Ministry of Electronics and Technology (MeitY) has set up a high-level technical committee, which include top researchers from world over, professors from University of Berkeley and Stanford, to evaluate the 67 proposals received for building an indigenously-designed and developed artificial intelligence (AI) foundational model, Additional Secretary, Abhishek Singh shared at Storyboard18 DNPA Conclave 2025.
Speaking on India AI Mission in New Delhi on February 27, Singh also talked about the importance of AI literacy and how the Ministry is working with various government departments, including Prasar Bharati, to share datasets to build its own large language models (LLMs) under India AI mission. Singh also mentioned how public, restricted and private data sets will be available under India AI datasets platform, version of which will be launched in the coming 10 days.
The Ministry this week received 67 applications for developing artificial intelligence (AI) foundational model and building 20 LLMs, under the government’s flagship Rs 10,370 crore India AI Mission. Among the applicants were homegrown AI startups such as Sarvam AI, CoRover.ai and Ola’s Krutrim.
Highlighting how this is a tall task for the Ministry to evaluate and assess each of the proposals, Singh shared that the Ministry will complete the evaluation in a month.
“It’s going to be a very, very tall order, but we have involved top researchers from world over. We have professors from Berkeley and Stanford, and other institutions are also being involved. People of Indian origin who have volunteered to help us in assessing the quality of these proposals because we may not have the same level of expertise which is required because this is very high cutting edge research,” he shared.
In a fireside chat with Puneet Gupt, COO, Times Internet (Vice Chairperson, DNPA), Singh also highlighted that MeitY is collaborating with various Ministries and departments for collection and aggregation of datasets available to build a database with all non-personal information that can be used for the India AI datasets platform.
“We have been trying to speak to Prasar Bharati because All India Radio has very rich data sets with news broadcast in all languages. Any such content, which has the same content in different languages, spoken by different sets of people, is a very valuable data set if you are building a voice enabled services. We are also trying to get datasets from all sources, whether it’s DPIIT, Ministry of Finance, Agriculture, etc.- but right now it is all in silos and spread all over.”
Read more: Without fair compensation, quality journalism will be compromised: Sanjay Jaju
Upon being asked if India data sets platform will be available to developers across the world or just in India, Singh answered saying that every government department which is publishing the data will have the right to decide what they want to keep in an open domain, what they want to keep in a licensed domain or restricted domain, and what they want to keep in prohibited domain.
“All users can see what data sets are available. But to access the data set, they will have to register and make a request so the concerned owner of the data, who will decide whether they will keep it open for everyone, or whether they will keep it restricted.”
Further discussing the importance of AI literacy in times of misinformation, Singh shared that the government has launched a course called “AI For All (a very basic 101 AI course) available on MeitY’s portal and is available in nine Indian languages.
Additionally, the government is working with the CBSE and has introduced AI cybersecurity as a chapter in science subject. Speaking on how news media companies are also part of the AI ecosystem, he remarked that challenges are faced when the lines between the reality and AI gets blurred.
“Given at the a time when the technology evolves to such a stage when it becomes very difficult to distinguish or detect between what is real and what is not real, maybe we will need to have a techno legal way of regulation, or maybe some provisions which will be able to limit the harm that can be caused. Our regulatory approach so far has been to look at it from the point of view of user harm, and try to see how we can harness the benefits and limit the harms can that can happen,” Singh concluded.
Held on February 27, in New Delhi, this year’s Storyboard18 DNPA Conclave was themed around ‘Media Transformations in the AI Age’, and served as a pivotal gathering for policymakers, media leaders, and tech experts to examine the ongoing impact of AI.
DNPA, representing India’s top news publishers from both print and electronic media, has been at the forefront of promoting credible journalism and fostering the growth of the news industry. As AI revolutionises industries and transforms traditional paradigms, the DNPA Conclave 2025 serves as an essential forum for stakeholders to shape the future of digital media.
Discussing the India AI Mission, Singh also shared that the potential of this technology is immense with the impact it can make for improving public services, health care, agriculture, enabling educational services accessible to each.
“Under the mission, there is a focus on the compute pillar. Today we have availability of almost 18,000 GPUs (graphical processing units which are essential for building AI applications), out of which 14,000 have already been installed. So one bottleneck is solved to some extent, but we will need to do much more on the compute front.”
The second pillar under the mission is about data sets- how to make data sets available for building AI applications. For that, MeitY in the final stages of building a datasets platform, and the version one of the platform will be launched within week to ten days from today.
The third key pillar of the mission is application development. “We published a call Innovation Challenge across five themes including climate change, agriculture, learning disabilities, governance and health care. And we got 900 entries for that, out of which 30 applications have been selected for scaling up for deploying across the country. There are AI applications which can assist farmers, can help diagnose tuberculosis, can help children with learning disabilities and build up their personalised learning plans,” Singh shared.
The other key pillars of the India AI Mission include skilling, startup financing, foundation models and AI innovation center, and safe and trusted AI.