Archive · 23 Sep 2026
India’s AI Strategy Must Move Beyond the Application Layer
The United States and China are leading different parts of artificial intelligence.
The United States is strongest at creating advanced AI models.
China is especially strong at using AI in factories, robots, logistics, and other industries.
India has become very good at building software and applying technology to local needs.
The authors say this approach is useful now but could make India dependent on foreign AI technologies later.
They want India to create a short-term plan and a 10-year plan for becoming an AI creator.
They also suggest building a new research institution that can attract leading scientists.
Finally, they recommend investing in computing power, data, semiconductors, and robotics.
The United States leads in frontier foundation models, while China leads in industrial AI and robotics.
India is focusing on small language models, multilingual tools, digital public infrastructure, and sector-specific applications.
The authors warn that remaining an AI application-layer economy could create long-term dependence on the United States and China.
They propose separate short-term and long-term national AI strategies supported by parliamentary and expert committees.
The authors recommend a globally competitive Indian Institute of Artificial Intelligence and major investments in compute, data, and foundational technologies.
- Who
- The article is written by Soumya Kanti Ghosh and Pulak Ghosh and focuses on India, the United States, and China.
- What
- It examines India’s position in the global AI ecosystem and proposes ways for India to move beyond applications toward frontier AI capabilities.
- When
- The authors propose a short-term roadmap covering two to three years and a long-term roadmap covering 10 years.
- Where
- The discussion concerns the global AI landscape, with particular focus on India, the United States, and China.
- Why
- The authors argue that controlling foundational models, computing infrastructure, semiconductors, and robotics will become increasingly important for economic and technological power.
Application-First Strategy
Frontier-Capability Strategy
Near-term economic strategy
Application-First Strategy
Adapting existing AI technologies to local needs is economically sensible for India given its resource constraints.
Frontier-Capability Strategy
A strategy focused only on applications could leave India dependent on foreign foundational technologies in the long run.
India’s technological role
Application-First Strategy
India can build on its strengths in software services, systems integration, digital infrastructure, and sector-specific deployment.
Frontier-Capability Strategy
India should develop ownership of foundational models, computing infrastructure, semiconductors, and robotics platforms.
Research institutions
Application-First Strategy
India can continue relying substantially on existing institutions such as the Indian Institutes of Technology and the Indian Institute of Science.
Frontier-Capability Strategy
The country should establish a dedicated, autonomous Indian Institute of Artificial Intelligence capable of competing globally for researchers.
- United States strength
- Frontier foundation models, advanced large language models, reasoning models, and agentic AI systems.
- China strength
- Industrial AI, robotics, manufacturing, logistics, autonomous systems, and industrial automation.
- India’s current advantage
- Digital public infrastructure, software engineering talent, and AI applications.
- Short-term roadmap
- A two-to-three-year plan to strengthen AI infrastructure, talent, and institutional capacity.
- Long-term roadmap
- A 10-year plan to position India as a creator of frontier AI technologies.
- Proposed institution
- An Indian Institute of Artificial Intelligence with operational autonomy and internationally benchmarked salaries.
- Research priorities
- Public compute infrastructure, high-quality datasets, privacy-preserving data sharing, and international collaboration.