China is reportedly considering measures to restrict foreign access to its advanced artificial intelligence (AI) models, following similar actions by the United States.
On July 7, Reuters cited three sources familiar with the matter, stating that Chinese authorities have held meetings with major technology companies over the past month to discuss potential restrictions on foreign access to the country’s cutting-edge AI models. Some of these models have not yet been publicly disclosed.
Participants in the meetings included Alibaba, ByteDance, and the AI startup Z.ai. The discussions were led by the Ministry of Commerce, with representatives from the National Development and Reform Commission also in attendance.
According to the sources, the meetings covered restrictions on both closed and more open models of advanced AI. There was also discussion about treating the theft or leakage of proprietary AI technology as a violation of China’s strict national security laws.
There is a possibility that new measures could limit who can invest in AI startups within China. However, the specific scope, implementation, and timing of any potential restrictions remain unclear. Some sources indicated that the limitations might only apply to models that are released in the future.
This discussion in China comes in the wake of the U.S. imposing restrictions on foreign access to advanced AI models. In June, the Trump administration ordered that foreign nationals be barred from accessing Anthropic's advanced models, Fable and Mythos. While export controls on the publicly available Fable have since been lifted, access to the cybersecurity-focused Mythos remains limited to select 'trusted' U.S. institutions.
Chinese authorities are particularly concerned that Mythos could be exploited to take advantage of software vulnerabilities or used by the U.S. in ways that could harm China's interests.
China has already been strengthening its protective measures for domestic AI technology. In April, the National Development and Reform Commission ordered Meta to withdraw its $2 billion acquisition of the Chinese AI startup Manus, and in early June, it announced new regulations to tighten controls on overseas transactions involving Chinese investors, technology, data, and national security.
Reuters reported that it remains unclear how the restrictions on foreign access to Chinese AI models would be implemented. However, a summary from a roundtable published in May in the official journal of the Supreme People's Court indicated that Chinese legal experts proposed a tiered regulatory system. This would involve simple reporting procedures for basic open-source tools, security reviews for more advanced technologies, and outright bans or domestic use restrictions for the most sensitive frontier models.
* This article has been translated by AI.
Copyright ⓒ Aju Press All rights reserved.