Archive · 1 Jun 2025
IIT-Bombay Develops AI Datasets to Advance Research and Indian Languages
Imagine IIT-Bombay created a bunch of smart data sets to help computers understand Indian languages and cultures!
These datasets contain lots of sentences and words from old textbooks on topics like medicine and math, making it easier for computers to learn.
It's like giving AI a huge library of Indian knowledge.
This helps researchers and businesses build new AI programs.
They also created tools that teach practical skills like making toys from waste and understanding Sanskrit.
The goal is to improve AI and promote Indian languages.
IIT-Bombay contributed AI datasets to the government's AIKosh portal.
The datasets include over 2.18 lakh sentences from diverse subjects.
BharatGen, led by IIT-Bombay, launched various AI models on the portal.
The project aims to build sovereign AI models tailored for India.
Datasets are designed for training and fine-tuning AI for various tasks.
- Who
- IIT-Bombay, BharatGen consortium, and the Department of Science and Technology.
- What
- IIT-Bombay curated AI datasets for AI research and Indian knowledge systems.
- When
- The datasets are now available.
- Where
- On the government's AIKosh portal.
- Why
- To simplify research in Indian knowledge systems, aid AI-based innovation, and create sovereign AI models.
- Lead Institution
- IIT-Bombay
- Platform
- AIKosh Portal
- Sentences in Datasets
- 2.18 lakh
- Consortium Partner
- BharatGen
- Focus
- AI research and Indian languages
- Goal
- Sovereign AI models
