Driving Indias AI Future Growth Innovation and Impact

So this conversation here and the couple of conversations we are going to have over the next one hour or so are aligned with the summit’s goal of bridging the global AI divide. So AI drives economic growth, social empowerment, and of course, global leadership for India. This is not just a presentation, it is a call to action. I’m your host, Mridu Bhandari, senior anchor and consulting editor at Network 18 with brands like CNBC and Forbes India, and I’ll be guiding you through this next 55 -minute journey that we’re on. To set the tone of this morning, we’re going to begin with framing the execution pathway of AI adoption and scaling it up from an industry vantage point.

Our leadership keynote theme today is architecting India’s AI leadership, a blueprint for transformation. To deliver this knowledge, we’re going to be talking about the key points of AI adoption and scaling it up from an Please join me in welcoming on stage Dr. Vivek Mohindra, special advisor to the vice chairman and COO of Dell Technologies Global. Dr. Mohindra, please join us here.

Thank you, Mridu, and thank you, everyone, for joining us for the unveiling of this important blueprint. As we have heard over the most of this week, India is at the cusp of very significant changes and progress on the back of AI with very bold aspirations, which are not only bold for India, but they’re very bold when you put it in the context of global aspirations that lots of other countries have. Dell has had a presence in India for over 30 years. We have partnered very closely with several government agencies as well as companies and the broader ecosystem to bring the broader set of capabilities that we have, which are across the board. We have a board covering server storage, networking, PCs.

and we are the number one AI infrastructure provider to enterprises globally. So leveraging our global presence and leveraging our deep knowledge of India, we have put all of that thought into putting forth, as Milu described, an AI blueprint, which is a practical guide for what we think not only the country but also companies need to do to be able to take advantage of this particular opportunity. The growth in terms of compute expected on the back of AI in India is expected to be well greater than 10 exo -flops, and that is a significant amount of growth. And the AI workloads in India are growing at over 30 % compound annual growth rate over the next few years, which is extremely significant.

So as we step back and look at what are the key elements of what a country and companies need to do, there really are three key elements. The first element is investments. And the investment really goes at the heart of the compute infrastructure that a country needs to put in place to ensure that everybody has access to that infrastructure, including MSMEs who sometimes do not have the capacity to be able to put their infrastructure in place. Investment also includes energy infrastructure, because without energy, there is really no compute infrastructure you can put in place which can run on that. So those are some of the key areas of the invest pillar of that. And there are other several other areas that I will encourage you to read through our blueprint that you will see both from a policy perspective and practical perspective that we think needs to get done.

The innovate side really comes down to. Areas like skilling, which I know when Minister Chaudhry joins us, we will get into that in quite some detail. But innovating around how the skilling occurs all the way from schools to colleges to workforce entering employment and employers themselves, what role they play across a whole spectrum of mediums to deliver that skilling is a key part of the innovate pillar. And then the last one, evolve, revolves all around governance aspects of that. And governance covers multiple areas. One of the key areas within governance is fundamentally the regulatory framework that needs to exist and that countries need to put in place. The pace of change is so significant with AI and how rapidly the technologies are moving that one of the fundamental balances that countries need to strike vis -a -vis, regulations, is striking the balance between innovation and responsibility while anchoring it to responsibility.

That is one of the key. regulatory principles that needs to be in place. And the regulations have to be agile because the technology is moving at such a fast pace that you cannot anchor the regulatory framework to yesterday’s technologies. And at the heart of it, I hope what you will take away from our blueprint is realizing sovereign AI potential for any country, including India, is really about the public -private partnership. And it’s really about marrying the public resources with private innovation. And that really is what the key to unlocking the full potential of AI and sovereign AI is in this country. So, again, I would encourage you to read through the blueprint, and we look forward to your feedback, and we look forward to partnering closely with Indian ecosystem to help India realize its aspirations with AI.

Thank you very much.

