In May 2026, Beijing was not a city of Cold War tensions, but neither was it a place of complete reconciliation. The red carpet of the Great Hall of the People and the serene pathways of Tiananmen Park reflected a significant tension and restraint surrounding the 21st-century global order.
The summit between U.S. President Donald Trump and Chinese President Xi Jinping projected messages of stability and cooperation. However, the real focus of the talks was not tariffs but rather artificial intelligence (AI) and semiconductors. The global power dynamic has already shifted from oil, steel, and automobiles to data, computational power, and advanced semiconductors.
A notable moment during the visit was the presence of Jensen Huang, CEO of NVIDIA. After completing his schedule in Alaska, President Trump effectively brought Huang directly to Beijing. This was not merely a business leader's visit; it symbolized the U.S. beginning to manage AI semiconductors as a strategic national asset.
At one time, the center of U.S.-China tensions was the tariff war. High tariffs imposed during Trump's first term shook both Chinese manufacturing and the U.S. consumer market. The U.S. cited trade deficit reduction and the correction of China's unfair trade practices, while China retaliated with counter-tariffs. The world became accustomed to the term 'tariff war.'
However, in just a few years, global attention has shifted dramatically. The key question is no longer “who can sell cheaper goods?” but rather “who can design the future civilization?” At the heart of this is AI and semiconductors.
Semiconductors are no longer just electronic components; in the AI era, they represent national power. A nation's military strength, financial systems, cloud industries, autonomous vehicles, robotics, aerospace, and biotechnology all rely on high-performance semiconductors. Notably, NVIDIA's GPUs are now referred to as the 'oil of the AI era.'
The area where the U.S. is applying the most pressure on China is advanced AI semiconductors. The U.S. has restricted exports of top-tier AI chips like the H100 and H200 and has tightened controls on advanced semiconductor equipment and software. Additionally, companies like ASML from the Netherlands and Japanese semiconductor equipment firms have partially joined the U.S. strategy, putting significant pressure on China’s access to advanced processes.
However, China has not retreated easily. Instead, it has elevated semiconductor self-sufficiency to a national survival strategy in response to U.S. pressure. Development of AI chips centered around Huawei, the establishment of a domestic GPU ecosystem, memory independence, and the nurturing of Chinese semiconductor equipment are all progressing simultaneously. Recently, the rise of Chinese AI company DeepSeek has sent shockwaves through the global industry.
DeepSeek has achieved significant AI performance through its own optimization technology and efficient computational structure, even with limited access to U.S. advanced chips. This challenges the assumption that “without American chips, Chinese AI cannot grow.”
This situation deepens the dilemma for the U.S. While excessive pressure on China may yield short-term advantages for the U.S., it risks accelerating China's self-sufficiency in the long run. Historically, technology blockades have often led to the growth of independent ecosystems in the targeted countries.
The cautious handling of AI and semiconductor issues during the Beijing summit reflects this reality. The U.S. is wary of China's military AI ambitions while also seeking to avoid a complete loss of its companies' access to the Chinese market. For U.S. firms like NVIDIA, Apple, and Tesla, China remains one of the largest markets globally. Conversely, China understands that it is challenging to grow without completely excluding U.S. advanced technologies and the global financial system.
Thus, the atmosphere of this summit differed significantly from past tariff conflicts. Previously, both sides raised tariffs and fought over trade surpluses and deficits. Now, a much deeper competition has begun over AI hegemony and leadership in future civilization. While they may smile and shake hands on the surface, a quiet war is underway over “who will control the operating system of the AI era.”
In fact, competition for AI data centers and GPU acquisition is intensifying in the U.S. Microsoft, Google, Amazon, Meta, and even Wall Street financial firms are pouring astronomical amounts into AI infrastructure investments. The U.S. still maintains its position as the global leader in design, software, and advanced GPU technology.
Meanwhile, China is accelerating its pursuit, leveraging its vast domestic market, state-led investments, and manufacturing base. Local governments in China are pouring significant funds into AI industrial parks, and aggressive efforts are underway to secure semiconductor talent and develop domestic equipment. Numerous AI startups and semiconductor companies are already clustered in Beijing, Shanghai, and Shenzhen. The competition in AI and semiconductors is not merely an industrial rivalry; it is a contest for the future national system and global order. In this context, the tariff wars have receded into the background.
Tariffs remain important, but they are a relic of the industrial age. The core of the AI era lies in data, computational power, and semiconductor supply chains. Ultimately, the world is transitioning from an era defined by “who can produce cheaper” to one defined by “who can control more computational power and algorithms.”
The global economy is also likely to be significantly shaken by this trend. If the U.S. and China can maintain a certain level of cooperation and trade in AI and semiconductors, the global semiconductor market may stabilize. Conversely, if tensions escalate again, supply chain fragmentation and technological blockades will deepen, potentially dividing the world into U.S.-led and China-led technological spheres.
South Korea's concerns are even more profound. While South Korea is one of the world's leading memory semiconductor nations, it is also required to maintain a strategic balance between the U.S. security alliance and the Chinese market. Samsung Electronics and SK Hynix must comply with U.S. advanced semiconductor regulations but cannot afford to abandon their production bases and markets in China.
Ultimately, South Korea must rise to become a key technology nation in the AI era, not just a production base. It needs to expand its competitiveness beyond memory-centric structures to include AI semiconductor design, software, power semiconductors, and advanced packaging. At the same time, a strategic diplomatic sense that does not allow itself to be swept away by either the U.S. or China is essential.
