The semiconductor revolution laid the silicon foundation for today’s digital economy. Now, AI has become the next force poised to reshape industries.
For India, which missed the train on semiconductors, the AI revolution offers a second chance. Yet, challenges like limited computing infrastructure, brain drain, and a lack in deep-tech R&D – all barriers in the previous semiconductor race – may rear their heads again.
Will India seize the AI opportunity or will history repeat itself? Will it miss the “code” just as it once missed the “chip?”
Ghosts of semiconductors past
East Asia’s semiconductor success wasn’t an accident.
Japan took the lead with government-backed research and development and strategic industry coordination in the 1970s to 1980s. Taiwan created lasting dominance through its multinational semiconductor firm TSMC’s foundry model, which now controls 70% of global foundry revenue. And while China entered late, it had a large war chest of up to US$332 billion.
Meanwhile, India – which had a promising start – faltered repeatedly.
When Fairchild Semiconductor, a founding pillar of Silicon Valley, considered setting up a packaging unit in India in the late 1950s, bureaucratic hurdles sent the firm to Malaysia instead.
India’s ambitious Semiconductor Complex Ltd (SCL), a government-owned firm that manufactures chips and was launched in 1984, initially kept pace with global standards but was derailed by a facility fire in 1989.
The critical failure wasn’t the incident itself but the eight-year delay that resulted from it – all while competitors raced ahead. This pattern of inconsistent vision, underinvestment, and bureaucratic inertia has repeatedly blocked the country’s semiconductor progress.
New frontiers, familiar challenges
The rise of AI offers India a fresh opportunity. Unlike semiconductors, AI appears more accessible, with open-source models allowing engineers in Bangalore to tinker with the same tools as their peers in Palo Alto.
The nation also enters this era with strengths: a large pool of STEM graduates, 20% of the world’s semiconductor design talent, and a thriving digital economy.
Yet, the deeper infrastructure challenges remain. Now, the bottleneck lies not in building fabs but in securing high-end computing power, which is still capital-intensive and largely foreign-controlled.
Other critical gaps persist, too.
While the IndiaAI Mission – a government-led initiative to boost the national AI ecosystem via investments in infrastructure, data, and talent – has secured over 18,000 GPUs, the country still accounts for less than 2% of global computing infrastructure.
China, in contrast, is operating at a different scale by building its own computing power network to optimize and unify computing resources nationwide.
Estimates suggest China’s overall computing capacity dwarfs India’s – China contributes to a combined 58% to 59% share of global computing infrastructure when paired with the US.
Innovation metrics tell a similar story, according to a report from the Center for Security and Emerging Technology, a US-based think tank. India ranks third in the number of AI research papers published, but it’s 15th in citations. The country’s contributions to the most prestigious and highly competitive international academic conferences are also minimal (~1.4% vs. China’s ~22%).
This means that while India produces a large quantity of AI research, its global impact and recognition for cutting-edge innovation are limited.
And despite growth in AI patents, India trails far behind China, especially in generative AI. Startup investment patterns reflect the gap: Just 5% of India’s VC funding in 2023 went into deep tech compared to China’s 35%.
The initial promise of opening up AI access risks becoming an illusion if computing power, talent, and deep-tech ecosystems aren’t strengthened.
Learning from the dragon
China’s AI strategy stands apart from the rest of the world.
For one, its intentions are explicit. The 2017 New Generation Artificial Intelligence Development Plan set a bold target: to make China the world’s primary AI innovation center by 2030.
For the country, AI isn’t just about tech leadership; it’s framed as essential for economic transformation, national security, and social governance.
China’s government is also all-in. Through a mix of direct R&D funding, tax breaks for high-tech firms, state-backed venture capital, and massive infrastructure projects, the nation is pouring resources into AI.
While exact figures vary, China alone is expected to commit up to US$1.4 trillion in AI-related investments by 2030, with US$140 billion already pledged.
And unlike laissez-faire approaches, China orchestrates its ecosystem. Private giants like Alibaba, Tencent, Baidu, and Huawei are appointed as “national AI teams” to lead key sectors.
Startups often emerge from state-supported programs, and universities like Tsinghua churn out talent and spin-offs aligned with national goals. Public-private partnerships aren’t just encouraged but systematically engineered, too.
Importantly, China is building national computing infrastructure and cultivating a robust domestic talent pool, securing vast datasets, and advancing AI applications across autonomous vehicles, healthcare, finance, and smart cities. Developing competitive domestic foundation models is a top national priority.
Still, despite its strengths, China’s model isn’t without flaws.
US export controls targeting advanced chips pose serious external bottlenecks. Internally, heavy state direction could mean bureaucratic overlap, inefficiency, and potential stifling of bottom-up, disruptive innovation.
Some metrics, like patent counts, may overstate true breakthroughs. Global trust issues around surveillance and data governance under an authoritarian regime further complicate China’s AI ambitions abroad.
Will history repeat itself?
India’s AI strategy must start with infrastructure. The IndiaAI Mission’s move to democratize GPU access is critical, but execution will decide success.
Computing power must also reach startups, researchers, and students, not just a privileged few.
This requires deliberate, well-structured action like setting clear public eligibility criteria, having a transparent review process led by rotating experts, and creating fair access policies like usage caps and queuing. The country could also publish anonymized usage data to ensure accountability and guide future allocation.
Importantly, India must address brain drain by creating globally competitive opportunities at home with better research environments, practical industry-academia collaboration, and deeper, targeted skill-building beyond basic AI literacy.
Initiatives like FutureSkills, for instance, need aggressive scaling to nurture mid-tier experts and top-tier innovators alike.
In R&D, India should break free from its low-investment cycle and strongly incentivize private sector participation in AI research. Deep-tech AI startups should be actively nurtured, not sidelined.
In addition, protecting intellectual property and encouraging international research collaborations are key to lifting the impact of AI innovation in the country.
In fact, India’s greatest AI asset may not be imitation but invention.
With world-class digital public infrastructure (DPI) initiatives like Aadhaar, UPI, and ONDC, India holds a unique advantage few others can match. DPI offers a rare platform to build AI solutions at a large scale, creating massive datasets, enabling inclusive public services, and catalyzing bottom-up innovation.
Combined with India’s large, linguistically diverse market, there’s a real opportunity to build globally competitive expertise, particularly in domains where Western models often fall short.
India’s semiconductor history offers a cautionary tale. In AI, the early signs are mixed: promising initiatives like the IndiaAI Mission exist, but infrastructure bottlenecks, talent migration, and limited deep-tech R&D still threaten momentum.
The next decade will determine whether India merely participates in the global AI race or shapes it. The window is open, but it won’t stay open forever.
