‘Let Silicon Valley build LLMs; India should lead in AI use cases,’ says, Infosys Chairman Nandan Nilekani
Meta’s Build with AI summit in Bengaluru, Nilekani stressed that India's AI trajectory should be less about competing in LLM development and more about building an infrastructure to gather and utilise data, driving meaningful innovation across various sectors, highlighted the publication."Our aim is not to build yet another LLM," Nilekani remarked, referencing the significant investments by Silicon Valley giants in creating vast language models. “Let the big players handle that.
We should instead prioritise synthetic data generation, small language models, and training tailored for real-world challenges in India."Reportedly, Nilekani highlighted data as the core component of this approach, urging Indian tech firms to concentrate on establishing robust systems for collecting and curating data. He expressed a vision for India to become a global “use case capital" for AI, where solutions are scaled with speed and efficiency, often described as ‘frugal innovation.’Nilekani’s outlook marks a distinct path in India’s AI journey.
“The Indian way is different," he noted, referring to the country’s preference for deploying practical, everyday solutions over participating in what he called an “arms race" to develop larger LLMs. “We are here to make a difference," he added, emphasising accessibility and impact over competing with tech giants focused on LLMs.The Infosys co-founder also praised Meta for making its Llama models open source, calling it a “game changer" for India’s AI landscape.
Meta recently updated Llama’s licensing terms, allowing developers to leverage its synthetic data for building diverse applications. In September, Meta released the Llama 3.2 model with multimodal capabilities, facilitating AI to interpret both
. Read more on livemint.com
