Potential US curbs on AI model exports 'hegemonic' action, experts say

Chinese experts said on Thursday that potential export restrictions by the US on artificial intelligence (AI) models mainly target China, and the reported move is essentially a case of self-isolation and a hegemonic action, which will hinder the swift evolution of the global AI industry and dampen previous strides in globalization.

The so-called security concerns on the use of advanced AI models are completely political rhetoric meant to stoke fears over China's development, amid the intensifying US crackdown on Chinese products, observers said.

Reuters has reported that the US government is considering a new regulatory push to restrict exports of proprietary or closed-source AI models, whose software, and the data they train on, are kept under wraps.

The action was due to concerns that "US adversaries could use the models… to wage aggressive cyber-attacks or even create potent biological weapons."

The move adds to restrictions Washington has put in place to block exports of sophisticated AI technologies to China.

As China's AI sector booms, the US government has been acutely aware of the competitive pressure exerted by its biggest rival. With AI models emerging as a new focal point of competition, the US has continuously used the "national security threat" rhetoric to stoke fears over Chinese products, in a bid to hinder China's industrial upgrading, Ma Jihua, a veteran telecom industry observer, told the Global Times on Thursday.

The US announced in April it was adding four Chinese firms to an export blacklist for allegedly aiding Chinese entities in acquiring AI chips in violation of US regulations, according to media reports. Analysts noted that the action was part of the intensifying US crackdown on China's high-tech sector in recent years.

China's Foreign Ministry said in April that "we strongly oppose the illegal US unilateral sanctions" and vowed to do "what is necessary to firmly safeguard the lawful rights and interests of Chinese companies while urging the US to stop politicizing trade and tech issues and turning them into weapons."

Regardless of the measures the US may ultimately enact, the impact on China's AI sector is expected to be minimal, analysts said. Instead, the likely move would drive Chinese users toward home-grown large language models (LLMs), which are gaining robust momentum, Pan Helin, a member of the Expert Committee for Information and Communication Economy under the Ministry of Industry and Information Technology, told the Global Times on Thursday.

China's LLM sector is highly competitive, with a wide array of options available for its huge market. Given the disparities between Chinese and English contexts, homegrown AI products are better suited to Chinese consumers than their US counterparts, Pan said, highlighting the vast potential of the country's high-tech sector.

The Chinese Embassy in the US responded to inquiries by denouncing the reported action as a "typical act of economic coercion and unilateral bullying, which China firmly opposes," adding that it would take "necessary measures" to protect its interests, Reuters reported.

China is stepping up efforts to boost AI development and has witnessed fruitful outcomes. As of April, 117 LLMs had been registered in China for generative AI services, according to the Cyberspace Administration of China, reflecting the fierce competition in this burgeoning sector.

This year's Government Work Report noted AI's crucial role in accelerating new quality productive forces, highlighting the country's continuous push for the high-quality development of the digital economy by stepping up R&D and application of big data and AI, and planning to launch an AI Plus Initiative.