New Delhi: Artificial Intelligence (AI) deployment in India should be economically grounded and socially responsive, the Economic Survey 2025-26 tabled Thursday in Parliament by Finance Minister Nirmala Sitharaman has stated.
In an extensive chapter— ‘Evolution of the AI Ecosystem in India’—the document discusses at length how AI is reshaping the global economy and outlines a pragmatic strategy in an environment marked by rapid technological changes and persistent uncertainties.
The survey has called for a “bottom-up, application-focused” strategy that prioritises economic and social impact and aligns with solving real-life problems. It made the case that India should take a cautious approach to AI in light of increased unpredictability and resource limitations, stressing that the nation’s AI strategy must be properly planned to prevent early lock-ins or regulatory overreach.
The report has paved a roadmap that suggests building coordination first and capability next, before focusing on drafting policy last.
In order to facilitate effective resource utilisation, the report has suggested concentrating on compact, open-weight models that are sector- or application-specific.
The survey has described a central challenge for India: what it builds domestically, what it sources globally, what it regulates early, and what it deliberately allows to evolve. The benefit of hindsight allows India to design AI systems that are more resource-efficient and aligned with public objectives from the outset, sequencing regulation alongside deployment. It offers the country the opportunity to pursue a more resilient and inclusive AI trajectory, according to the report.
Also Read: WIEF 2026: How global AI race is shifting & why India is better positioned than it looks
‘AI Economic Council’
The Economic Survey 2025-26 has also proposed the creation of an ‘AI Economic Council’ to guide the use of artificial intelligence across the country. Compared with governance in the United States and the United Kingdom, the document has said that India needs to develop AI laws faster in the changing scenario. The council will work independently from other AI governance bodies, with its purview expanding beyond ethical and social use of the developing tech.
The survey has further stated that as AI usage grows, regulators must exercise control over where and how it is employed. It has cautioned against excessive use of generative AI, particularly as a substitute for thinking and creativity, by students, since that can be damaging in the long term.
It has observed that AI use in universities is rapidly increasing, with many students relying on AI for cognitive activities. When combined with excessive social media use, this can impair skills such as reading, critical thinking, and analytical writing. Over time, this may deteriorate mental health, diminish productivity, and limit future work opportunities.
AI startups & data usage
The Economic Survey 2025-26, which has referred to data as a strategic resource in the age of AI, has moreover proposed a new framework to oversee how Indian data is used, shared, and monetised while preserving startups and innovation. It, however, has added that smaller companies, research laboratories, and firms developing Indian or sovereign AI models should face less stringent compliance requirements to avoid stifling innovation.
According to the survey, India has more than 100 million internet users, making it one of the most digitally dense markets in the world. With such a massive user base, India generates a lot of data. This volume and variety of data is a significant advantage for companies developing AI products for the Indian market.
So far, India has eschewed strong data localisation requirements in order to facilitate corporate operations, attract investment, and benefit from global data flows. However, the economic survey has stated that this approach should be expanded so that India can derive greater value from its own data.
Rather than requiring all data to remain within India, the survey has suggested a change towards “accountable portability.” This technique allows data to traverse borders, but enterprises that process Indian data on a “big” scale, particularly for training large or general-purpose AI models, must stay answerable to Indian regulators.
Such organisations must ensure that Indian data is auditable, traceable, and retrievable, even if processed outside of the nation.
(Edited by Madhurita Goswami)
Also Read: As tech and tariffs collide, pharmaceuticals become the next weapons in global trade