US export rule forces Southeast Asian data centers to get licenses for AI chips
A Commerce Department interim rule requires licenses for advanced AI chip purchases in Singapore, Malaysia, Thailand and Indonesia, with 90 days for facilities serving Chinese customers to comply.

Why it matters
- Commerce Department interim rule released Monday requires licenses for advanced AI chip purchases by data centers in Singapore, Malaysia, Thailand and Indonesia.
- Facilities serving large Chinese customers have 90 days to apply for a license or face export restrictions.
- About 75% of China's overseas AI compute was forecast to sit in Southeast Asia by end of 2026, per the Institute for Progress.
- Capital Economics estimates the restrictions could cut 0.2 percentage points off Malaysia's GDP growth forecasts through 2028.
- Oracle canceled a multibillion-dollar deal giving Tencent access to about 100,000 advanced AI chips in early November.
Data centers in Singapore, Malaysia, Thailand and Indonesia will need a U.S. government license to buy advanced AI chips, including Nvidia's, under an interim rule the Commerce Department released Monday. Facilities already serving large Chinese customers have 90 days to apply for a license or face future export restrictions. The rule targets the pipeline through which Chinese technology firms have rented computing capacity abroad — a pipeline that, by October estimates from the Washington-based think tank Institute for Progress, was set to concentrate about 75% of China's overseas AI compute in Southeast Asia by the end of 2026.
The stakes are straightforward. U.S. restrictions on China's access to Nvidia's most advanced chips pushed Chinese firms, including Alibaba Group and ByteDance, to rent capacity in overseas data centers. That demand fueled a construction surge across the region, and Washington now wants to shut the workaround down. The rule lands while the U.S. presses ahead with its investigation into Nvidia's former CEO Jensen Huang, who left the company he co-founded in 1993 after reports of alleged extramarital affairs, including one with a suspected Chinese spy believed to have facilitated billions of dollars worth of Nvidia chips into China.
Who pays the cost?
The new restrictions will likely make building and operating these data centers "slower and costlier," said Jessica Brandt, a senior fellow for technology and national security at the Council on Foreign Relations. The rules could undercut the business model of neoclouds — newer cloud providers built around AI computing — in the region, she said. Operators should also expect more rigorous regulatory and national security requirements, according to Reema Bhattacharya, head of Asia research at risk intelligence company Verisk Maplecroft.
The caveat matters as much as the promise: greater scrutiny of investment in these projects could weigh on Asian economies that have benefited heavily from the build-out.
Why is Malaysia most exposed?
Malaysia stands to lose the most, analysts say, given the scale of capacity being built in Johor and the sector's reliance on Chinese customers. More than half of Malaysia's data center capacity sits in colocation facilities (data centers that rent space to third parties), with Chinese firms accounting for "a sizeable share of these workloads," said Leah Fahy, senior China economist at Capital Economics.
Restricting access to overseas cloud facilities would weaken the commercial case for further expansion and weigh on investment from both Chinese and U.S. companies, Fahy said. "Any resulting slowdown in investment would add to the drag from an anticipated cooling in the global AI export cycle," she said, potentially knocking 0.2 percentage points off Malaysia's GDP growth forecasts through 2028.
Goldman Sachs analysts offered a split verdict. Malaysian data centers that rely heavily on Chinese customers — particularly those built for AI training — may struggle to lease space, fill capacity and could be left with idle power. Operators with a mix of U.S. hyperscaler (the giant cloud platforms such as Amazon and Microsoft), sovereign and enterprise clients should be relatively insulated.
What is Thailand doing?
Thailand, with more than 60 data centers operating and about 30 under construction, is bracing for tighter scrutiny. Vice minister for commerce Kirida Bhaopichitr told CNBC's JP Ong in Manila on Sunday that the country would scrutinize data centers importing high-end chips and "crack down" on those that re-export them to third countries.
New projects that fail to secure U.S. approval for advanced chips may "not be able to complete construction," the vice minister said — though existing facilities providing AI services could be less affected. "We will have to discuss with the U.S. to make sure that the chips are coming," and are allowed for use in data centers operating in Thailand, Bhaopichitr said.
Regional governments are signaling compliance rather than resistance. Malaysia's investment, trade and industry minister, Johari Abdul Ghani, said Tuesday: "If you want to be in the game, you have to follow U.S. regulations." Singapore plans to negotiate "trusted access" to advanced AI chips for its companies.
Where does the demand go now?
The rule exempts U.S.-headquartered companies, analysts noted, leaving a narrow path for Chinese firms to keep renting advanced computing power from U.S. cloud giants — though those providers may grow more cautious about leasing to Chinese customers. "Potentially, U.S. firms may receive an advantage," Chong said.
Chris McGuire, a senior fellow at the Council on Foreign Relations and former National Security Council official, put the loophole bluntly: "There is a significant risk that Chinese companies will pivot from renting compute from data centers in Southeast Asia to renting compute from data centers in Virginia." He urged Washington to close that loophole before it is exploited.
Bhattacharya said some high-end capacity may shift to Japan, South Korea and Taiwan, though power and land constraints may limit how much those markets can absorb. Chinese companies are already moving: CNBC reported earlier this month that Chinese developer Moonshot AI tried to access Nvidia chips through data centers in Europe. Oracle, according to the Financial Times, canceled a multibillion-dollar chip-leasing agreement with Tencent in early November — one that would have given the Chinese company access to about 100,000 advanced AI chips — bowing to pressure from U.S. administration officials.
The takeaway in plain words: Washington has turned Southeast Asia's data center boom from a growth story into a compliance test, and Malaysia's bet on Chinese demand now carries a measurable macroeconomic cost.
Source: CNBC Tech
More from James Calloway
Show full bio
News editor covering people, launches and day-to-day practice in artificial intelligence and machine learning at AI In Context.
749 articles
Related articles
- Chinese AI Firms Rush to Lock Up Overseas Nvidia Capacity
- US Moves to Close Cloud Loophole as China Builds Distillation Pipeline
- Nvidia Chips Reach China's AI Labs Through a Legal Rental Route
- Smuggled Chips and Server Loopholes: How China Gets Nvidia GPUs
- Chinese AI Data Centers Still Specify Restricted Nvidia Chips