The 2026 Beijing summit clearly demonstrated this shift. The axis of global hegemony is now moving from tariffs to AI and semiconductors. And this quiet war has already begun.
The summit between U.S. President Donald Trump and Chinese President Xi Jinping projected messages of stability and cooperation. However, the real focus of the talks was not tariffs but rather artificial intelligence (AI) and semiconductors. The global power dynamic has already shifted from oil, steel, and automobiles to data, computational power, and advanced semiconductors.
A notable moment during the visit was the presence of Jensen Huang, CEO of NVIDIA. After completing his schedule in Alaska, President Trump effectively brought Huang directly to Beijing. This was not merely a business leader's visit; it symbolized the U.S. beginning to manage AI semiconductors as a strategic national asset.
At one time, the center of U.S.-China tensions was the tariff war. High tariffs imposed during Trump's first term shook both Chinese manufacturing and the U.S. consumer market. The U.S. cited trade deficit reduction and the correction of China's unfair trade practices, while China retaliated with counter-tariffs. The world became accustomed to the term 'tariff war.'
However, in just a few years, global attention has shifted dramatically. The key question is no longer “who can sell cheaper goods?” but rather “who can design the future civilization?” At the heart of this is AI and semiconductors.
Semiconductors are no longer just electronic components; in the AI era, they represent national power. A nation's military strength, financial systems, cloud industries, autonomous vehicles, robotics, aerospace, and biotechnology all rely on high-performance semiconductors. Notably, NVIDIA's GPUs are now referred to as the 'oil of the AI era.'
The area where the U.S. is applying the most pressure on China is advanced AI semiconductors. The U.S. has restricted exports of top-tier AI chips like the H100 and H200 and has tightened controls on advanced semiconductor equipment and software. Additionally, companies like ASML from the Netherlands and Japanese semiconductor equipment firms have partially joined the U.S. strategy, putting significant pressure on China’s access to advanced processes.
However, China has not retreated easily. Instead, it has elevated semiconductor self-sufficiency to a national survival strategy in response to U.S. pressure. Development of AI chips centered around Huawei, the establishment of a domestic GPU ecosystem, memory independence, and the nurturing of Chinese semiconductor equipment are all progressing simultaneously. Recently, the rise of Chinese AI company DeepSeek has sent shockwaves through the global industry.
DeepSeek has achieved significant AI performance through its own optimization technology and efficient computational structure, even with limited access to U.S. advanced chips. This challenges the assumption that “without American chips, Chinese AI cannot grow.”
This situation deepens the dilemma for the U.S. While excessive pressure on China may yield short-term advantages for the U.S., it risks accelerating China's self-sufficiency in the long run. Historically, technology blockades have often led to the growth of independent ecosystems in the targeted countries.
The cautious handling of AI and semiconductor issues during the Beijing summit reflects this reality. The U.S. is wary of China's military AI ambitions while also seeking to avoid a complete loss of its companies' access to the Chinese market. For U.S. firms like NVIDIA, Apple, and Tesla, China remains one of the largest markets globally. Conversely, China understands that it is challenging to grow without completely excluding U.S. advanced technologies and the global financial system.
Thus, the atmosphere of this summit differed significantly from past tariff conflicts. Previously, both sides raised tariffs and fought over trade surpluses and deficits. Now, a much deeper competition has begun over AI hegemony and leadership in future civilization. While they may smile and shake hands on the surface, a quiet war is underway over “who will control the operating system of the AI era.”
In fact, competition for AI data centers and GPU acquisition is intensifying in the U.S. Microsoft, Google, Amazon, Meta, and even Wall Street financial firms are pouring astronomical amounts into AI infrastructure investments. The U.S. still maintains its position as the global leader in design, software, and advanced GPU technology.
Meanwhile, China is accelerating its pursuit, leveraging its vast domestic market, state-led investments, and manufacturing base. Local governments in China are pouring significant funds into AI industrial parks, and aggressive efforts are underway to secure semiconductor talent and develop domestic equipment. Numerous AI startups and semiconductor companies are already clustered in Beijing, Shanghai, and Shenzhen. The competition in AI and semiconductors is not merely an industrial rivalry; it is a contest for the future national system and global order. In this context, the tariff wars have receded into the background.
Tariffs remain important, but they are a relic of the industrial age. The core of the AI era lies in data, computational power, and semiconductor supply chains. Ultimately, the world is transitioning from an era defined by “who can produce cheaper” to one defined by “who can control more computational power and algorithms.”
The global economy is also likely to be significantly shaken by this trend. If the U.S. and China can maintain a certain level of cooperation and trade in AI and semiconductors, the global semiconductor market may stabilize. Conversely, if tensions escalate again, supply chain fragmentation and technological blockades will deepen, potentially dividing the world into U.S.-led and China-led technological spheres.
South Korea's concerns are even more profound. While South Korea is one of the world's leading memory semiconductor nations, it is also required to maintain a strategic balance between the U.S. security alliance and the Chinese market. Samsung Electronics and SK Hynix must comply with U.S. advanced semiconductor regulations but cannot afford to abandon their production bases and markets in China.
Ultimately, South Korea must rise to become a key technology nation in the AI era, not just a production base. It needs to expand its competitiveness beyond memory-centric structures to include AI semiconductor design, software, power semiconductors, and advanced packaging. At the same time, a strategic diplomatic sense that does not allow itself to be swept away by either the U.S. or China is essential.
The 2026 Beijing summit clearly demonstrated this shift. The axis of global hegemony is now moving from tariffs to AI and semiconductors. And this quiet war has already begun.
* This article has been translated by AI.
Copyright ⓒ Aju Press All rights reserved.