India's AI Paradox: Open Source Freedom, Hardware Dependency

India is aggressively charting a course toward technological self-reliance, particularly in the burgeoning field of artificial intelligence. The nation’s ambitious IndiaAI Mission, launched in 2024 with a substantial outlay, is designed to democratize AI access, foster job creation, and solidify India’s position as a global AI leader. Central to this vision is the drive to cultivate indigenous AI models, reducing reliance on foreign-made open-source alternatives and asserting digital sovereignty in a geopolitically complex world.

Under the “Making AI in India” and “Making AI Work for India and the World” philosophy, the IndiaAI Mission champions the development of foundational models tailored to India’s unique linguistic and cultural diversity. Initiatives like AIKosh, a unified platform, already offer over 1,200 India-specific datasets and 217 AI models, fostering a robust local ecosystem. The government’s commitment extends to providing accessible computing power, making 34,000 Graphics Processing Units (GPUs) available at an exceptionally affordable rate, less than $1 per hour, for innovators and researchers. Efforts are also underway to develop homegrown large language models (LLMs), with Bengaluru-based Sarvam AI receiving government backing for this critical endeavor, aimed at ensuring future access to powerful AI tools without external dependencies.

However, India’s quest for AI autonomy faces a significant paradox: while striving for software independence, the nation remains heavily reliant on foreign-made hardware, particularly advanced AI chips and high-end GPUs. This dependency presents a strategic vulnerability, amplified by geopolitical tensions and potential export controls that could limit India’s access to crucial AI hardware. Reports indicate that the US has, for instance, imposed limitations on GPU exports to countries like India, capping transfers at fewer than 1,700 GPUs annually per company. Such restrictions could impede India’s aspirations for large-scale AI data center deployment, potentially putting Indian enterprises at a disadvantage. The nation also grapples with a shortage of high-end AI hardware and a lack of deeply rooted hardware manufacturing expertise.

Recognizing this critical gap, India is embarking on an aggressive strategy to build a robust domestic AI hardware ecosystem. A key initiative is the development of an indigenous AI chip, expected to be ready by 2027, designed in collaboration with government institutions using the open-source RISC-V architecture. Complementing this, the Indian government has approved the establishment of six semiconductor manufacturing facilities across Gujarat, Assam, and Uttar Pradesh, attracting investments exceeding $15 billion. These include a major fab facility by Tata Electronics in Dholera, Gujarat, a joint venture with Powerchip Semiconductor Manufacturing Corporation, and other units focusing on assembly, testing, marking, and packaging (ATMP) and chip design. Furthermore, India has inaugurated its first 3nm chip design centers in Noida and Bengaluru, signaling its entry into the elite sphere of cutting-edge semiconductor design. The overarching Semicon India policy, launched with a substantial outlay, offers financial incentives and support for semiconductor fabrication and design, aiming to make India a global manufacturing hub and reduce import reliance. With 20% of the global semiconductor design workforce already based in India, the talent pool is a significant asset.

The journey toward complete self-reliance in AI, encompassing both software and hardware, is undeniably complex and capital-intensive. While India has made commendable strides in fostering indigenous AI model development and is aggressively investing in semiconductor manufacturing, the road ahead involves overcoming challenges such as high development costs, limited advanced fabrication infrastructure, and the imperative to retain top talent. Yet, with strategic policies, significant investments, and a burgeoning ecosystem, India is poised to transform from a consumer to a key contributor in the global AI hardware landscape, striving to resolve the paradox at the heart of its AI mission.