India’s relative lateness to the AI (artificial intelligence) boom is not a handicap but an advantage, according to the Economic Survey 2025-26.
“Being a late mover gives India the benefit of hindsight,” the Survey notes, allowing policy and innovation choices to be made with greater intentionality.
Early leaders, it argues, scaled AI under conditions of “abundant capital and weak regulatory frameworks” and are now “locked into a system characterised by high energy intensity, opaque development practices, ballooning financial commitments, and uncertain revenue models”.
India, by contrast, has an opportunity to “avoid costly path dependencies and unsustainable design choices that have been observed elsewhere”.
In this context, the Survey makes an ambitious proposal: An “AI-OS initiative”, where the sovereign is a monetary shareholder in the effort, similar to UPI and Aadhaar, thereby turning AI into a public good.
The aim is not to outspend Silicon Valley, but to build interoperable rails that let innovation spread across sectors, firms and states—on terms aligned with India’s economic realities.
Trying to replicate that model would impose prohibitive fiscal costs on India and lock scarce resources into an approach that is increasingly becoming unsustainable. The real strategic choice, the Survey argues, lies elsewhere: “Deploying those resources more effectively towards domain-specific AI systems aligned with domestic economic priorities.”
It said the core governance principles for such an institution would involve human primacy and economic purpose; labour-market sensitivity by design; sequencing over speed; co-evolution of technology and human capital; and public interest safeguards and ethical non-negotiables.
The Survey said the AI-OS would see the sovereign act as “an enabler and coordinator”, helping to expand structured, anonymised and machine-readable datasets in priority sectors, pool existing data centre capacity, and create shared platforms where AI development can be “coordinated and audited”. The objective is to turn AI into a public good.
This design reflects lessons drawn directly from global excess. Centralised, hyperscale compute, the Survey warns, is exposed to hardware bottlenecks, energy stress and financial risk. Smaller, application-specific models—capable of running on local hardware—offer a more resilient and inclusive path.
India’s response, it says, must be about pacing. Rapid adoption may raise output, but it can also “displace segments of the workforce faster than the economy can reabsorb them”. Smaller models, deployed sector by sector, are better suited for India’s needs.
Recognising that AI is devaluing entry-level cognitive tasks, the Survey also called for a complete overhaul of the education system. It advocated for “Earn-and-Learn” initiatives that integrate high school education with corporate apprenticeships, acknowledging that underground knowledge cannot be taught in a classroom.
“As AI applications diffuse into critical sectors and public institutions, these dependencies carry systemic risks,” according to the Survey. Yet, complete technological self-sufficiency is neither feasible nor efficient. The trade-off, therefore, lies between strategic autonomy and continued integration with global innovation networks, according to the Survey. “India’s policy choices must navigate this balance carefully, preserving openness where it enhances capability while insulating critical functions from external shocks.”
The Survey bats for India’s AI strategy to be sequenced carefully so as to avoid premature lock-ins or regulatory overreach. “The objective is to build coordination first, capacity next and binding policy leverage last, allowing institutions and markets to co-evolve.”
Read Forbes India's complete Budget 2026-27 coverage here
First Published: Jan 29, 2026, 17:25
Subscribe Now