Breaking News

Up to 200,000 people estimated to travel to Vermont for total solar eclipse How fast will April’s total solar eclipse travel? The UN Security Council demands a ceasefire in Gaza during Ramadan Mexico in the emerging world order Pennsylvania State Guard Organizes Lithuanian Foreign Minister US Abstention from UN Security Council Resolution on Gaza – US State Department USA beats Mexico 2-0 thanks to goals from Adams and Reyna to win 3rd consecutive CONCACAF Nations League Mexico x United States | Highlights Meaning | The Case for American Intervention in Haiti Julian Assange to hear results of key US extradition ruling

In October, the Biden administration announced restrictions on the export of semiconductors from China. Denying access to the chips is necessary, the administration said, to protect against Chinese weapons and protect US national security. The new policy is wrong, however, and will harm American security instead of protecting it. By cutting off China’s access to advanced chips today, the United States is giving up its long-term ability to develop artificial intelligence in China and China’s acceleration toward chip independence. The recent U.S. export controls are the latest step in “bonding” the technology relationship between the U.S. and China, yet the disconnect is not enough to secure U.S. competitive interests in the long term. The best approach would be to keep China dependent on American technology, which would give the United States the ability to deny China access to critical technology when necessary.

The latest U.S. export controls are a major escalation of technology competition between the U.S. and China. In September, National Security Adviser Jake Sullivan announced a shift from the US goal of being “just two generations ahead of China” in key technologies to maintain “as big a lead as possible.” The Biden administration’s new semiconductor export controls have been implemented in principle, with the aim of stopping China’s progress in its tracks. But important elements of politics are likely to fall behind.

Semiconductors, or computer chips, are critical to today’s global economy, and trends in machine learning suggest they are poised to become an increasingly important strategic resource in the future. The recent recession has highlighted the importance of semiconductors, and in August Congress approved a $52 billion stimulus package for US semiconductor manufacturing. Yet the chips are more important than most policy makers or CEOs. Computer hardware is one of four key battlegrounds, along with data, talent, and institutions, that will determine which countries will lead the AI-driven future.

In October, the Biden administration announced restrictions on the export of semiconductors from China. Denying access to the chips is necessary, the administration said, to protect against Chinese weapons and protect US national security. The new policy is wrong, however, and will harm American security instead of protecting it. By cutting off China’s access to advanced chips today, the United States is giving up its long-term ability to develop artificial intelligence in China and China’s acceleration toward chip independence. The recent U.S. export controls are the latest step in “bonding” the technology relationship between the U.S. and China, yet the disconnect is not enough to secure U.S. competitive interests in the long term. The best approach would be to keep China dependent on American technology, which would give the United States the ability to deny China access to critical technology when necessary.

The latest U.S. export controls are a major escalation of technology competition between the U.S. and China. In September, National Security Adviser Jake Sullivan announced a shift from the US goal of being “just two generations ahead of China” in key technologies to maintain “as big a lead as possible.” The Biden administration’s new semiconductor export controls have been implemented in principle, with the aim of stopping China’s progress in its tracks. But important elements of politics are likely to fall behind.

Semiconductors, or computer chips, are critical to today’s global economy, and trends in machine learning suggest they are poised to become an increasingly important strategic resource in the future. The recent recession has highlighted the importance of semiconductors, and in August Congress approved a $52 billion stimulus package for US semiconductor manufacturing. Yet the chips are more important than most policy makers or CEOs. Computer hardware is one of four key battlegrounds, along with data, talent, and institutions, that will determine which countries will lead the AI-driven future.

China is a global AI center with high-level companies such as Baidu, Alibaba, Tencent, SenseTime, and iFlyTek, yet China faces a huge gap in AI equipment, importing more than $400 billion in chips annually. China is working hard to reduce its dependence on foreigners, which it sees as a strategic vulnerability. More than 90 percent of the chips used in China are made abroad or by foreign companies that produce in China. China has been building domestic facilities for the production of chips, or clothing. Before 2020, China accounted for the fastest growing share of global semiconductor production.

The Biden administration’s new export administration is restricting sales of U.S. semiconductor manufacturing equipment to China, slowing chip development in China. The United States, the Netherlands, and Japan control more than 90 percent of the semiconductor manufacturing equipment market. If the Netherlands and Japan adopt similar controls, the three countries could effectively block China from the technology needed to build a high-quality textile. The Biden administration is also reportedly considering restrictions on US investment in Chinese AI and semiconductor companies. These controls aim to prevent China from achieving chip independence, while keeping China dependent on foreign supplies. Yet as part of the same move, the Biden administration also restricted China’s access to advanced foreign chips, including graphics-processing units (GPUs) used in artificial intelligence, data centers, and advanced computing. . In doing so, the Biden administration undercuts the United States’ major influence on China’s AI development.

The United States has a major advantage in the emerging geographic competition for AI equipment: The United States and allied companies dominate key points in the global supply chain, especially the equipment and software needed to produce chips. Even chips made outside the US in Taiwan or South Korea rely on US equipment for production.