Thank you so much, Dr. Mohindra. I’m going to request you to please stay back on stage. I’d also like to invite Manish Gupta, President and Managing Director of Dell Technologies India, to join us here. This is the big moment, ladies and gentlemen. We are ready for the unveiling of the Dell Technologies Blueprint to accelerate India’s AI growth. Yes, that’s a photo moment for everyone. Thank you. Thank you so much, gentlemen. Thank you, Dr. Mohindra. Thank you, Mr. Gupta. Well, this blueprint advances India’s vision of Vixit Bharat 2047, positioning AI as a foundational engine for productivity, modernized public services, opportunity expansion and strategic autonomy. It centers on three pillars that we’ve been discussing. Invest, invest in sovereign, scalable compute and data foundations, innovate with collaboration and with a future ready workforce and evolve, evolve into a responsible, agile, security first governance structure.

So our next panel today will go inside this blueprint and India’s AI future to unpack how to convert this ambition. into nation scale execution. And that’s quite a mean feat for a country as diverse and as huge as India. So let’s welcome the panelists for all the tough questions this morning. Raj Gopal, AS, Managing Director and Chief Executive Officer of NextGen Cloud Technologies. Bhaskar Chakravarti, Dean of Global Business, the Fletcher School of Law and Diplomacy at Tufts University. Please have a seat, sir. And once again on stage, Manish Gupta, President and Managing Director of Dell Technologies India. And I will be moderating this session for you. Welcome, gentlemen. We are here this morning to really translate the invest, innovate and evolve pillars into very actionable steps that we all can take together to grow India’s AI ecosystem.

So I’ll begin with targeted questions to each of you. And of course, you are free to jump in to add thoughts to each other. It’s a candid, free -flowing conversation. Mr. Rajgopal, if I can start with you. So startups and MSMEs, they are the engine of our economy. They are also the engine of our innovation, especially as far as AI is concerned. But access to very reliable, affordable AI compute and cloud at scale continues to be a barrier for many of the small and medium enterprises. Now, in your opinion, what are some of the policies, some of the infrastructure, some of the market interventions that we need today to really unlock this access at scale?

A. S. Rajgopal:

Yeah, actually, I think across many other countries that we have seen, India has got a much comprehensive approach to this. I mean, in terms of actually they started this India mission, which is across seven pillars. I actually don’t see many startups actually using the facilities that are there in the sense that you could, you know, you apply for GPU infrastructure and you would get it at a subsidized rate. And some of them even got 100 percent of the GPUs that they need. So I think from India’s side, we’ve got a little less number of GPUs compared to what we really need. Maybe we need about 200 ,000 GPUs now, and we have about 40 ,000 to 50 ,000 now.

So we all need to really invest more and then deploy more. But the most important thing that you should see is that there is a good ecosystem. There is a system policy available for MSMEs and startups to leverage this. So there are a lot of innovative AI solutions being built. Most importantly, I also see that government actually setting pace in terms of actually leveraging some of these. We ourselves, I mean, we did one job for government, which is very, very unique. Like we serve Election Commission of India. They came to us and said, can you deduplicate and look at all the photographs that we have? This is like, you know, 90 -year -old pictures, right?

I mean, so. So humanly it was not possible to deduplicate. You can’t check one photograph with 90 crore others. We did that in a matter of 51 hours, and then we responded to them as to whether they had complications and all that.

Wow, that deserves some applause. That’s a humongous task.

A. S. Rajgopal:

So what I see in this country is that I don’t think we will be pure play, this chatbot, I mean, the generative AI, the way it has been envisaged, I mean, will be the primary use case. I think we are going far beyond what AI can be applied in terms of actually improving the productivity of citizen services and also give use cases that these small and medium enterprises can actually use. In terms of actually enabling more GPUs, I think we have, you see, we need a lot of money. I mean, we are investing about one hundredth of what U.S. is investing or even less. so for us to do more I think we need to remove certain bottlenecks that are there one of the things I believe that can be done is I’m sure everybody is familiar we all pay GST but basically when we import servers when we pay GST on it and then when we deliver service I get that as an input and we do the I mean we only pay the value -added piece back to the government but the government gets the GST either way and I get an input so the way to one thing that we could look at is whether we can wave off the GST up front and just take the GST when the services are being delivered what that would do is it will deliver it will reduce my upfront infrastructure cost to by about you know you know 18 percent you know I don’t have to fund that up front and then pay interest on it you know or raise equity and you know pay more expectation on that I mean these things could really help I mean there are some of these things that government should look at and I think that’s a good point.

