Chen Yixin warned AI could threaten political stability and cybersecurity
Chen said China had over 500 million AI users as of June 2026
Chen said bad actors use AI-generated content to spread political rumors
Chen warned U.S. AI models are fueling malware and hacking capabilities
AI is expected to be a key topic when Trump and Xi meet Sept. 24
China's national security authority said AI advancement could threaten political stability as the U.S. and China prepare for AI safety talks ahead of a planned Trump-Xi meeting later this month.
In an article posted Sunday on the state-run China Cyberspace magazine, State Security Minister Chen Yixin wrote that AI is reshaping global power while posing serious risks to political stability and cybersecurity.
Chen said that AI has been widely used in China, with over 500 million users as of June 2026 and daily token usage exceeding 140 trillion. The official said bad actors have used AI-generated text and images to form "troll armies" to spread political rumors.
The national security official warned that U.S. models, including Anthropic's Claude Mythos and OpenAI's GPT-5.5-Cyber, are fueling the development of malware and hacking capabilities.
Chen also said that foreign intelligence agencies are using AI to crawl and mine data in an effort to collect state secrets.
"Some domestic users lack sufficient security awareness and use overseas AI products to process sensitive information and send data abroad, causing large-scale leakage of domestic data," Chen wrote.
The national security chief argued that AI is fundamentally transforming warfare, citing the use of AI for targeting, decision support, and cognitive operations. "Whoever can use algorithms to achieve finer perception, more accurate judgment, and more precise strikes will seize the initiative on the battlefield," Chen said.
Chen's comments come as U.S. President Donald Trump and Chinese President Xi Jinping are scheduled to meet on Sept. 24 at the White House, with AI expected to be a key topic. At their previous meeting in May, the two leaders discussed AI guardrails but produced limited results.
Meanwhile, over the weekend, Anthropic CEO Dario Amodei called for slowing the pace of frontier AI development, citing growing safety risks. OpenAI's Sam Altman, Google DeepMind's Demis Hassabis, and xAI founder Elon Musk also expressed similar concerns or backed the idea.
Last week, several U.S. federal agencies alleged that Chinese AI companies systematically extracted capabilities from American AI models through distillation. Anthropic has also accused Moonshot and DeepSeek of silently routing user requests to Claude models and displaying the responses to users as its own.
China has pushed back against the allegations, calling them groundless and arguing that the U.S. move "politicizes and instrumentalizes distillation," which it said was a "common practice" in the AI field.
Despite the tensions, an Atlantic Council researcher suggested that shared concerns about AI slipping out of control could provide a rare diplomatic opening.
"There are strong signals that Washington and Beijing share an understanding of frontier risks," said Kenton Thibaut, a senior resident China fellow at the Atlantic Council. "[Loss] of control may offer a rare basis for cooperation - provided both sides can keep AI safety from becoming another front in their technology contest."
"China has technical reasons to engage: Chinese developers lose visibility when users deploy their predominantly open-weight models on infrastructure outside China, and evidence of abuse could expose vulnerabilities and improve safeguards," Thibaut added.
