What smart people are saying about China’s role in ‘pacing the frontier’

China has become the biggest question in one of AI's most important debates.

The debate took on a new dimension this week after China's top intelligence official published a rare essay warning that advanced AI poses major risks and requires stronger oversight, even as Beijing pushes to become a global AI leader.

The comments came days after Anthropic CEO Dario Amodei reignited the debate over "pacing the frontier." He argued that leading AI labs should slow the development of the most powerful models so safety can catch up, while maintaining a sufficient US lead over China through measures such as export controls and stronger AI security.

The debate also comes ahead of President Donald Trump's White House meeting with Chinese President Xi Jinping, with AI emerging as a key issue in US-China relations.

Here's what smart people are saying about China's role in pacing the AI frontier.

Kenton Thibaut, senior China fellow at the Atlantic Council's Digital Forensic Research Lab

A sweeping agreement between Washington and Beijing on slowing frontier AI development remains unlikely, even as both countries increasingly recognize the technology's risks, Kenton Thibaut, a senior China fellow at the Atlantic Council's Digital Forensic Research Lab, said.

China worries AI safety could become another tool for the US to preserve its technological lead, making a major breakthrough on AI safety "highly unlikely," Thibaut wrote.

Instead, she said the more realistic path is limited cooperation on preventing AI misuse by nonstate actors and building technical systems to monitor risks and respond to emergencies.

"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," she wrote.

Efforts to slow the development of frontier AI are unlikely to succeed without global coordination, Mizuho analysts led by Vijay Rakesh wrote in a note on Monday.

The firm says unilateral guardrails risk ceding ground to Chinese competitors, which could close the gap with US frontier models while advancing their own computing capabilities.