Nvidia has significantly reduced the number of Asian companies authorized to purchase AI processors directly, compiling a new list of approved customers after an extensive vetting process to prevent advanced chips from ending up in China via third countries. The stricter selection criteria apply to customers in Singapore, Malaysia, and Japan, among other countries, with more than half of the existing customers reportedly failing the first round of evaluation. Specialized cloud providers offering AI infrastructure, known as “neocloud” providers, are particularly affected. Nvidia has expanded its monitoring procedures in recent months, now inspecting customers’ data centers, reviewing contracts, and speaking with end users to verify hardware delivery. The U.S. Department of Commerce is also reportedly involved in these checks, aligning with the Trump administration’s policy. In May, the department published additional guidelines to prevent powerful AI chips from being supplied to foreign subsidiaries of Chinese companies, highlighting risks involving Nvidia’s latest Blackwell processors. Earlier this year, U.S. prosecutors indicted a co-founder of Supermicro and several employees for alleged involvement in a network that funneled AI chips to China via Southeast Asia, with the hardware valued at approximately $2.5 billion. Tightened controls have made it increasingly difficult to obtain advanced AI processors, and Nvidia’s H200 processor is virtually unavailable on the Chinese market.