provide a prominent example of domestic fabrication investment. The company's first Arizona facility entered high-volume production at the end of 2024. Construction of its second facility was ongoing, while work on a third facility began in 2025. TSMC also reported access to up to USD 6.6 billion in direct funding under the CHIPS Act, as described in its 2025 Form 20-F filed with the U.S. Securities and Exchange Commission.
Domestic fabrication also depends on capabilities beyond wafer production. Advanced packaging connects and integrates semiconductor components to support performance, power efficiency, and compact device designs. Investment in 3D semiconductor packaging is therefore relevant to developing a more complete domestic supply chain.
These developments demonstrate how investment incentives and trade policy can influence the location of future production. However, building fabrication capacity requires substantial capital, specialized equipment, skilled labor, and time. Announced commitments and construction activity should therefore be distinguished from completed facilities and operational production capacity.
What Else Drives Foundry Demand Besides Tariffs?
Trade policy is one factor influencing fabrication decisions. Demand from AI computing, automotive electronics, and other technology applications also shapes the types of chips that foundries produce and the processes they prioritize.
AI workloads require processors capable of handling complex calculations at high speed. This demand supports the use of advanced nodes, which offer higher transistor density and improved power efficiency. Chip designers developing AI accelerators and related processors therefore depend on foundries that can provide suitable process technology, production quality, and capacity.
Automotive electronics represent another important source of demand. Modern vehicles use semiconductors for power management, driver assistance, connectivity, infotainment, battery management, and other functions. As vehicle platforms incorporate more electronic systems, manufacturers require a dependable supply of chips that meet automotive performance and reliability requirements.
The International Energy Agency's Global EV Outlook 2026 reported that electric car sales increased by 20% to exceed 20 million units in 2025, representing approximately one-quarter of new car sales worldwide. However, global electric car sales declined by 8% year over year in the first quarter of 2026, reaching approximately 3.9 million units. The IEA attributed much of the decline to lower sales in China and the United States following policy changes.
This short-term decline illustrates an important limitation. Electrification supports long-term demand for automotive semiconductors, but vehicle sales and policy conditions can fluctuate. Foundries serving automotive customers must therefore account for changing production schedules, qualification requirements, and demand cycles alongside longer-term technology trends.
Conclusion
U.S. chip policy is influencing semiconductor foundry decisions through targeted duties, separate trade actions, and incentives for domestic production. The advanced computing chip measure has a defined product scope and exemptions, while the Section 301 action concerning China's semiconductor industry includes a scheduled future increase. Separate forced-labor-related trade measures should be assessed according to their own country coverage and product exclusions.
At the same time, investment commitments and fabrication projects are supporting efforts to expand U.S. production. The outcome will depend on how quickly announced investments translate into operational capacity, how trade measures are implemented, and how demand develops across AI computing and automotive applications.
For foundries, chip designers, and suppliers, the key considerations include production location, access to advanced processes, tariff classification, and the reliability of cross-border supply arrangements. Companies will also need to monitor the scheduled Section 301 change concerning Chinese semiconductors and any future decisions on broader import duties.
Explore the semiconductor foundry market with Kings Research for further analysis of technology developments, regional activity, competitive positioning, and the factors influencing long-term industry growth.
Frequently Asked Questions
How do U.S. chip tariffs affect semiconductor foundries?
Tariffs can influence sourcing costs, customer purchasing decisions, and the location of fabrication activity. Their effect depends on product classification, country of origin, destination, and applicable exemptions. Foundries and chip buyers should assess each relevant trade measure against their own supply chains.
What is the difference between a pure-play foundry and an IDM?
A pure-play foundry manufactures chips for external customers using their designs. An integrated device manufacturer manages both chip design and production, although it may also outsource some fabrication. Fabless companies generally rely on external foundries because they focus on design rather than operating fabrication plants.
Which advanced computing chips are covered by the targeted U.S. duty?
The targeted duty includes certain advanced computing chips, including NVIDIA H200 and AMD MI325X products. The measure also specifies exemptions for qualifying uses, such as certain U.S. data center, consumer, industrial, and research applications. Importers should review the applicable product and use requirements.
What is the status of broader U.S. semiconductor tariffs?
The United States has considered further semiconductor duties under Section 232. The January 2026 proclamation established a targeted duty and directed continued monitoring and review. Through November 6, 2026, the reviewed Federal Register notices showed no separate implementation notice for broader semiconductor duties.
What does the U.S. investment arrangement with Taiwanese companies involve?
The January 2026 memorandum of understanding outlines new direct investment by Taiwanese semiconductor and technology companies in the United States, alongside credit guarantees intended to support eligible projects. The MOU and the separate Agreement on Reciprocal Trade have distinct implementation statuses.
Why are advanced nodes important for AI chip production?
Advanced nodes allow manufacturers to produce chips with higher transistor density, faster processing, and lower power consumption. These characteristics are important for AI accelerators and other processors used in demanding computing workloads. Foundries with advanced process capabilities are therefore important partners for chip designers developing these products.


