
Indian data-centre operators could face stricter customer verification and compliance requirements following new US restrictions targeting remote access to advanced Artificial Intelligence chips. The measures are designed to prevent sensitive US-origin semiconductor technology from reaching restricted entities, particularly companies based in China and Macau, through overseas computing infrastructure.
The US Bureau of Industry and Security (BIS) released a 132-page interim final rule on November 16 titled “Addressing Remote Access to Advanced Semiconductors”. The policy expands controls over the end-users and end-uses of advanced chips, potentially placing greater responsibility on overseas data-centre providers that rent out graphics processing units (GPUs) and computing capacity.
For India, the development could create additional compliance obligations at a time when demand for AI infrastructure is increasing. Operators may need to examine not only the companies directly renting their computing resources but also their ultimate beneficial owners, parent organisations and actual end-users.
The rules could also affect the speed at which some data-centre projects and AI computing services are deployed. Industry experts have warned that identifying the real users of computing infrastructure can be difficult when services are accessed through subsidiaries, intermediaries or companies registered in different countries.
What the New US AI Chip Restrictions Mean for Indian Data Centres
The new US policy changes export-control requirements relating to remote access to advanced semiconductors. It seeks to address situations in which companies outside restricted countries obtain access to advanced US-origin chips and subsequently provide computing services to entities that may not be authorised to use them.
Under the rule described in the report, Data Centres using advanced US chips exceeding 12 million Total Processing Performance (TPP) must obtain a licence if they intend to export them to companies headquartered in China or Macau.
This could have implications for Indian operators that purchase advanced US-origin GPUs and rent computing capacity to businesses developing or training AI models.
For example, an Indian data-centre company may purchase advanced GPUs from a US supplier and provide access to an Indian customer. If a Chinese company subsequently seeks to use that computing infrastructure to train its AI models, the arrangement could trigger US authorisation requirements under the applicable rules.
The central issue is that access to computing infrastructure may now receive closer scrutiny, even when the physical chips remain inside an Indian data centre. Companies will need to understand who is using the infrastructure and whether the ultimate beneficiary is subject to US export restrictions.
Why KYC Requirements Could Become More Stringent
The expanded controls could place greater responsibility on data-centre service providers to verify the identities and business relationships of their customers.
Traditionally, customer onboarding may involve checking a company’s registration documents, business activities and contractual commitments. However, the new regulatory environment could require operators to look beyond the immediate customer and establish who ultimately controls or benefits from access to advanced computing resources.
Counterpoint Research co-founder Neil Shah told Moneycontrol that the responsibility would increasingly fall on service providers renting out GPUs and computing capacity. According to him, operators would need to conduct due diligence on the entities accessing their infrastructure, creating a significant Know Your Customer (KYC) and compliance burden.
This means Indian data-centre companies may need to strengthen their customer verification processes, ownership checks and monitoring systems. They could also have to establish whether a customer is acting independently or providing computing access to another organisation.
The distinction is important because a company may appear to be an ordinary domestic customer while its ownership structure, funding or business arrangements involve entities operating in restricted jurisdictions.
Why the US Is Targeting Remote Access to Advanced AI Chips
Advanced AI chips are critical components of modern artificial intelligence infrastructure. They provide the computing power required to train large AI models, process complex workloads and operate sophisticated applications.
Access to these chips has become an important strategic issue for the United States because advanced semiconductor technology can have commercial, military and intelligence applications.
US export controls are intended to restrict the transfer of sensitive technology to entities considered potential security risks. However, controlling physical shipments alone may not be sufficient when companies can access computing resources remotely through cloud platforms and overseas data centres.
A business does not necessarily need to own a physical GPU to benefit from its computing power. It may rent access to a server or virtual machine hosted in another country and use that infrastructure to train or operate AI models.
This creates a regulatory challenge because the location of the hardware, the location of the service provider and the location of the ultimate user may all be different.
The latest measures seek to address this gap by placing greater emphasis on remote access and end-use verification. For data-centre operators, this could mean that compliance responsibilities extend beyond the physical movement of chips to the digital services delivered through them.
