Nandan Nilekani, co-founder of Infosys and architect of India’s Aadhaar digital identity system, has called for a shift in focus from developing massive AI models to creating practical, smaller-scale solutions that address real-world problems.
Speaking in a recent interview, Nilekani emphasized that while large language models (LLMs) like those from OpenAI, Google, and Chinese tech firms have dominated discussions, the true value of AI lies in its application. “Model-building will become very common. The real challenge is making AI useful to people,” he stated.
Nilekani highlighted three key areas for AI application: consumer, enterprise, and societal uses. He noted particular interest in language-based AI for India’s diverse linguistic landscape, where solutions could help non-English speakers access services in their native languages.
The technocrat also expressed support for open-source AI development, citing initiatives like “AI for Bharat” that collect localized language data. He cautioned against over-reliance on proprietary systems controlled by major tech companies.
While acknowledging AI’s potential risks, including misinformation and deepfakes, Nilekani maintained that the technology represents a meaningful advancement if applied responsibly. He compared current AI enthusiasm to past technology hype cycles but argued this innovation has more substantial potential to improve lives.
Nilekani’s comments come amid ongoing global competition in AI development, particularly between U.S. and Chinese tech firms. His perspective suggests an alternative approach focused on practical implementation rather than sheer model size.
