India Unveils BharatGen Project for Generative AI in Local Languages

India on Monday launched BharatGen, an initiative to make generative AI available to citizens in different Indian languages, with Science and Technology Minister Jitendra Singh asserting that it was the world's first State-funded project of its kind. Spearheaded by IIT Bombay under the National Mission on Interdisciplinary Cyber-Physical Systems (NM-ICPS), the initiative will create generative AI systems that can generate high-quality text and multimodal content in various Indian languages.

The initiative was launched by Principal Scientific Advisor A K Sood, Department of Science and Technology Secretary Abhay Karandikar, Department of Telecommunications Secretary Neeraj Mittal among others. The S&T Minister has sent a video message for the event.

Sood also launched the four thematic hubs (T-Hubs) for Quantum Computing, Quantum Communication, Quantum Sensing and Metrology, and Quantum Materials and Devices, under the National Quantum Mission (NQM).

"The hubs will be at the forefront of research and innovation, setting the stage for India's leadership in quantum computing, communication, sensing, and materials," Singh said.

The four T-Hubs have been set up in Indian Institute of Science (IISc) Bengaluru, Indian Institute of Technology (IIT) Madras along with Centre for Development of Telematics New Delhi, Indian Institute of Technology (IIT) Bombay and Indian Institute of Technology (IIT) Delhi. They comprises 17 Technical Groups across 17 states and two union territories.

Each T-Hub will operate under the Hub-Spoke-Spike model, supporting a cluster-based network of research projects (Spokes) and individual research groups (Spikes) alongside these central hubs to enhance coordination among research institutions, enabling them to pool resources and expertise.

Singh said the BharatGen initiative was the world's first government-funded multimodal large language model project focused on creating efficient and inclusive AI in Indian languages.

A key element of BharatGen is its open-source foundational models, which will help democratise AI across India, he said. By making AI more accessible, a collaborative ecosystem would be created, where researchers and developers can work together to build innovative solutions, the minister said.

The project is expected to be completed in two years along with plans to benefit several government, private, educational, and research institutions. A core feature of BharatGen is its focus on data-efficient learning, particularly for Indian languages with limited digital presence.

Through fundamental research and collaboration with academic institutions, the initiative will develop models that are effective with minimal data, a critical need for languages under served by global AI initiatives.