Five Countries Face New US Restrictions
The Trump administration has introduced additional restrictions involving Singapore, Indonesia, Brazil, Malaysia and Thailand. The measures follow concerns about the possible diversion of advanced US-origin AI chips through third countries.
The restrictions reflect Washington’s broader effort to prevent companies in China from obtaining access to advanced computing technology through indirect supply arrangements.
Concerns about chip diversion have become more prominent as AI infrastructure expands across Asia and other regions. Companies may purchase advanced processors through overseas subsidiaries or use cloud computing services provided by businesses operating outside their home markets.
Such arrangements can make it harder for regulators and suppliers to establish the final destination and purpose of the technology.
For India, the new requirements are particularly relevant because the country is seeking to expand its data-centre capacity and attract investment in AI infrastructure. Operators will need to understand how the updated US rules apply to their equipment, customers and service agreements.
Indian Data Centres Could Face Greater Scrutiny of Ownership Structures
One of the major challenges for Indian data-centre operators will be identifying the ultimate beneficial owners of companies seeking access to advanced computing infrastructure.
An ultimate beneficial owner is the individual or organisation that ultimately owns, controls or benefits from a business, even when the ownership is held through multiple companies or intermediaries.
According to an industry executive cited by Moneycontrol, contracts restricting access by Chinese and North Korean entities are already common in the sector. However, enforcing those restrictions can be difficult when computing resources are accessed through subsidiaries or shell companies operating in other countries.
The executive said a virtual machine sold to an Indian entity could potentially be accessed by related companies elsewhere. These intermediaries could operate in Southeast Asian countries, while the ultimate users of the computing resources may be located in China.
This illustrates why checking only the immediate customer’s registration details may not be sufficient. Operators could need to examine ownership structures, business relationships and the parties receiving actual access to their infrastructure.
Such checks may be particularly challenging when customers operate through several layers of subsidiaries or have complex international ownership arrangements.
Key Compliance Checks Data-Centre Operators May Need to Strengthen
Industry sources have identified several areas where data-centre operators could face greater scrutiny under the expanded US controls.
- Ultimate beneficial ownership: Operators may need to identify the actual owners and parent companies behind customers, particularly where businesses have links to China, Macau or other restricted jurisdictions.
- End-user verification: Companies may need to establish which individuals and organisations ultimately access their GPUs and computing infrastructure.
- Intermediary arrangements: Operators may have to examine whether non-Chinese companies are providing computing services to Chinese entities through indirect arrangements.
- Customer documentation: Maintaining KYC records, ownership disclosures and end-use declarations could become more important.
- Access monitoring: Data-centre providers may need stronger mechanisms to monitor how customers use computing resources and identify potentially restricted access.
These measures could require changes to customer onboarding procedures, contractual arrangements and internal compliance systems. Operators may also need specialised teams to assess the regulatory risks associated with different customers and transactions.
For companies serving international clients, the process could become more complicated because customer relationships may involve multiple jurisdictions and different regulatory requirements.
Enforcement Could Be Difficult Despite Stricter Rules
Although the new requirements could strengthen oversight, enforcing them consistently may prove challenging.
Neil Shah pointed out that the regulation does not provide sufficient clarity on the frequency of audits or the extent to which service providers are expected to verify their customers. This uncertainty could make it difficult for operators to determine exactly how much due diligence is necessary.
For example, a relatively small company registered in India may rent computing capacity from a data centre to fine-tune an AI model. However, foreign investors or related entities could have an interest in the company, or the model being developed could eventually be used by organisations in another country.
In such circumstances, identifying the actual end-user may require more than a routine KYC check. Operators may need to examine ownership records, contractual relationships, funding arrangements and the purpose for which computing capacity is being used.
There is also a practical question about how frequently data-centre providers should review customer information. Ownership structures and business relationships can change, meaning that verification conducted during initial onboarding may not always reflect the current situation.
