India’s late entry into the artificial intelligence race could work in its favour, according to the Economic Survey. The Survey notes that early adopters who scaled AI when capital was cheap and regulation was limited are now locked into energy-intensive architectures, rising financial commitments and revenue models that remain uncertain.
As the size of these bets has grown, the Survey points out that risks have also increased. In advanced economies, discussions around government backstops have emerged as possible insurance against a broader fallout if large AI investments fail to deliver expected returns.
The Economic Survey argues that India does not need to replicate this trajectory. Its comparative advantage in the AI era, the Survey says, does not lie in building frontier-scale models. Instead, it lies in application-led innovation, the productive use of domestic data, the depth of human capital, and the ability of public institutions to coordinate distributed efforts across sectors.
According to the Survey, a bottom-up strategy built on open and interoperable systems, sector-specific models, and shared physical and digital infrastructure offers a more credible pathway to value creation. India, it adds, benefits from hindsight and can learn from the experience of early movers, avoiding dependencies that are costly and difficult to unwind.
This, the Economic Survey says, allows India to design AI systems that are more resource-efficient and aligned with public objectives, rather than being driven purely by scale.
Economic Survey’s overview comes days ahead of the Budget 2026-27. In Budget 2025–26, the government allocated ₹2,200 crore specifically for artificial intelligence to strengthen India’s AI ecosystem and support centres of excellence. The flagship IndiaAI Mission received ₹2,000 crore under its multi-year plan.
Additional funds supported AI Centres of Excellence, research infrastructure expansion, and education/innovation efforts. Besides, total deeptech commitments in Budget 2025-26 included ₹10,000 crore for the Fund of Funds for startups and ₹20,000 crore focused on deeptech research and innovation encompassing AI, biotech, and advanced manufacturing.
Data centre spending and the risk of financial spillovers
The Economic Survey also flags risks arising from the physical infrastructure underpinning AI. It notes that global experience shows AI-driven data centre expansion can place significant strain on existing energy systems. The Survey references concerns raised by the CEO of IBM over the financial viability of the capital expenditure being undertaken for large-scale data centre expansion.
The Survey warns that with some firms projected to burn nearly half a trillion dollars in cash by 2030 while building AI compute infrastructure, the risk of financial contagion from an unprecedented, debt-fuelled expansion remains high.
For India, the Economic Survey highlights that power availability, access to finance and especially water remain binding constraints. Scaling compute indiscriminately, it says, carries opportunity costs, as investment in AI infrastructure competes directly with households and industries for scarce resources.
This, the Survey concludes, creates a trade-off between centralised scale and distributed efficiency, strengthening the case for smaller, task-specific models that can run on limited hardware and decentralised compute networks—an approach the Survey sees as consistent with India’s late-mover advantage in AI.
India’s data centre capacity is set to quintuple to ~8 GW by 2030 from ~1.5–1.7 GW in 2025, with under-construction and planned projects reaching ~3.3 GW by 2028. Major projects include Reliance & Digital Connexion’s $11 billion, 1 GW centre in Visakhapatnam, ARCPL ₹4,500 crore in Visakhapatnam, UPC Volt ₹5,000 crore in Telangana, Blackstone ₹10,000 crore in Chennai, AdaniConneX & Google $15 billion AI campus, and CtrlS 612 MW Hyderabad site.
