Beijing Reads Anthropic CEO’s AI Safety Essay As A ‘Cold War’ Attack

Beijing’s state media alleged that an American tech giant CEO’s recent safety warning about artificial intelligence was “in essence” a plot to contain China.

Chinese state-backed news outlet Global Times called Anthropic CEO Dario Amodei’s essay advocating for limits on frontier AI development a “Cold War” strategy to maintain U.S. dominance in the industry, according to a Sunday editorial. These comments land just before President Donald Trump’s anticipated Sept. 24 summit with Chinese President Xi Jinping, where AI safety is among the topics the leaders plan to discuss, Reuters reported.

“On the surface, this appears to be a ‘rational statement’ focused on global AI security,” the Global Times wrote. “A close reading, however, reveals its hidden agenda: The article is packed with containment provisions targeting China and is, in essence, a ‘Cold War playbook’ for the AI sector.”

The Chinese Communist Party (CCP) linked publication said Amodei’s critique was “hypocritical and short-sighted” and was instead a push for regulatory barriers on Chinese AI that “politicizes” issues in the industry. 

“Pacing within democracies will be limited by the lead that US companies have over authoritarian regimes, chiefly the [CCP],” Amodei wrote in his Saturday essay. “If we slow down by more than this amount, then (unpaced) CCP-associated projects will pull ahead, creating significant national security risk.”

Anthropic and the Chinese Embassy in the U.S. did not immediately respond to the Daily Caller News Foundation’s request for comment.

Beijing’s state security minister, Chen Yixin, said U.S. AI models threaten the national security of China, South China Morning Post reported Monday. AI is “a new arena for strategic rivalry among major powers,” according to Yixin, who said the technology threatens the ideological safety of China if bad actors generate mass amounts of false information.

The president, who said his meeting with Xi will be “very exciting,” recently emphasized the need to stay ahead in the AI race.

“We’re leading China in AI. We’re the most sophisticated country in the world and, frankly, I want to keep it that way — because whoever wins AI wins,” Trump said in a video shared Sunday on X by Rapid Response 47.

Trump added that while guardrails could make AI safer, “negative forces” are talking about these measures as well.

Anthropic, a critic of China, said it found that Chinese AI companies had distilled its Claude model to train their own models in February.

“We have identified industrial-scale campaigns by three AI laboratories — DeepSeek, Moonshot, and MiniMax —to illicitly extract Claude’s capabilities to improve their own models,” the company said in a Feb. 23 press release.

A joint cybersecurity advisory, comprised of the National Security Agency (NSA), Cybersecurity and Infrastructure Security Agency (CISA) and the Federal Bureau of Investigation (FBI), warned U.S. AI companies about the threat of China and other potential bad actors stealing model tokens on Sept. 8, according to a press release.

“We strongly urge AI companies to take immediate steps to safeguard their platforms against knowledge distillation campaigns that threaten to close the gap in advancements made by American companies,” the advisory stated.