India's AI Development Strategy Proposal

Economic Survey 2025-26: Key Highlights

AI Development Proposal

• Initiative: Introduction of an “AI-OS” initiative, positioning the government as a monetary shareholder in AI infrastructure.
• Objective: Treat AI similar to public goods like UPI and Aadhaar, focusing on sector-specific applications rather than large language models.
• Centralized Repository: Establishment of a centralized code repository (similar to GitHub) to foster collaborative development under the IndiaAI mission.
• Switch to Small Models: Shift from large frontier models to smaller application-specific models utilizing locally available hardware, promoting efficiency and accessibility.

Educational Reform

• “Earn-and-Learn” Initiative: Proposal for a radical shift in educational structure allowing students from Class 11 to earn academic credits and financial compensation through apprenticeships and project placements.
• Co-designed Fellowships: Industry-academic collaborations to design structured fellowships providing both practical experience and contributions to formal educational qualifications.

Job Creation Focus

• Sectoral Job Mapping: Identification of high-skill jobs outside white-collar sectors that are often understaffed, including:
  • Nursing and geriatric care
  • Culinary sciences
  • Advanced metalwork
• Aimed at addressing India’s dependency ratio set to double in the next decade.

Structural Challenges Identified

• Limited AI Infrastructure:
  • India's share of training data is only 2%, with significant disparities compared to the US and EU.
  • Over 70% of data centres located in high-income countries; India holds just 3%.
• Financial Constraints: Scarce resources for large-scale model training, and reliance on global GPU supply chains limits domestic capacities.

Recommendations and Trade-Offs

• AI Approach: Balancing pursuit of resource-intensive large models against effectively utilizing domain-specific AI systems.
• Job Market Dynamics: Addressing the potential for job displacement caused by rapid AI deployment against the need for productivity gains.
• Open vs Closed AI Models: Striking a balance between proprietary systems and open-source flexibility to enhance domestic innovation.
• Environmental Resource Allocation: Recognizing the high resource demands for AI development and advocating for smaller, efficient models to optimize shared resources like power and water.
• Regulatory Balance: Finding an appropriate regulatory framework that promotes innovation while ensuring safety and risk management.

Global Considerations

• Geopolitical Asset: AI is increasingly viewed as a strategic asset. India must ensure that while pursuing technological integration, it also maintains strategic autonomy to mitigate systemic risks associated with over-reliance on foreign technology.

International Comparisons and Examples

• China's Talent Program: Mentioned as a comparative model for enhancing domestic research productivity.
• EU Collaborations: Emphasis on collaborations between industry and academia as key to unlocking potential in the workforce and economy.

Conclusion

The Economic Survey emphasizes a tailored, proactive approach to AI development in India, fostering collaboration of educational, industrial stakeholders and a focus on creating sustainable jobs while addressing sectoral skills shortages.