KUALA LUMPUR: Most Malaysian businesses are unlikely to face immediate operational disruptions from the United States’ latest restrictions on advanced artificial intelligence (AI) chips, says IP ServerOne Solutions Sdn Bhd.
In a statement, the Malaysian cloud and infrastructure provider said conventional cloud, hosting and business IT services serving non-restricted customers were less exposed to the new measures.
However, providers offering advanced graphics processing unit (GPU) computing services could face additional regulatory requirements.
The restrictions, introduced by the US on Nov 16, require companies in Malaysia and four other countries to obtain US approval to access certain advanced AI chips, including through rented computing capacity.
The development coincides with Malaysia’s proposed data-centre legislation, announced on Nov 12, which is expected to require operators to disclose more information about their facilities, ownership and imported chips.
IP ServerOne said the two developments underscored the growing importance of transparency, regulatory compliance and data security as businesses expand their adoption of cloud and AI services.
The developments come amid substantial investments in Malaysia’s digital infrastructure.
Data centre and cloud-computing projects accounted for RM95.8 billion, or nearly 44 per cent, of the country’s RM218.5 billion in approved investments in the first half of 2026, according to Malaysian Investment Development Authority.
IP ServerOne founder Lee Cheung Loong said most of the company’s customers would continue their day-to-day operations as usual.
However, he noted that businesses needed to pay closer attention to where their data was hosted, who had access to computing resources and whether service providers complied with relevant requirements.
“For most of our customers, daily work goes on as normal. What has changed is the questions people should ask. Where is my data kept? Who runs the system? Is it run the right way? These now matter as much as price,” he said.
The company said about 98 per cent of its customers were Malaysian businesses, with demand for cloud and AI computing continuing to grow.
More customers, particularly healthcare providers and professional services firms, were exploring its NovaGPU GPU-as-a-Service offering as they sought to process confidential information without relying on public AI tools.
It said the most powerful AI computing resources under NovaGPU were offered exclusively to Malaysian businesses, with customers verified before access was granted.
Chips covered by US restrictions would only be made available after the necessary compliance reviews had been completed, it added.
According to MBSB Investment Bank Bhd, providers of GPU computing for rent, operators serving customers in jurisdictions subject to additional restrictions and certain new AI projects could face tighter regulatory requirements and potential delays while licences were obtained.
The implications could also extend to lenders and investors, as the risk profile of data-centre operators would increasingly depend on their customer base.
Equipment suppliers could likewise face longer approval timelines when shipping advanced AI servers into Malaysia.
Launched in April 2025, NovaGPU provides dedicated GPU computing resources from IP ServerOne’s data centres in Cyberjaya and Sarawak.
The company also offers private AI solutions that allow organisations to process their own data in dedicated environments, reducing their reliance on public chatbots for handling confidential information.
