China is facing a delicate regulatory balancing act as it weighs the security risks of open-weight AI models against an innovation strategy that has been crucial to its race for technological supremacy with the United States, researchers said.
Traditionally, open-weight models – which allow anyone to download code for free and run it on local hardware – have lagged months behind proprietary frontier models. But recent releases from Chinese labs have significantly narrowed that gap.
Zhipu AI’s GLM-5.2 recently became the first Chinese large language model (LLM) to rank among the top three globally on leading benchmarks, earning praise from users as the country’s first open-weight model reliable enough for daily coding workflows.
But as these AI models advance rapidly, analysts warned that a regulatory response could be on the horizon.
“As open-weight models approach the kind of cyber and biosecurity risks of Mythos and other leading-edge models, China may make the same calculation as the US and find them to be too dangerous to be released, especially in open form,” said Mark Witzke, a non-resident scholar at the University of California San Diego who researches US-China tech policy.
The emerging concerns stem from Anthropic’s Claude Mythos. Announced in April, the powerful LLM rattled global industries with its ability to autonomously identify and exploit cybersecurity vulnerabilities. Amid warnings that its advanced reasoning capabilities could lower the technical barriers to developing bioweapons, the US lab restricted Mythos to a select group of American organisations.
