On Monday, the US government announced plans to further restrict the export of artificial intelligence (AI) chips and technology. The move aims to divide the world into strategic zones, ensuring that advanced computing power remains within the United States and its allies. This new set of restrictions will focus on limiting China’s access to these critical technologies.
By implementing these restrictions, the US seeks to maintain its technological edge and prevent rival nations, particularly China, from gaining access to powerful AI capabilities. The government believes that keeping these technologies away from certain regions will safeguard national security interests and preserve the balance of power in global technological competition.
The restrictions will specifically target high-performance AI chips, which are essential for developing cutting-edge AI systems. These chips are vital for training large language models and running complex algorithms, making them a key asset in both commercial and military applications. The US intends to keep this technology confined to nations it considers trustworthy.
In addition to limiting exports to China, the US is exploring other avenues to prevent unauthorized access to sensitive AI technology. These measures could involve diplomatic efforts, collaborations with international partners, and stricter enforcement of existing regulations. The goal is to isolate China’s technological development while strengthening partnerships with allied nations.
This strategy reflects a growing concern over China’s advancements in AI and its potential to surpass Western nations in AI capabilities. By restricting access to these powerful technologies, the US aims to maintain its leadership in the AI race and prevent any challenges to its dominance in global technology innovation.
The new regulations will impose limits on the number of AI chips that can be exported to most countries, while granting unlimited access to US AI technology for America’s closest allies. At the same time, the rules will maintain a ban on exports to China, Russia, Iran, and North Korea.
Introduced during the final days of President Joe Biden’s administration, the comprehensive new rules extend beyond China, aiming to preserve the United States’ dominance in AI by regulating its distribution globally. These measures reflect a strategic effort to control the spread of AI technology and ensure US leadership in the field.
“The US leads in AI — both in development and chip design, and it’s crucial that we maintain this leadership,” said Commerce Secretary Gina Raimondo. Her statement highlights the importance of safeguarding the US’s technological advantage in the competitive AI sector.
The regulations represent the culmination of a four-year effort by the Biden administration to limit China’s access to advanced chips that could strengthen its military capabilities. By closing loopholes and implementing new safeguards, the US aims to reinforce its leadership in AI and prevent adversaries from accessing critical technologies.
With these new measures, the US seeks to secure its position as the global leader in AI while limiting the ability of rival nations to challenge its technological supremacy. The regulations are a response to growing concerns about the strategic use of AI and its potential implications for national security.
While it remains uncertain how President-elect Donald Trump’s incoming administration will enforce the new rules, both the Trump and Biden administrations share similar concerns about the competitive threat posed by China. The regulation will take effect 120 days after publication, providing time for the Trump administration to evaluate the policy.
New restrictions will target advanced graphics processing units (GPUs), which power the data centers essential for training AI models. The majority of these GPUs are produced by Nvidia, based in Santa Clara, California, though Advanced Micro Devices (AMD) also manufactures AI chips. Following the announcement, Nvidia’s shares dropped by about 5%, while AMD’s shares fell by approximately 1% during morning trading.
Major cloud service providers, including Microsoft, Google, and Amazon, will be able to apply for global authorizations to build data centers. Once approved, these companies will no longer need export licenses for AI chips, enabling them to establish data centers in countries that face limitations on chip imports due to US-imposed quotas.
Shares of all three companies saw a decline of about 1%.
To receive approval, authorized companies must meet strict conditions, including security requirements, regular reporting, and a demonstrated commitment to human rights. These measures ensure that the companies adhere to the regulations while maintaining their ability to operate globally.
Until now, the Biden administration had implemented broad restrictions on China’s access to advanced chips and the equipment used to produce them. These controls were updated annually, with each revision tightening restrictions and expanding efforts to include countries that might divert the technology to China.
NVIDIA EXPRESSIONS CONCERNS ABOUT ‘OVERREACH’
The new rules, which reshape the global landscape for AI chips and data centers, faced significant criticism from powerful industry voices even before their official release. Nvidia labeled the plan as “sweeping overreach” on Monday, arguing that the White House was restricting “technology that is already available in mainstream gaming PCs and consumer hardware.” Data center provider Oracle warned earlier this month that the regulations would hand “most of the global AI and GPU market to our Chinese competitors.” Notably, the restrictions do not apply to gaming chips.
The regulations impose worldwide licensing requirements on advanced chips, with certain exceptions, and establish controls over the “model weights” of the most advanced “closed-weight” AI models. Model weights, which are crucial for decision-making in machine learning, are considered among the most valuable components of an AI model.
The rules divide the world into three tiers. About 18 countries, including Japan, Britain, South Korea, and the Netherlands, will largely be exempt from the regulations. Around 120 other countries, such as Singapore, Israel, Saudi Arabia, and the United Arab Emirates, will face country-specific caps. Countries under arms embargoes, such as Russia, China, and Iran, will be completely barred from receiving the technology.
Additionally, US-based providers, including Amazon Web Services and Microsoft, likely to receive global authorizations, will be restricted in deploying AI computing power outside the United States. They can only deploy up to 50% of their total AI computing power abroad, no more than 25% outside Tier 1 countries, and no more than 7% in any single non-Tier 1 country.
Meghan Harris, a national security official from the first Trump administration, remarked, “How effective the rule ends up being in the next 10 to 15 years is now up to the incoming team.” She emphasized the importance of maintaining a dominant domestic industry to compete with China. In response, China’s Commerce Ministry stated that it would take necessary measures to protect its “legitimate rights and interests.”
AI has the potential to revolutionize sectors such as healthcare, education, and food security, but it also presents risks, including its use in developing biological and other weapons, enabling cyberattacks, and supporting surveillance and human rights violations. “The US has to be prepared for rapid increases in AI’s capability in the coming years, which could have a transformative impact on the economy and on our national security,” said US National Security Adviser Jake Sullivan.
