The Biden administration has introduced new export restrictions on U.S.-made computer chips that power artificial intelligence (AI) systems, marking a significant effort to limit China’s access to cutting-edge technology. These measures, announced just a week before President-elect Donald Trump’s inauguration, are seen as the culmination of years of efforts to hinder China’s advancements in military and industrial leadership.
The move is expected to escalate tensions between Washington and Beijing, which have already been strained ahead of the upcoming presidential transition. The restrictions have drawn sharp criticism from major U.S. tech firms, including Nvidia and Oracle, who argue that the measures could undermine American competitiveness in the global tech race.
U.S. Commerce Secretary Gina Raimondo emphasized that the new rules are designed to “protect the most advanced AI technologies and ensure they do not fall into the hands of foreign adversaries.” However, she also noted that the regulations aim to facilitate broader collaboration and sharing of benefits with allied nations. The updated export framework categorizes countries into three tiers, with no new restrictions imposed on partners like Australia, Japan, South Korea, and Taiwan.
China and Russia, placed in the second tier, will face additional bans on purchasing the most powerful “closed” AI models, which are not publicly available. The third tier, encompassing most other countries, will soon face limits on the amount of computing power they can acquire, though they may apply for additional quotas based on specific security needs. Analysts suggest these changes are intended to prevent China from accessing AI chips through third-party countries, particularly in the Middle East.
The restrictions come amid surging global demand for AI chips produced by companies like Nvidia, AMD, and Intel. With just days left in Biden’s term, the rules will enter a 120-day comment period but are expected to take effect before its conclusion. Raimondo expressed hope that the incoming administration would fully utilize this period to gather input from experts, industry stakeholders, and partner nations, potentially leading to further adjustments.
Senior Biden administration officials, speaking on background, acknowledged ongoing discussions with the incoming Trump administration but provided limited details. One official noted, “We are in a critical moment, especially in relation to China. Every minute counts.”
The latest measures follow a series of export controls imposed by the previous administration, including bans on the sale of semiconductor manufacturing equipment and restrictions on Chinese companies’ access to U.S. technology. Since October 2022, the Biden administration has intensified efforts to target Beijing, as Chinese leader Xi Jinping has prioritized self-reliance in technology to position China as a global tech superpower.
However, the new restrictions have faced backlash from the tech industry. Nvidia, Oracle, and influential semiconductor industry groups have criticized the Biden administration for bureaucratic overreach, arguing that the rules could harm U.S. competitiveness. In a blog post, Nvidia’s Vice President of Government Affairs, Ned Finkle, warned that the regulations threaten global innovation and economic growth, stating, “The Biden administration is attempting to restrict access to mainstream computing applications, risking derailing global progress.”
Oracle’s Executive Vice President Ken Glueck echoed these concerns, calling the rules “an extreme regulatory overreach that prioritizes bureaucratic goals over American interests and the security of our partners and allies.” The Semiconductor Industry Association also expressed alarm over the “extraordinary scope and complexity” of the regulations, warning they could significantly diminish U.S. leadership in semiconductor technology and advanced AI systems.
As the global tech landscape becomes increasingly competitive, these restrictions highlight the delicate balance between national security and economic interests, with far-reaching implications for the future of AI and semiconductor innovation.
