India Should Prioritize Practical AI Over Costly LLM Race, Mohandas Pai Said

Mohandas Pai said India should prioritize the development of practical AI models rather than competing in the expensive race for large language models ^[1].

This shift in strategy would align technological ambitions with the country's current economic and technical realities. By focusing on utility over scale, India could avoid the massive financial drains associated with the global LLM arms race while still advancing its digital infrastructure.

Pai said these remarks Wednesday in Karnataka during the launch of ToneTag’s eKosha platform ^[1]. He said that the pursuit of massive language models is often resource-intensive and may not yield the most immediate benefits for the domestic economy.

"India should not chase costly LLM race; instead, it should focus on practical AI models," Pai said ^[1].

The recommendation stems from specific limitations within the Indian tech ecosystem. Pai said that constraints in capital and computing infrastructure make the pursuit of the largest models a risky endeavor ^[1]. Instead of attempting to outspend global giants, he suggested a focus on AI applications that solve concrete problems.

Practical AI models typically require less computing power and lower investment than the foundational models used by the world's largest tech firms. By targeting specific industry needs, India can leverage its existing software strengths without needing the prohibitive hardware clusters required for massive LLM training ^[1].

This approach emphasizes sustainable growth over speculative competition. By building specialized tools, the country can integrate AI into governance, healthcare, and agriculture more efficiently than by attempting to build a general-purpose model from scratch ^[1].