New Delhi, March 6 In a bid to make access to quality datasets easier for developing Artificial Intelligence (AI) technologies, the Central government on Thursday took a big step in its IndiaAI Mission by launching AIKosha, the country’s sovereign datasets platform, and an AI Compute Portal.
Along with these initiatives, the government has also set a target of developing GPUs domestically within the next three to five years. This is part of the larger Rs 10,000-crore IndiaAI Mission.
AIKosha has been designed to provide high-quality datasets for training AI models. The AI Compute Portal will allow companies to access subsidised GPUs for research and development.
As of now, about 14,000 GPUs have gone live, with another 4,000 in the pipeline.
Earlier, the Ministry of Electronics and Information Technology (MeitY) had announced the procurement of 18,693 GPUs from selected vendors.
The core datasets for AIKosha are being contributed by various government ministries. Private entities are also being consulted to share non-personal datasets.
Union Minister for Railways, Electronics and Information Technology Ashwini Vaishnaw said that the ministries related to agriculture, weather forecasting, and Bhashini have already provided data to the platform.
"India's own AI foundational model is progressing well; we have 67 applications, of which 22 are for developing LLMs,” Vaishnaw said.
He added that the significant part of our compute capacity will be used for developing sovereign AI models.
“They will be developed at a fraction of the cost of the same in advanced economies," Vaishnaw added.
"India is to be among the top 5 AI nations during PM Modi's third term. We will have its own chipsets in the next 3-4 years," the Union Minister mentioned.
He also mentioned that the government will continue to add more AI compute capacity every quarter.
As part of the IndiaAI Mission, AI training courses for civil servants have also been introduced under the capacity building mission (CBM).
These courses aim to equip government officials with the knowledge needed to understand and implement AI-based solutions.
Disclaimer: This post has been auto-published from an agency feed without any modifications to the text and has not been reviewed by an editor