Without clearer guidance on audit expectations and monitoring responsibilities, companies could face uncertainty about how to demonstrate compliance.
Higher Compliance Costs Could Slow AI Infrastructure Deployment
The new restrictions could increase the operational costs of running data-centre businesses in India. Stronger KYC checks, ownership verification and access monitoring may require additional investment in compliance technology and specialised personnel.
Operators may also need to spend more time reviewing customer applications and assessing whether proposed computing arrangements comply with US export-control requirements.
Neil Shah said the measures would result in significantly more compliance work and could slow the deployment and rollout of infrastructure.
This could affect businesses that depend on quick access to advanced GPUs for AI development. Startups, research organisations and enterprises may experience longer onboarding procedures if data-centre providers introduce more detailed customer verification requirements.
However, the impact could vary depending on the operator’s customer base, the type of chips being used and the nature of the computing services being offered.
Large providers with established compliance teams may be better equipped to absorb the additional requirements. Smaller operators could face a greater burden because they may have fewer resources to manage complex international regulatory obligations.
India’s Participation in Pax Silica Adds Strategic Significance
The timing of the US rule is also significant because it was released on the same day that signatories to the US-led Pax Silica initiative signed a declaration on safeguarding AI supply chains and critical infrastructure.
India is among the countries participating in the initiative, which focuses on cooperation around the supply chains and systems supporting artificial intelligence.
The declaration highlighted concerns about the theft, illicit transfer or acquisition of AI-related supply-chain capabilities by hostile actors. Participants recognised that such activities could provide significant military or intelligence advantages or create undue economic leverage.
It also emphasised that the misuse of critical AI infrastructure could pose risks to the economic and national security of participating countries.
The parallel developments underline the growing importance of trusted supply chains in the global AI industry. Governments are increasingly treating advanced chips, computing infrastructure and related technologies as strategic assets rather than ordinary commercial products.
For India, participation in international cooperation initiatives could strengthen its position in discussions about AI supply-chain security. At the same time, domestic data-centre operators may need to align their business practices with evolving international expectations.
What the New Rules Mean for India’s AI Industry
India’s expanding data-centre sector is likely to remain an important part of the country’s AI ambitions. Advanced computing infrastructure is required for training AI models, developing applications and supporting enterprise adoption.
However, access to advanced US-origin chips is subject to export-control requirements that can affect how infrastructure is purchased, operated and made available to customers.
The latest restrictions could make compliance a more important part of the data-centre business model. Operators may need to establish stronger processes for verifying customers, identifying beneficial owners and monitoring the ultimate use of computing resources.
These requirements could increase costs and introduce delays, but they may also encourage greater transparency in the market. Businesses that develop robust compliance systems could be better positioned to serve international customers and maintain access to advanced technology.
The central challenge will be balancing regulatory compliance with the need to expand AI infrastructure quickly. Indian operators will need clarity on their responsibilities, particularly regarding audits, customer verification and indirect access arrangements.
Indian Data-Centre Operators Face a New Compliance Test
The US decision to strengthen controls on remote access to advanced semiconductors signals a broader shift in how governments regulate AI infrastructure. The focus is no longer limited to where chips are shipped or which companies purchase them. Increasing attention is being given to who ultimately uses the computing power and how that access is provided.
For Indian data-centre operators, this could mean more detailed KYC procedures, greater scrutiny of corporate ownership and stronger monitoring of customers using advanced GPUs.
While the measures are intended to prevent restricted entities from accessing sensitive technology, their implementation could create operational challenges for legitimate businesses as well. Uncertainty around audit requirements and the verification of indirect users may remain a major concern.
As India expands its role in the global AI ecosystem, the ability to maintain trusted and transparent computing infrastructure could become an important competitive advantage. The next challenge for data-centre companies will be ensuring that stronger compliance does not unnecessarily slow the country’s broader AI development ambitions.
For breaking news and live news updates, like us on Facebook or follow us on Twitter and Instagram. Read more on Latest Business on thefoxdaily.com.

COMMENTS 0