What is the New 'Remote Access' Rule?
On November 16, 2026, the U.S. Department of Commerce's Bureau of Industry and Security (BIS) issued an interim final rule to close a significant loophole in its AI chip export controls. Previously, restrictions focused on the physical sale and shipment
of high-performance AI chips to specific countries, most notably China. However, companies in restricted nations could still access the power of these chips by renting computing time from data centers located in third countries. The new rule treats this remote access as a regulated transfer. It requires a license to export advanced AI chips to any data center in the world if there's knowledge it will sell significant computing power to companies headquartered in restricted destinations, such as China, Russia, or Iran.
The Geopolitical 'Why' Behind the Rule
This policy move is a direct extension of the U.S. strategy to limit the technological advancement of strategic rivals. Washington aims to prevent its sensitive, American-made technology from being used in ways that could undermine its national security or foreign policy interests. The rule is designed to ensure that the formidable power of top-tier AI chips is not leveraged by restricted entities to train advanced AI models, particularly for military or surveillance purposes. The Commerce Department has stated the rule is intended to stop circumvention of existing controls, ensuring that what cannot be bought directly cannot simply be rented through the cloud.
The 'Know Your Customer' Burden on Cloud Providers
The rule places a significant new compliance burden on Infrastructure-as-a-Service (IaaS) providers and data center operators globally. They are now responsible for conducting due diligence on their customers. This means identifying not just the direct client but also their ultimate parent company's headquarters to ensure they are not from a restricted country. Data centers already servicing such clients have a 90-day grace period to obtain a license. This 'know-your-customer' requirement extends the compliance chain far beyond the chip manufacturer to anyone who owns or operates the infrastructure where these powerful chips are deployed.
Compliance Questions for Indian and Global Firms
For Indian technology firms and global corporations with operations in India, this rule raises immediate questions. Many companies rely on global cloud providers for their AI development and data processing needs. The key challenge is navigating a web of overlapping and sometimes conflicting international regulations. Firms must now verify where their cloud provider's data centers are located and what safeguards they have in place to comply with these US rules. Furthermore, the rule adds new direct license requirements for exporting advanced chips to several countries, including Singapore and Malaysia, which are key hubs in Asia's tech supply chain. This could complicate hardware procurement and increase operational costs for businesses that operate across these jurisdictions.
Navigating a Fragmented Regulatory World
The era of seamless global AI infrastructure is facing a major stress test. Companies must now prioritize robust data governance and create a clear cross-border data strategy. This includes mapping AI systems, documenting data flows, and understanding exactly which jurisdictions' laws apply to their operations. For companies in India, a nation aiming to build significant AI and data center capacity, the US rules add another layer of complexity to these ambitions. While some previous US frameworks placed India in a Tier 2 category with controlled access, the new remote-access rule applies universally, demanding vigilance from all users of global cloud infrastructure. The key is no longer just where you are, but who you might be providing services to, even indirectly.
















