The Trump administration moved to cut off Chinese companies’ remote access to advanced American AI chips, issuing a rule on Nov. 16 that restricts exports to overseas data centers used to train artificial intelligence models. [2]
The Commerce Department’s Bureau of Industry and Security published the interim final rule, which bars data centers or networks capable of delivering over 12 million total processing performance — roughly the output of 750 Nvidia H100 or H200 chips — from supplying that compute to restricted users. [3][4]
Companies in Singapore, Thailand, Malaysia, Indonesia and Brazil must now apply for U.S. export licenses to receive the most advanced chips, because those countries “present a higher-than-normal risk of diversion,” the Commerce Department said. [2][3]
The rule also covers access through brokers or go-betweens, and requires exporters and data center operators to monitor customers’ ownership structures and activities. [3]
“The rule is designed to ensure that existing restrictions on access to sensitive American technology cannot be circumvented through remote computing arrangements involving data centers in third countries,” the bureau said in a statement. [2][4]
Commerce Secretary Howard Lutnick received wide discretion under the rule, including the authority to inform chip sellers “orally” of stricter export requirements on specific items if the department identifies an unacceptable risk. [1]
Senator Maria Cantwell weighed in on enforcement. “America cannot maintain its leadership in AI if the agency responsible for protecting that advantage fails to enforce its own controls,” Cantwell said. [5]



