India's Push for AI Semiconductor Self-Reliance and Fissures in the Global Tech Hegemony

Executive Summary

India's drive to develop its own AI semiconductors stems from a long-held perception that technological dependence equates to security vulnerability—a lesson learned after the United States refused to export a supercomputer to India in 1985 and now being revisited in the age of AI. The Modi government is investing $15.4 billion in its Semicon 2.0 program, pursuing a hedging strategy to resist pressure from both the United States and China to align with their respective AI platforms. However, given the semiconductor industry's inherent need for long-term capital investment and technological accumulation, a significant gap between policy announcements and actual mass-production capabilities is likely to persist for some time. As India is more likely to establish itself as a key node in a diversified supply chain rather than achieve complete self-sufficiency, South Korea should approach India as both an alternative supply chain partner and a source of demand for cooperation in GPUs and foundry services. The trend of spillover from the U.S.-China tech rivalry extending to the industrial policies of third countries holds significant implications for South Korea's own trade and technology security strategies.

I. Analysis of the Current Situation

India's Push to Develop Domestic AI Semiconductors: An Analysis of the Current Situation

1. Background and Developments

India's push for semiconductor self-reliance is not a recent trend. In October 1985, the United States refused to sell a supercomputer to India, citing concerns about its potential diversion for nuclear weapons development[1]. This prompted the Indian government to establish the Centre for Development of Advanced Computing (C-Dac) in March 1988[1]. In August 1991, C-Dac succeeded in developing India's first indigenous supercomputer, the Param-8000[1]. This experience serves as the archetypal case that cemented the conviction within Indian policy circles that external technological dependence directly translates into security vulnerability.

Nearly 40 years later, this historical experience is being replayed in the realm of AI semiconductors. The Indian government projects that its dependence on semiconductor imports will swell to $240 billion annually by 2035[8]. On this basis, Prime Minister Narendra Modi's government is accelerating its policies to foster the semiconductor industry[8]. The focus of these policies has expanded beyond simple import substitution to building indigenous capabilities across the entire process, from design and manufacturing to packaging.

2. Current Situation

India's Ministry of Electronics and Information Technology (MeitY) has launched the second phase of its semiconductor development initiative, the Semicon 2.0 program, with a budget of 1.28 trillion rupees (approximately $15.4 billion)[5][8]. Aiming to attract modern fabs that use 12-inch standard wafers, the program has shifted away from large direct government subsidies, instead offering cash incentives of up to 40%[5]. Eligibility requirements have been tightened, with support restricted to proven manufacturers capable of investing at least 200 billion rupees ($2.1 billion) of their own capital[5]. Following the Tata Group's first commercial fab, the government is now seeking investment from global companies with the goal of attracting a second fab by 2031[5].

Nikkei Asia, reporting on India's aggressive moves, also conveyed expert opinions pointing to 'latecomer risk'[8]. The concern is that because the semiconductor industry requires long-term, large-scale capital investment and technological accumulation, a significant gap exists between policy announcements and the actual capability for mass production[8]. Within India, the government's shift from direct subsidies to a model focused on attracting private capital can be interpreted as a policy lesson learned from past experiences where large-scale financial support failed to produce timely, tangible results.

Meanwhile, India's attempt at semiconductor self-reliance is being pursued in parallel in adjacent electronics industries, such as smartphones. Minister of Electronics and Information Technology Ashwini Vaishnaw announced that through a 6.25 trillion rupee mobile phone manufacturing program, an indigenous Indian smartphone brand could emerge by mid-2027[12]. This is a response to the dominance of Chinese brands in the Indian market over the past decade and is in line with a broader national agenda to secure 'domestic brands and domestic intellectual property' not just in semiconductors but across the electronics industry as a whole.

3. Key Actors and Positions

The Indian Government frames this policy as an essential task from an economic security perspective. As evidenced by the move to develop indigenous SCADA systems for its power grid, vigilance against cyber threats from China underpins the government's broader policy of localizing critical infrastructure[16]. Semiconductor self-reliance is an extension of this security logic.

The United States and China are competing to draw India into their respective AI platform ecosystems. The Council on Foreign Relations (CFR) analyzes that India faces a dual challenge: it needs foreign investment and partnerships to address the structural vulnerabilities of its domestic AI ecosystem, while simultaneously needing to manage its relationships with both Washington and Beijing[4]. This suggests that India is likely to pursue a strategic hedging policy to extract practical benefits from both sides, rather than fully aligning with either camp.

