Resource inefficiency biggest obstacle for large-scale AI adoption in medium term: Eco Survey

The Survey tabled at the parliament on January 31, stated that “India’s demographic advantage and diverse economic landscape position it uniquely to benefit from AI

Even as the Indian government prioritises advancement of artificial intelligence (AI) for population-scale use cases, the Economic Survey 2025 noted that there will be several barriers including concerns over reliability, resource inefficiencies, and infrastructure deficits that will hinder large-scale adoption in the medium term.

The Survey tabled at the parliament on January 31, stated that “India’s demographic advantage and diverse economic landscape position it uniquely to benefit from AI. However, achieving these benefits requires significant investments in education and workforce skilling, supported by enabling, insuring, and stewarding institutions.”

The uncertainties surrounding the AI technology and the fact that the speed of research and development is outpacing the regulatory and ethical frameworks needed to manage its risks has been a key concern for the government.

“For AI to scale effectively, technological advancements and performance gains must be coupled with significant reductions in costs and more efficient utilisation of scarce resources—an achievement that remains elusive. The challenge is compounded by the fact that modern AI systems, still in their developmental infancy, demand enormous investment in research and development,” the survey said.

Moreover, with AI research and development currently concentrated in the hands of a few, very large companies such as Google, Microsoft, OpenAI and others, that control the resources has created high entry barriers. AI adoption in place of humans also presents the risk of concentrating the benefits of automation, the report said.

Risks for labour

The survey stated that “technological developments that worsen inequality can erode any possible benefit that the innovation brings, leaving the public sector responsible for addressing the cost of the transition.”

This has led to calls for a more responsible approach to AI adoption, especially in a country like India where employment numbers make the magnitude of impact something worth paying attention to.

In fact, the Indian information technology (IT) workforce employed in low-value-added services are the most susceptible to automation, the survey warned. IT services sector is one of the biggest employers in the country.

The effects of automation are already visible in the case of tech firms in India and the business process outsourcing (BPO) sector in the Philippines.