India's AI infrastructure expansion reveals regulatory and technology dependency risks under looser US policy.
India stands to benefit from the Trump administration's AI deregulation in the short term, but analysts warn the policy shift could deepen long-term reliance on American technology providers. The Trump administration released its AI Action Plan on July 23, 2025, following Executive Order 14179 issued in January 2025, directing officials to draft a strategy to strengthen US leadership in artificial intelligence and reduce regulatory barriers. According to Dr. Danish Faruqui, CEO of Fab Economics, "In the short-term three to five-year period, Trump's AI Action Plan supports the AI infrastructure and AI-for-everyone growth across Asia, including India, with easy access supported by deregulation." The policy marks a reversal from Biden-era restrictions, clearing key US-headquartered companies—Nvidia, AMD, Intel, Micron, and Seagate—to ship previously sanctioned products including the Nvidia H20, AMD MI 308, Intel Gaudi, and Micron's HBM3E to Asian markets.
This opening creates substantial growth opportunities. Fab Economics' proprietary modeling indicates the Asian datacenter business from 2025 to 2029 holds potential for over US$1.2 trillion in AI chip consumption across compute, networking, memory, and storage, driven by over 30 GW of AI infrastructure projects. US-headquartered players stand to capture the majority of this expansion, generating over US$345 billion in cumulative annual revenue growth. However, Faruqui cautioned that "In the longer term 7 to 10 years effect, it dampens the urgency for developing domestic solutions for AI chip design, architectures, semiconductors, connectivity solutions, LLM and other components across the AI value chain, as all Asian nations like in the pre-2020 era would now after five years again be reliant on US headquartered AI players for AI infrastructure needs."
India's US$10 trillion economic vision centers on AI-led growth, yet that vision depends heavily on scaling AI infrastructure through imports. Faruqui noted that "India's US$10 trillion economic vision is predicated on AI-led growth, which in turn relies on domestic AI hardware infrastructure scale-up to drive AI use cases across public, enterprise, and individual levels," and "AI hardware infrastructure supply currently needs to be imported in India, which is mostly produced by US-headquartered AI value chain players across compute, networking, memory, connectivity, and storage datacenter infrastructure components like Nvidia, AMD, Intel, Micron, and Seagate."
To avoid excessive reliance, Fab Economics advises India to pursue a dual approach. "India must recalibrate and develop an AI datacenter investment policy amidst recent changes, one that is attractive globally and meets all success factors required for AI infrastructure site selection and setup," Faruqui said. Simultaneously, "India must focus on aggressive development of domestic AI hardware and software stack components by garnering support from both US and non-US partners, though treading carefully to balance geopolitical guardrails on AI technology to ensure sustained access." As Faruqui described it, "This is a seize-it-all moment, as it competes for AI infrastructure, domestic and FDI-based projects with other Asian countries."