The United States is moving to tighten its artificial intelligence policies toward China, a shift that comes as the AI company Anthropic publicly calls on Washington to extend its technological lead. The dual developments signal a hardening of the competitive landscape between the two nations over advanced computing.
New restrictions on the horizon
According to sources familiar with the matter, the Biden administration is preparing additional curbs on the export of AI-related technology to China. The measures are expected to target advanced chips, semiconductor manufacturing equipment, and possibly cloud-computing services that could be used to train powerful AI models. These would build on earlier rules issued in October 2022 and updated in 2023 that restricted sales of Nvidia’s A100 and H100 chips to China.
The new policies aim to close loopholes that Chinese firms have used to obtain restricted hardware through third countries or via cloud access. Officials argue that without tighter controls, the US risks losing its edge in a technology that could reshape economies and militaries.
Anthropic’s warning
Anthropic, the AI safety company behind the Claude model, has weighed in with a clear message: the US must do more to maintain its advantage. In a recent policy paper, the company argued that the window for the US to lead in AI is narrowing and that proactive government action is needed to ensure American companies stay ahead of Chinese competitors.
The company did not endorse specific export controls but stressed the importance of investing in domestic AI research, infrastructure, and talent. Anthropic’s call comes as other tech firms have expressed mixed views on the pace and scope of restrictions.
Industry and policy reactions
The tightening has drawn a split response. Some lawmakers and national security officials support aggressive measures, warning that China is already using AI for surveillance and military applications. Others in the tech industry worry that overly broad rules could harm US companies and push allies toward Chinese suppliers.
Anthropic’s intervention adds a notable voice from within the AI sector itself — one that prioritizes safety but also recognizes the strategic stakes. The company has previously advocated for responsible AI development and has been involved in White House voluntary commitments on AI safety.
The exact scope of the new rules remains unclear, but they are expected in the coming weeks. Officials are reportedly weighing how to balance national security concerns with the need to keep American AI firms competitive globally.
What’s next
The Commerce Department is expected to release a proposed rule that would expand the list of controlled technologies and tighten licensing requirements for exports to China. Public comment periods and interagency reviews will follow. Meanwhile, Anthropic and other AI labs are likely to continue lobbying for policies that they believe will secure US leadership without stifling innovation.
One unresolved question is how China will respond. Beijing has already retaliated against earlier chip curbs with export controls on gallium and germanium. Further US restrictions could escalate the tech war, with consequences for global supply chains and the pace of AI development worldwide.