Global Big Tech Companies The actions of global big tech companies are another variable influencing India's policy calculations. Companies like Google, Amazon, Microsoft, Meta, and OpenAI are developing their own AI chips to reduce their reliance on Nvidia's GPUs[10]. This trend away from Nvidia signals a diversification of the AI semiconductor supply chain, which is currently concentrated among a few firms, raising hopes among latecomers like India that a window for market entry may be opening. At the same time, the intensifying competition among leading firms to develop in-house hardware—as illustrated by reports that Anthropic considered acquiring the AI chip startup MatX[14]—also poses a risk for new entrants like India, as it could further widen the technology gap.

Key Supply Chain Node Countries like Taiwan and the Netherlands The positions of these countries also highlight the limitations of India's self-reliance ambitions. The existence of ASML, the sole supplier of EUV lithography equipment, demonstrates a structural bottleneck, proving that self-sufficiency in advanced semiconductors cannot be achieved through the policy will of a single nation alone[13]. Furthermore, with companies like Taiwan's Powertech preparing for mass production of panel-level packaging by 2027[15], Taiwan is also leading the race in advanced packaging, making it difficult for India to close this gap in the short term.

4. Key Issues

The first key issue is the gap between policy announcements and execution capabilities. While the financial scale and targets of Semicon 2.0 are clear, the 'latecomer risk' identified by Nikkei Asia raises fundamental questions about whether self-reliance is truly feasible, given India's lack of an accumulated technological foundation across design, materials, and equipment[8].

The second issue is India's positioning between the United States and China. While India needs foreign investment to build its domestic AI ecosystem, it faces pressure from both Washington and Beijing to integrate into their respective platforms[4]. This means India's push for semiconductor self-reliance is not merely an industrial policy but also a geopolitical balancing act.

The third issue is the structural constraint posed by supply chain bottlenecks. As exemplified by ASML's monopoly on EUV equipment[13], advanced semiconductor manufacturing involves choke points controlled by a small number of companies and countries. It remains uncertain whether India's goal of self-reliance can reach a point where it can circumvent these structural limitations.

The fourth issue is the signaling effect of India's policies on other third countries. Amid the tech rivalry sparked by U.S. export controls on China[11], India's actions could serve as a model for other middle powers seeking to avoid complete dependence on either Washington or Beijing. This, in turn, could become a reference case for major semiconductor-producing nations, including South Korea, as they formulate their own supply chain strategies.

II. In-Depth Analysis

India's Push to Develop Domestic AI Semiconductors: An In-Depth Analysis

1. Analysis of Root Causes

To view India's push for semiconductor self-reliance as merely an industrial policy is to miss the core issue. Underlying this drive is a lesson repeatedly learned in New Delhi's policy circles: 'technological dependence is security vulnerability.' The U.S. refusal to export a supercomputer in 1985 is the classic example. At the time, Washington imposed the restriction citing the risk of diversion for India's nuclear program[1]. India perceived this as a humiliating experience and promptly established C-Dac to develop its own supercomputer[1]. This incident left an archetypal lesson for the Indian bureaucracy: the understanding that relying on foreign sources for advanced technology leaves the country vulnerable to geopolitical leverage at any time.

Forty years later, this same perception is replaying in the realm of AI semiconductors. As the U.S.-China tech rivalry intensifies, export controls on advanced AI chips, including Nvidia's GPUs, have become a standard policy tool. In this dynamic, India is positioned as a third party, aligned with neither the United States nor China. The CFR notes the possibility that India will face simultaneous pressure from both sides to integrate into their respective AI platforms[4]. From New Delhi's perspective, developing indigenous AI chips is a strategic choice aimed at avoiding this binary choice altogether. The determination to not become subordinate to any single great power's supply chain is the fundamental motivation behind the policy.

Economic factors are another root cause. India's semiconductor import bill is projected to swell to $240 billion annually by 2035[8], a level that could place a significant strain on the country's current account balance. This pressure for import substitution is a key reason behind the Modi government's decision to invest 1.28 trillion rupees in the Semicon 2.0 program[5][8].

2. Structural Context

Political Structure

For the Modi government, semiconductor self-reliance is directly linked to the domestic political narrative of 'Atmanirbhar Bharat' (Self-Reliant India). Minister of Electronics and Information Technology Ashwini Vaishnaw's target of launching an indigenous smartphone brand by mid-2027 fits within this same context[12]. The localization of the semiconductor and electronics industries transcends mere industrial policy, carrying political symbolism tied to Hindutva nationalist sentiment. Consequently, there is little disagreement between the ruling and opposition parties on the need to sustain this policy. However, the government has shifted away from its past approach of large-scale direct subsidies. Under Semicon 2.0, it now requires a minimum private capital investment of 200 billion rupees[5]. This change can be seen as a result of both political pressure for fiscal prudence and lessons learned from past experiences where government support led to delayed outcomes.

