In a move that combines strategic firmness with economic pragmatism, President Donald Trump has intensified restrictions on semiconductor production in China, directly impacting giants like Intel, Samsung, and SK Hynix. Simultaneously, he has chosen to postpone the imposition of a 25% tariff on hardware imports from that country, offering a temporary relief to the U.S. market. This dual decision, announced by the Department of Commerce and the U.S. Trade Representative’s Office (USTR), reflects the Trump administration’s determination to limit China’s technological advances in critical areas such as artificial intelligence, without immediately harming American consumers’ wallets. It’s a delicate balancing act that could reshape the global chip industry landscape, with repercussions extending from Seoul to Silicon Valley.
The announcement has caused a stir in the tech sector, where analysts and executives see these measures as a further step in the trade war between Washington and Beijing. Trump, true to his direct and confrontational style, hasn’t hesitated to criticize his predecessor Joe Biden’s policies, accusing them of leaving “loopholes” that favored foreign competitors. “We are closing the doors that others left open,” could summarize the message from the White House, though not in those exact words. The reality is that, with these actions, the U.S. aims to reaffirm its dominance in the semiconductor supply chain—an essential element for national security and emerging technologies.
The Blow to Production in China: End of Exemptions for Samsung, SK Hynix, and Intel
The Department of Commerce, through its Bureau of Industry and Security (BIS), has been the first to deliver the blow. In a concise but firm statement, BIS has eliminated an exemption allowing certain foreign companies to export semiconductor manufacturing equipment and technology to China without special licenses. This “Biden-era legal loophole,” as the agency has called it, primarily benefited a handful of companies, including Samsung, SK Hynix, and Intel. Now, these firms must seek licenses for any U.S. technology export, placing them on equal footing with their American competitors.
For Samsung and SK Hynix, two pillars of the South Korean memory industry, the impact is immediate and significant. Both companies have 120 days before existing exemptions expire entirely. After that period, their factories in China can continue operations but won’t be able to expand or modernize with U.S.-origin technology. This limits them to conventional chip production, excluding advanced products like high-bandwidth memory (HBM), which is crucial for AI applications and high-performance computing.
Samsung, for example, produces up to 40% of its NAND flash in its Xi’an plant, a strategic facility key to its global market position. Similarly, SK Hynix concentrates around 40% of its DRAM production in Wuxi and 20% of its NAND in Dalian. Industry sources familiar with operations indicate that license delays could extend from three to nine months, as each application must undergo Washington’s scrutiny. This will not only halt technological upgrades but could force these companies to reconsider their production strategies. Some experts suggest that, in the medium term, parts of these operations might relocate back to South Korea, though this would entail high costs and supply chain disruptions.
Intel is also affected, albeit to a lesser extent in terms of production volume in China. The American company has invested in manufacturing facilities in China, but the new restrictions limit its ability to import advanced equipment. This move aligns with Trump’s broader policy of restricting sensitive technology flow to China, aiming to curb Beijing’s progress in cutting-edge semiconductors. In a context where AI increasingly relies on components like HBM, these limitations could slow the development of Chinese GPUs for AI, which had begun gaining ground due to their low cost and high capacity.
The rationale behind this decision is clear: the U.S. seeks to undermine China’s position in advanced chip production—a sector Trump considers vital for technological supremacy. During his first term, he imposed similar restrictions, now tightened, criticizing Biden’s policies for allowing exceptions that he claims undermine national security. “It’s time to put America first,” has been a recurring slogan of his speeches, and these measures exemplify that stance.
Extension of Tariffs: A Breathing Space for the U.S. Market
Alongside these restrictions, the USTR has announced the extension of tariff exemptions under Section 301, covering imports of hardware and components from China. An initial 25% tariff was slated for these products, but the extension—its second—delays its application until November 29, 2025. This affects 178 specific exclusions, which now have an additional 90-day window, preventing immediate price hikes for consumers and businesses in the U.S.
This decision is described by some analysts as a “kick forward,” allowing Washington to buy time while strengthening its own manufacturing capacity. Trump has promoted initiatives like the CHIPS Act, aimed at building more American fabs to reduce reliance on Asian suppliers. By postponing tariffs, the risk of price increases in electronics—ranging from computers to server parts—is mitigated, especially as inflation remains a concern for voters.
The balancing act is evident: while tightening controls over Chinese production, the U.S. protects its domestic market from economic shocks. For instance, products like NVIDIA graphics cards, crucial for AI, could have seen significant price hikes if tariffs had been enacted immediately. Trump has previously indicated interest in redirecting a portion of AI hardware sales—such as 15% of NVIDIA’s sales—back to the government, though this has not materialized. The extension provides time for these policies to solidify while keeping pressure on China without escalating costs for Americans.
Global Implications and Expected Responses
This dual announcement isn’t isolated; it fits into a broader strategy that Trump has revived since returning to the White House. His first term saw hefty tariffs on China, Mexico, and Canada, creating trade tensions affecting multiple sectors. Now, with renewed focus on semiconductors, the U.S. seeks not only to curb China but also to attract investments domestically. Companies like TSMC and Samsung have already announced plans to expand operations in the U.S., driven by government subsidies.
However, the strategy is risky. China, which has heavily invested in its chip industry via initiatives like Made in China 2025, could retaliate. Beijing has condemned these restrictions as “discriminatory” and may accelerate efforts to develop indigenous technology, reducing dependence on foreign suppliers. For Samsung and SK Hynix, both with strong government ties in South Korea, the situation raises geopolitical dilemmas: balancing Washington’s demands with their strategic interests in the Chinese market, which accounts for a significant share of their revenue.
Industry analysts warn these measures could cause disruptions in the global supply chain, potentially leading to higher memory prices and delays in product delivery. In a world where AI chip demand is exploding—driven by companies like OpenAI and Google—limiting Chinese production could benefit competitors in Taiwan or the U.S., but also cause temporary shortages. With his typical style, Trump seems to favor a hardening approach that strengthens U.S. positions in the long term, albeit with inherent risks.
In summary, these decisions continue the saga of sino-American rivalry. By shutting down Chinese factory modernization for Intel, Samsung, and SK Hynix, and delaying tariffs, Trump sends a clear message: the U.S. controls the future of semiconductors but does so without sacrificing immediate economic stability. How Beijing responds—and whether these measures accelerate a broader technological decoupling—are questions that will shape the geopolitical landscape in the coming months.
Frequently Asked Questions
- What are the immediate consequences for Samsung and SK Hynix in China?
They have 120 days to adapt; after that, they cannot import U.S. technology to expand or upgrade their factories. They will be limited to conventional chips, affecting products like AI HBM memory, and may face delays of three to nine months, possibly relocating some operations back to South Korea. - Why are the 25% tariffs on Chinese hardware being postponed?
The extension until November 29, 2025, aims to prevent price hikes in the U.S. while strengthening domestic manufacturing. It covers 178 exclusions under Section 301, giving time to reduce dependence on Chinese imports without impacting consumers and businesses. - How does this affect China’s AI development?
Restrictions limit access to advanced components like HBM, vital for AI GPUs. This could slow China’s progress in low-cost AI hardware, benefiting the U.S. and allies, but might also prompt Beijing to accelerate its indigenous tech efforts. - What role does Trump’s policy play compared to Biden’s?
Trump criticizes Biden’s “loopholes” that allowed exemptions, tightening restrictions to prioritize national security. His approach seeks to shift manufacturing to the U.S. via laws like the CHIPS Act, contrasting with more permissive previous policies.
via: elchapuzasinformatico and bloomberg