Maybe India Doesn’t Need Another OpenAI

After my last post I did some research on this because I wanted to understand whether India is actually building its own AI or we are mostly using models built somewhere else.

That’s when I came across Sarvam AI.

What I found interesting is that they are building AI specifically around Indian languages and the way people in India actually use them. Their models cover Indian languages across speech to text text to speech translation document AI and LLMs.

They also work on things like code mixed speech which is actually a big part of how people communicate in India.

This made me think differently about my last question.

Maybe India does not need to build another copy of OpenAI.

Maybe the bigger opportunity is to build AI that understands India better than anyone else.

Sarvam is one example that makes me think we are already moving in that direction.

But the bigger question still remains.

Can an Indian AI company take this from solving Indian problems to competing globally?