Economic Structure

The structural vulnerability of India's semiconductor industry is that while it has considerable fabless design capabilities, it has virtually no actual manufacturing (foundry) capacity. The Tata Group's construction of the country's first commercial fab is an attempt to fill this void, and the government is now seeking investment from global firms to attract a second fab by 2031[5]. The challenge is that the semiconductor industry demands long-term capital accumulation and a steep technological learning curve. Nikkei Asia describes this as 'latecomer risk,' highlighting the significant gap between policy announcements and actual mass-production capabilities[8]. This points to a structural limitation: it will be difficult for India to quickly catch up to the process know-how and equipment ecosystems that South Korea, Taiwan, and China have built over several decades.

Security Structure

India's move toward semiconductor self-reliance is aligned with its efforts to localize technology in the cybersecurity domain. The Central Electricity Authority's (CEA) roadmap for developing indigenous SCADA systems for the power grid is driven by vigilance against cyberattacks originating from China[16]. This demonstrates that the Indian government views semiconductors not as mere industrial components but as strategic assets directly linked to the security of the nation's critical infrastructure. AI chips are viewed through the same lens. As the use of AI in defense and security expands, the reliance on foreign-made chips for the underlying computing power is itself considered a security risk.

3. Comparison with Historical Precedents and Similar Cases

India's current actions show several common patterns when compared with similar cases in other countries.

The first point of comparison is China. In response to U.S. export controls, China plans to increase its self-sufficiency in advanced-node semiconductors to 66% by 2035[13]. However, analyses suggest that even China is unlikely to overcome the bottleneck posed by ASML's EUV lithography equipment[13]. This serves as a reference case, suggesting that even as India pursues self-reliance, its dependence on specific bottleneck technologies for cutting-edge processes will likely be unavoidable for some time.

The second comparison is the trend of big tech companies developing their own chips. Google, Amazon, Microsoft, Meta, and OpenAI are all developing custom AI chips to reduce their reliance on Nvidia's GPUs[10]. Anthropic reportedly considered a $7 billion acquisition of MatX to secure its own silicon supply[14]. While these cases represent a pursuit of 'AI sovereignty' at the corporate rather than the national level, the underlying motivation is essentially the same as India's national strategy: an attempt to reduce dependence on a single supplier.

A third parallel is the connection to the discourse on 'sovereign AI,' as analyzed by EAI. The South Korean government has also established AI sovereignty as a key national priority, aligning with the assessment that 'when it comes to AI, not only our country but also the United States, China, and nations worldwide are now fiercely competing with a profound sense of hegemonic purpose'[6]. India's development of indigenous AI chips is an example of how this competition for sovereign AI is expanding beyond the U.S.-China duopoly to include middle powers and emerging economies.

The most direct historical precedent is India's own experience: the successful establishment of C-Dac and the development of the Param-8000 supercomputer following the 1985 U.S. export denial[1]. This precedent appears to have instilled confidence in Indian policymakers that a self-reliant response to a blockade is possible. However, the supercomputers of the 1980s and today's AI semiconductors are worlds apart in terms of technological complexity and ecosystem scale. Back then, the task was to replicate specific hardware through the concentrated investment of a single institution; today, it requires building an entire ecosystem encompassing design, foundry, packaging, and the software stack.

4. Key Variables Shaping Future Developments

The first variable is the success of efforts to attract foundries. The key question is whether the second commercial fab targeted by the Semicon 2.0 program can become operational by 2031[5]. An early litmus test will be whether a global company capable of meeting the minimum 200 billion rupee equity requirement actually materializes.

The second variable is the sustainability of India's balancing strategy between the United States and China. As the CFR notes, both Washington and Beijing will try to pull India into their respective AI platform ecosystems[4]. The extent to which India can maintain an independent path amid this pressure will determine the effectiveness of its self-reliance policy.

The third variable is securing supply chains for critical minerals and materials. The Carnegie Endowment for International Peace assesses that India is positioned to lead the de-risking of critical mineral supply chains alongside the United States and Japan[7]. Whether its push for semiconductor self-reliance can be coupled with securing the necessary back-end materials and minerals will be a key determinant of the policy's overall success.

The fourth variable is the pace of expansion for power and data center infrastructure. Boston Consulting Group (BCG) notes that the race for AI computing is effectively a race to secure electricity for data centers[3]. Even if India produces its own semiconductors, this will not translate into a meaningful enhancement of its AI capabilities without a supporting power grid and data center infrastructure. How India's efforts to develop an indigenous power grid[16] align with this variable will be a key point to monitor.