In a significant stride towards embracing AI that accommodates India's linguistic diversity, startups and tech giants have embarked on developing Local Language Models (LLMs) like BharatGPT and Hanooman. These models, crafted by Bengaluru-based CoRover and Seetha Mahalaxmi Healthcare in collaboration with IIT Bombay and other premier institutes, demonstrate a concerted effort to tailor AI technologies for India's linguistic richness.

BharatGPT, proficient in over 14 Indian languages, and Hanooman, designed for sectors like healthcare and education with initial support for 11 languages, signify a transformative step in making AI accessible to a broader segment of the Indian population.

The endeavour unfolds against the backdrop of the Indian government's advisory, requiring AI models to secure government permission before deployment. This regulatory intervention aims to ensure AI platforms' reliability, accountability, and non-bias, shedding light on the challenges posed by AI's ethical and societal implications.

Why does it matter?

Addressing AI bias, data scarcity in Indian languages, and the high costs associated with advanced AI systems becomes imperative, showcasing the complexities of navigating a responsible and inclusive AI ecosystem.

The advent of LLMs tailored for Indian languages and governmental oversight marks a new chapter in the evolution of AI within the country. These technologies, designed to bridge linguistic barriers, hold the potential to democratise access to digital services, improve governmental and healthcare delivery, and empower businesses to effectively serve a linguistically diverse customer base.

Despite the promises, the path forward necessitates addressing technical, ethical, and regulatory challenges to ensure equitable distribution of AI benefits across India's multifaceted linguistic and cultural landscape.

Moving ahead, collaborative efforts between the tech community, government, and academia will play a pivotal role in realising the vision of an AI-enabled India for all.