The United States should begin preparing for scenarios in which extreme measures could be required to prevent China from achieving artificial general intelligence first, including state-backed espionage and military strikes against Chinese data centres, according to a former White House national security official.
Jacob Stokes, deputy director of the Indo-Pacific Security Program at the Centre for a New American Security (CNAS), made the remarks during an online event on Thursday, days after a new CNAS report called for Washington to examine diplomatic, espionage, cyber and kinetic options to prevent a Chinese AGI breakthrough.
Stokes, who served as a national security staffer in the Obama administration, said US agencies including the Department of Defense and the National Security Agency should begin determining what intelligence would be required to justify such measures.
“Trying to think through the particulars of that will be especially important, in part because it will help policymakers … start to work backwards based on the unique nature of the technology, in the same way that in a past era, policymakers would learn about nuclear weapons and … work backwards from the science to the policy implications,” he said.
AGI refers to a hypothetical AI system capable of achieving human levels of performance across a wide range of tasks. Stokes’ CNAS report, titled “Superpowers and AGI”, examines how the potential arrival of such technology could affect competition between the United States and China.
The United States currently holds a months-long lead over China in frontier AI models, according to Epoch AI. But Stokes warned that China could achieve a “surprise technological breakthrough”, potentially bringing it close to or allowing it to reach AGI first. He pointed to Chinese advances in robotics and other forms of “embodied AI” as part of the reason for considering such a scenario.
If China were to reach AGI first, Stokes wrote that Washington should consider extreme measures, including military strikes against the data centres used to train and operate a Chinese AGI system. His assessment also included the possibility of preventive strikes before China develops AGI.
The proposal reflects the growing strategic importance of the infrastructure underpinning advanced artificial intelligence. Stokes argued that US agencies would need to determine which facilities could be targeted and what weapons could be used, while also assessing the possibility that such attacks could trigger a wider war between the two countries.
The prospect drew a sharp warning from William Hartung, an arms analyst at the Project on Government Oversight, a Washington non-profit organisation. Any attack on Chinese data centres, he said, could “risk starting a shooting war between two nuclear armed powers that could have tragic consequences for all concerned”.
“Striking Chinese data centres based on how they might or might not be used in the future would be the height of recklessness,” Hartung said.
Stokes also proposed another option: attempting to reverse engineer or “steal” China’s AGI technology. He argued that such action could be justified “legally and normatively” because of China’s alleged history of stealing US intellectual property.
At the same time, he acknowledged a major institutional obstacle, saying that the US government and intelligence apparatus have “not been organised for espionage to gather technological secrets for decades”.
The debate comes as US President Donald Trump and Chinese President Xi Jinping prepare for talks in Washington later this month on AI governance. The discussions are taking place as advances in artificial intelligence raise increasingly significant concerns over national and global security.
Stokes said the “best case scenario” for the talks would be for the two sides to find them “constructive”, rather than expecting major breakthroughs on key governance issues.
The competing approaches outlined by Stokes — from diplomacy and intelligence gathering to cyber operations and potential military action — reflect the increasingly high stakes attached to the possibility that either superpower could reach AGI first.
