In a pivotal policy shift, the United States has introduced restrictions on the import of advanced Graphics Processing Units (GPUs) to India as part of a broader Export Control Framework for Artificial Intelligence (AI) technologies. Aiming at regulating the global diffusion of AI capabilities, these restrictions impose licensing requirements for importing advanced AI chips.
U.S. providers such as Microsoft, Amazon, and Meta are anticipated to secure global authorizations, yet they will be restricted to deploying no more than 50% of their AI computing capacity outside the United States.
"The US imposing restrictions on import of GPU’s to India brings us back to the old licensing regime to haunt us again"
Reacting to the restrictions imposed by the US, Dr Ajai Chowdhry, Founder HCL, Chairman EPIC Foundation & MGB, National Quantum Mission of India commented, “The US imposing restrictions on import of GPU’s to India brings us back to the old licensing regime to haunt us again. And China can follow with other technologies. And the rest of the West too. Imagine export of semiconductor manufacturing machines being restricted tomorrow which will stop our semiconductor ambitions!
This is reminiscent of the export controls on Space and Atomic energy that we faced in the past. And we then emerged out of that successfully getting our brilliant engineers creating our own technologies.
I am sure most of the design of NVIDIA and AMD GPU’s must have been done in India by Indian engineers.
Recently we had written about how critically we have to look at strategic autonomy and become self-sufficient in key technologies like semiconductors, quantum technologies, security and defence, AI and drones.
This is the new Unipolar world where every country is on its own. And with the US and China vying for global dominance, India’s rise will resisted by both.
In semiconductor chips we have created our own RISCV technology in IITM for which we don’t need to keep paying licence fees. Using this we should design our own chips in RISCV to secure ourselves from future sanctions. This should be taken up in a mission mode with funds provided under ISM’s new DLI policy. And open up DLI for all Indian companies including corporates/ERCD exporters, MSME’S and Startup by increasing the size of the funding from the small 30 CR with generous 50 CR to 150 CR. Maybe we can create our own NVIDIA and AMD in the next ten years!”
Ashok Chandak, President of IESA (India Electronics and Semiconductor Association said, “The Export Control Framework for Artificial Intelligence (AI) diffusion, introduced by the Biden administration, is expected to significantly affect global AI development and deployment. These restrictions extend beyond U.S.-based companies, influencing nations like India, which now faces licensing requirements for importing advanced AI chips.”
The Advanced AI Chip Control from the U.S. will operate through a tiered access system.
1) The closest U.S. partners, including 18 countries, will be permitted unlimited access to advanced AI chips with profound export controls and sans any restrictions on verified companies.
2) Other non-embargoed countries will have new licensing pathways for chip production and development, with a low-volume exemption of up to 1,700 GPUs, valued at approximately $40-50 million, and license reviews for larger orders of up to $1 billion worth of chips.
3) Embargoed countries will face strict restrictions,
4) The provisions allow nations to shift between tiers following up the strategic relationships alongside export control commitments.
The Data Center Validation framework also lays out robust validation programs to ensure compliance:
1) Universal Validated End Users (UVEU) offers single authorization for the U.S. companies for global data centers, excluding embargoed countries. It requires match up the standards of NSA and FedRAMP guidance, with specific requirement for geographic chip deployment that is the minimum of 50% chips in the US and 75% for close allies, and a maximum of 7% in one select country.
2) National Validated End Users (NVEU) is designed for companies apart from the U.S. allies and partners. This program progressively limits chip allocations over the next few years with less than 100,000 units of H-100-equivalent GPUs allowed in 2025, increasing to 270,000 units in 2026, and 320,000 units in 2027.
Impact Of Restrictions On India
India's AI sector is likely to encounter several challenges due to the imposed restrictions. The restrictions to advanced AI chips could affect innovation and slow down the process of installations. Licensing requirements may also increase procurement costs and cause delays due to the need for authorizations. Additionally, Indian companies might become more dependent on global corporations for AI infrastructure, including data centers. However, in the short term, these restrictions is estimated to leave minimal impact.
Asok Chandak further added, “India could potentially secure General National Validated End User (NVEU) authorizations due to its status as neither a re-exporter of Compute ICs nor an advanced compute manufacturing hub. This distinction, combined with the presence of significant Indian design centers for GPU makers like NVIDIA and AMD and their management commitment of support to India, could position India favorably for license approvals.”
“India's National AI Mission aims to develop infrastructure with over 10,000 GPUs through public-private partnerships, supported by a ₹10,000 crore investment over five years. In the short term, the new export controls may not significantly impact India. However, the uncertainty of securing licenses and trade negotiations could challenge India's ambitions for large-scale AI hardware deployment.”
“India's National AI Mission aims to develop infrastructure with over 10,000 GPUs through public-private partnerships, supported by a ₹10,000 crore investment over five years. In the short term, the new export controls may not significantly impact India. However, the uncertainty of securing licenses and trade negotiations could challenge India's ambitions for large-scale AI hardware deployment,” he further added.
The expected National Validated End User (NVEU) restrictions will progressively limit the allocation of H-100-equivalent GPUs over the coming years. In 2025, companies can import less than 100,000 units, followed by 270,000 units in 2026 and 320,000 units in 2027. However, small quantities of less than 1,700 units per company per year will be eligible for a license exception, allowing for limited flexibility in smaller-scale deployments.
With aim to streamline the licensing under new rules for verified users, it also maintains security. However, there are administrative challenges associated with the restrictions. The licensing process for Validated End Users (VEUs) involves strict conditions, including continuous monitoring and required diligence, compliance to export restrictions and use policies, and a decade of record maintenance. Any failure to comply with these conditions necessitates immediate notification to the Bureau of Industry and Security (BIS).
“The export controls are set to take effect in 120 days, allowing the incoming administration under President-elect Trump to potentially amend these rules. Thus it is bit uncertain whether Mr. Trump administration will make it easy or pass the rule as is. Amid growing concerns from technology industries, the global AI landscape may witness a shift, impacting both U.S. technological leadership and India's growth trajectory in AI infrastructure. Though , In short term India may not have major impact but in long run scaling up by any Indian conglomerate could face the hurdles of quantity cap,” concluded Chandak.
𝐒𝐭𝐚𝐲 𝐢𝐧𝐟𝐨𝐫𝐦𝐞𝐝 𝐰𝐢𝐭𝐡 𝐨𝐮𝐫 𝐥𝐚𝐭𝐞𝐬𝐭 𝐮𝐩𝐝𝐚𝐭𝐞𝐬 𝐛𝐲 𝐣𝐨𝐢𝐧𝐢𝐧𝐠 𝐭𝐡𝐞 WhatsApp Channel now! 👈📲
𝑭𝒐𝒍𝒍𝒐𝒘 𝑶𝒖𝒓 𝑺𝒐𝒄𝒊𝒂𝒍 𝑴𝒆𝒅𝒊𝒂 𝑷𝒂𝒈𝒆𝐬 👉 Facebook, LinkedIn, Twitter, Instagram