A fresh round of sparring between Washington and Beijing over the future of artificial intelligence broke out this week after Anthropic's chief executive warned of the dangers of a Chinese lead in the technology. Beijing wasted little time firing back, calling the argument counterproductive.
What Amodei's essay actually says
In a lengthy essay posted Saturday to his personal website, Dario Amodei argued that a Chinese lead in artificial intelligence "would pose a grave danger" to both the United States and the rest of the world. He defended keeping current restrictions on selling cutting-edge semiconductors, and the equipment used to manufacture them, to China in place.
The essay went beyond export controls. Amodei also laid out a three-step plan meant to slow the industry's overall pace and said Anthropic would unilaterally give outside evaluators permanent, employee-level access to monitor its own systems. He warned that without similar coordination across other labs, AI agents could slip beyond human oversight — and potentially seize control of large parts of the internet — within six to twelve months, according to NPR's reporting. Amodei did add one caveat: pacing AI development worldwide will eventually require cooperation from China itself.
Beijing's blunt reply
China's Foreign Ministry responded Monday at its regular briefing. Spokesperson Guo Jiakun didn't mince words, warning that "fearmongering, confrontation and vicious competition will only disrupt global AI governance," a process he said serves no one's interest, and urging all sides to prioritize cooperation over a technological arms race.
The official response came alongside a broader media offensive. The state-run Global Times branded Amodei's argument "hostile" and "groundless," noting he referenced China roughly a dozen times while playing up alleged security threats. Separately, China's Commerce Ministry pushed back on U.S. claims that Chinese firms have leaned on "distillation" — a technique that trains one AI model using outputs from an already-trained one — to replicate American systems' capabilities. Beijing called those accusations baseless and accused Washington of chasing "technological hegemony" and trying to monopolize global computing power.
Why the timing isn't a coincidence
The clash lands just ten days before Trump is set to welcome Xi to U.S. soil, a meeting the president had flagged as far back as July for around September 24. Artificial intelligence is expected to be one of the central topics the two leaders discuss, alongside other pieces of the broader trade relationship. Beijing's decision to publicly rebut one of Silicon Valley's most influential executives just before the sit-down suggests a deliberate move to stake out a position ahead of negotiations.
For Washington, restrictions on advanced chips have for years been a centerpiece of its tech strategy toward China, with manufacturers like Nvidia subject to export licensing rules that have shifted from one administration to the next. Beijing, for its part, treats each new restriction as evidence that the U.S. is trying to contain its technological rise through political pressure rather than open market competition.
What comes next
With the summit fast approaching, it remains unclear whether chip export policy will end up as a bargaining chip in the Trump-Xi talks or whether Washington will hold its line regardless of what's discussed. As of this writing, neither the White House nor Anthropic had issued further comment on Beijing's remarks.
