In a move to safeguard its dominant position in the artificial intelligence (AI) sector, the US government has announced stringent new regulations aimed at restricting the global export of AI chips and related technologies. Unveiled recently, these regulations are designed to limit the number of AI chips that can be sold to most countries while providing unfettered access to the United States’ closest allies. At the same time, the restrictions will continue to block exports to countries such as China, Russia, Iran, and North Korea.
Commerce Secretary Gina Raimondo stated that the regulations are crucial to maintaining the United States’ leadership in AI development, both in terms of AI technology and AI chip design. “The US leads AI now – both AI development and AI chip design, and it’s critical that we keep it that way,” Raimondo remarked. The regulations build upon the efforts of the Biden administration to curtail China’s access to advanced AI chips that could potentially enhance its military capabilities.
These new rules, which mark the culmination of a four-year effort, are aimed at ensuring that the US retains control over the development of AI technology globally. While the regulations primarily target China, they also affect other nations that could potentially divert these advanced technologies to China. With the incoming Trump administration set to take office soon, it remains unclear how the new rules will be enforced. Nevertheless, both administrations share similar concerns about the competitive threat posed by China in the field of AI.
Set to take effect 120 days after publication, the regulations aim to limit the global access to advanced graphics processing units (GPUs), which are essential for powering data centres used to train AI models. These chips are predominantly manufactured by companies such as Nvidia, based in Santa Clara, California, and Advanced Micro Devices (AMD). Shares of Nvidia and AMD saw a slight decline of 5 per cent and 1 per cent, respectively, following the announcement.
Cloud service providers such as Microsoft, Google, and Amazon are expected to be given global authorisations to build data centres in countries that may not have access to sufficient AI chips due to the new US-imposed quotas. Once authorised, these companies would no longer need export licences to ship AI chips to such countries. However, authorised providers will be subject to strict security requirements, reporting obligations, and guidelines to ensure respect for human rights in the countries where they operate.
Previously, the Biden administration had implemented sweeping restrictions on China’s access to advanced AI chips and the equipment needed to manufacture them, with annual updates to tighten controls. The new regulations build on this approach, aiming to close any remaining loopholes and further limit China’s access to cutting-edge AI technology. However, industry leaders have expressed concerns about the potential overreach of these regulations. Nvidia, in particular, has criticised the rules as a “sweeping overreach,” arguing that the restrictions could impact technologies already available in mainstream gaming PCs and consumer hardware. Similarly, Oracle, a leading data centre provider, warned that the regulations could push the majority of the global AI and GPU market into the hands of Chinese competitors.
The new rules include provisions that divide the world into three distinct tiers. Countries like Japan, the UK, South Korea, and the Netherlands will effectively be exempt from the restrictions, while around 120 other countries, including Singapore, Israel, and the UAE, will face caps on the number of AI chips they can import. Nations under arms embargo, such as Russia, China, and Iran, will be entirely barred from receiving AI technologies.
Moreover, US-based cloud providers like Amazon Web Services and Microsoft will be allowed to deploy no more than 50 per cent of their AI computing power outside the United States, with no more than 25 per cent of that power deployed in Tier 1 countries and a maximum of 7 per cent in any single non-Tier 1 country.
More Read
As the world watches these developments unfold, national security experts like Meghan Harris have suggested that the effectiveness of the regulations will depend on how the new administration handles AI-related policies in the coming years. “They are well aware that ensuring a dominant domestic industry is a core element of competition with China,” Harris said.
In response, China’s Ministry of Commerce stated that it would take necessary measures to protect its “legitimate rights and interests,” signalling a potential escalation in the global race for AI dominance.
AI holds the potential to revolutionise sectors such as healthcare, education, and food security, among others. However, it also has the capability to be used for more nefarious purposes, including the development of biological weapons, cyberattacks, and surveillance. US National Security Adviser Jake Sullivan emphasised the importance of preparing for rapid advancements in AI, stating that these developments could have transformative impacts on both the economy and national security in the coming years.
As AI technology continues to evolve, the US government’s tighter controls over its global distribution are set to play a significant role in shaping the future of AI development and international relations.