In 2020, the Trump administration used this power to block the Chinese company Huawei’s bid for global capacity in 5G networks by allowing access to advanced chips. Huawei relied on chips manufactured in Taiwan, and while these were not made in American factories, they were made on American equipment. The latest U.S. regulations expand these restrictions, using the same export rules to cut off the supply of advanced AI chips to China as a whole, even when those chips are manufactured outside of the United States. This move slows China’s AI development today but undermines the US’s long-term strategic position.

The ability to deny China access to advanced chips is a powerful advantage whose value is growing exponentially. The past decade has seen an explosion in artificial intelligence. AI models are trained on data using computational hardware, or computing, often in the form of GPUs or specialized AI chips. Since 2010, the amount of computation used to train cutting-edge AI models has increased “by a factor of 10 billion.” Computing usage doubles every six months, compared to the 20-month doubling time under Moore’s Law. (The computation of the largest AI models doubles every 10 months.)

Today’s modern AI models are trained on thousands of GPUs running for weeks at a time. Their demanding mathematical requirements make them available only to samples with deep pockets. Three of the leading AI research labs, OpenAI, DeepMind, and Google Research, are supported by Microsoft or Alphabet. Microsoft is reportedly planning to invest $10 billion in OpenAI, the developer of ChatGPT. Meta is building an AI Research SuperCluster with over 6,000 GPUs, with plans to eventually grow to 16,000. Academics have rejected this trend, which devalues ​​research into large-scale AI models. The U.S. government has been working to expand access to computing resources for AI researchers, a key step in maintaining U.S. competitiveness. Computing is increasingly a strategic resource, the availability of which determines a company-or country’s ability to take advantage of the advanced capabilities of AI.

If China relies on foreign chips, the United States can control China’s access to increasingly strategic resources. The development of titles such as ChatGPT is only the tip of the iceberg of AI development. AI research continues to advance at breakneck speed, and the AI ​​capabilities of tomorrow will likely be more valuable than those of today. Using America’s current power to completely cut off China’s access to high-end chips is like selling stocks that double every six months. Using export controls today could make them ineffective in the future as companies adapt their supply chains to circumvent US restrictions.

The US government has effectively created a large market for an independent US semiconductor supply chain. The Chinese government has long pursued wasteful industrial policies to boost domestic semiconductor production. After US export controls, the Chinese government pledged another 140 billion in spending. US regulators greatly appreciate the efforts of the Chinese government in creating incentives for the private sector as well as, organizing the international market to accelerate China’s independence. Foreign companies have strong incentives to design US components outside their supply chains over time to bypass US restrictions and sell in the Chinese market. Changing US suppliers will take time, but all the incentives are now in place to do so.

Even if Japan and the Netherlands join the United States in banning equipment from China, foreign companies can build new clothing completely free of American equipment outside of China to serve the Chinese market. Controlling access to foreign chips has also turned China into $400 million of domestic purchasing power, boosting domestic chip production. Chinese business data and cloud companies may choose to turn to second-tier domestic chip suppliers, instead of continuing to rely on foreign chips for their uncertain supply. These market incentives can mobilize the private sector, inside and outside of China, to finally achieve the independence that the Chinese government has been seeking for so long.

More targeted controls on military applications and human rights abuses while allowing commercial use would be a positive step and could keep China dependent on foreign chips produced in US equipment. Targeted controls are more complicated, as the US is forced into a game of whack-a-mole while the Chinese military tries to circumvent US restrictions by using other buyers.

However, the United States has created a powerful tool to control China’s military integration. As part of the October 2022 controls, the Department of Commerce established that companies that do not cooperate with end-use verification will be placed on the Unverified List and, after 60 days, will be transferred to the Institutional List if they are still not in compliance. The US government has already successfully used this tool in December to verify and certify 27 Chinese companies that have demonstrated compliance with end-use restrictions while promoting others, including Chine YMTC, in existing List of Institutions. The United States has the tools it needs to enforce compliance with targeted sanctions. A more targeted system that allows the sale of high-end chips in the Chinese business sector can control the demand for chips made without US technology, delaying Chine’s independence from China.

The United States should seek to foster China’s strategic dependence on American technology. In the technological competition between the United States and China, the exploitation that the United States has used of Huawei’s clones is an invaluable strategic advantage. But the Biden administration’s new restrictions are undermining America’s long-term position. Because U.S. export controls will only have a temporary effect as global markets adjust over time, the U.S. is better off keeping these stocks for now. Artificial intelligence capabilities are advancing rapidly, and computer hardware is becoming increasingly valuable for AI capabilities.

The deep learning revolution is just over 10 years old, and the second decade looks to be even more dramatic than the first. The United States will be in a strong position in the coming changes if it retains the ability to deny China powerful AI capabilities, if necessary. Using today’s thin export controls could still deny China’s military access to advanced chips while keeping Chinese businesses dependent on foreign supplies. China’s reliance on American technology is a stronger strategy to diversify and will give the United States more control over China’s access to advanced technology in the long run.

Leave a Reply

Your email address will not be published. Required fields are marked *