Economic Survey 2025-26 Charts India's AI Path with Bold Proposals and Cautions
In a significant departure from global trends, the Economic Survey 2025-26 has outlined a comprehensive strategy for India's artificial intelligence development, emphasizing a bottom-up approach, educational restructuring, and sector-specific applications. The survey, released in New Delhi, proposes creating an 'AI-OS' initiative where the government acts as a monetary shareholder in AI infrastructure, akin to public goods like UPI and Aadhaar.
AI-OS: A Public Good for Distributed Innovation
The proposed AI-OS aims to transform artificial intelligence into a public good by establishing a centralized code repository under the IndiaAI mission. This repository would function similarly to a government-supported GitHub, enabling developers, researchers, and enterprises to share code and collaborate on projects. The government would provide shared infrastructure, standards, governance frameworks, and funding, fostering distributed innovation across sectors without stifling local creativity.
Instead of pursuing resource-intensive large language models, the Survey advocates for India to focus on application-specific small models tailored to defined sectoral needs. These computationally efficient models can run on locally available hardware, such as smartphones or personal computers, allowing for widespread innovation across firms, sectors, startups, research institutions, and public agencies without the need for expensive data center expansions.
Radical Education Restructuring: 'Earn-and-Learn' Initiative
The Survey proposes a radical overhaul of the education system through an 'Earn-and-Learn' initiative. Under this program, students would begin earning academic credits and paid work experience through apprenticeships and project placements as early as Class 11. Structured, credit-bearing industry fellowships would be co-designed by the private sector and academic institutions, with both practical experience and wages contributing to formal degrees.
This initiative is seen as a key way to unlock India's demographic dividend, drawing inspiration from international examples like China's Young Thousand Talents Program, EU industry-academia collaborations, and models such as Palantir's Meritocracy Fellowship. The Survey highlights that this approach could help bridge skill gaps and prepare students for the evolving AI economy.
Job Creation and Sectoral Mapping
The Survey calls for a comprehensive mapping of jobs outside the white-collar workspace that require high skills but are understaffed. This includes sectors like nursing, geriatric care, culinary sciences, advanced metalwork, experiential hospitality design, surgery, physiotherapy, advanced electrical work, and early childhood education. By identifying these areas, the education and skilling infrastructure can be upgraded to fill labor supply gaps and create meaningful jobs in the coming decade.
This focus on physical, human-centric, and hands-on jobs holds immense potential for job creation, potentially mitigating some of the risks AI poses to current employment opportunities.
Structural Challenges and IT Sector Caution
Despite India's strengths, including the world's second most AI-literate workforce according to Stanford University's Artificial Intelligence Index Report 2025, the Survey notes significant structural challenges. India's access to cutting-edge compute infrastructure is limited, with startups focused on training data representing only 2% of the global total. Over 70% of data centers are in high-income countries, with India accounting for just 3%.
Financial resources for large-scale model training are scarce, and private participation in foundational AI research is muted compared to global leaders. The Survey warns that India's AI infrastructure expansion could be constrained by global GPU supply chains, raising concerns about market power, technological dependence, and supply chain resilience.
It also raises a substantial question about the future of India's IT sector, as firms that once relied on India's comparative advantage may no longer need to do so. The Survey cautions that without adaptation, this could hollow out India's core value proposition, necessitating a comprehensive evolution to sustain competitive edge.
Key Trade-Offs in India's AI Journey
The Survey highlights several trade-offs India must navigate:
- LLM vs Applications: Choosing between chasing frontier-scale models or focusing on domain-specific AI systems aligned with domestic priorities.
- Labour Displacement vs Efficiency: Balancing productivity gains with employment absorption, pacing AI diffusion to facilitate labor augmentation.
- Open vs Closed AI Models: Striking a balance between openness and stewardship to ensure economic value from domestic data accrues within India.
- Compute Strength vs Resource Allocation: Weighing the resource intensity of data centers against the efficiency of smaller, task-specific models on limited hardware.
- Innovation vs Regulation: Finding the right regulatory approach to ensure safety and trust without stifling innovation in a resource-constrained landscape.
- Strategic Autonomy vs Global Integration: Preserving openness where it enhances capability while insulating critical functions from external shocks.
Overall, the Economic Survey 2025-26 presents a nuanced vision for India's AI future, leveraging its strengths while addressing challenges to foster inclusive and sustainable growth.