Thank you. The last point is that they’ve given tax holiday for delivering services to the world, but I think India has got a lot more to do within India than just look at world market. And I believe that we should get, Indian service providers should get the same benefits as the global providers would get when they host services in India. So maybe a GST waiver and some income tax benefits could be good.

Okay. GST waiver, income tax benefits, demands from the industry coming in. Professor Bhaskar, if I can come to you now. You’ve often argued that nations need to compete not just on technology, but on trust, on institutional strength, and of course, very, very inclusive digital participation. And that’s very critical for a country like India because we are a country of many countries and the rich -poor divide is quite huge. There are a lot of bridges. to gap here. There are a lot of gaps to bridge here. My apologies. Now, as India accelerates AI, what do you think are some of the biggest non -technical bottlenecks that we should be looking at addressing, which you believe could really, you know, limit this momentum that we are on currently?

And if we need larger societal good, what are some of the non -technical barriers that we need to immediately resolve?

Yeah, so thank you. Thank you for the question, and thank you for the invitation to join this terrific panel. I think the issue of what are the non -technical elements, it’s great that you have included that question in this discussion, because in all the excitement around the technical infrastructure, which is, of course, enormous that’s happening right here in India, it’s no secret that, you know, this is one of the biggest talent pools. In the world, growing very, very fast. In two years, one in three developers are going to be in India, the largest mobile data pool anywhere in the world, the third largest data pool, you know, anywhere in the world once you take mobile and everything else together.

Growth in compute, growth in energy, growth in workloads, all that is happening, you know, which is fantastic. Now, when you think about what are the other elements that drive demand, what we have found, we study 125 countries and try to understand the role of technology in shaping lives and livelihoods across 145 countries. The single most important determinant of what keeps a country on trajectory in terms of both the momentum of growth and also the state of their digital evolution is the demand side. So when I think about the demand side, obviously the core infrastructure, which has been talked about, is enormously important and is going to continue. So the demand side is going to continue to be a major contributor.

A second part, which has been talked about a lot in the Indian context, is the distribution infrastructure. With DPI and all the different platforms associated with it, we know that there’s a very powerful distribution system. Now, there’s a third infrastructure, which is the non -technical part, and that is what I would call the trust infrastructure. Now, when you think about trust, it’s a bit of a slippery concept. It’s very hard to define. Each one of us in our heads has an idea of what trust is. But if you force somebody to define it, we’ll struggle. The best thing about trust is I know what trust is when it is not there, when it is missing.

And then you have to ask the question from a human perspective, what really is trust? And how do I bake that into the policy systems, into the technical systems, into the marketing systems, into the narrative around the India ecosystem, which will then keep moving the system. And trust ultimately has to do with the the Do people have confidence, the people who are the grantors of trust, do they have confidence that this invisible transaction that I’m engaging in, whether I’m putting my data into a system or whether I have entered my financial information and I’m expecting something on the other side, that this whole thing is going to be completed and it’s going to be completed in a way that is reliable, that is repeatable and will not take advantage of me.

So when you think about that whole trust ecosystem, India starts in a great place. Relative to where I come from, the United States, India is a far more trusting country in terms of trust in digital systems overall and certainly in terms of AI. There’s a tremendous level of enthusiasm in terms of embracing all things AI. And we’ve seen this right here, just the sheer numbers of people who’ve attended the conference. It shows a level of trust that is probably unmatched anywhere in the world. Now, this is a tremendous asset. That’s it to start with. The challenge is that the institutional side of trust is still in the development process in India. So if you think about data governance, privacy, security, we are making progress, but we need to be much further along.

Other aspects of trust has to do with the fact that, say, the India AI mission, that is developed at the union level, at the center level. But the actual exercising of trust, the granting of trust happens at the district level. And the district varies depending on whether I’m in Telangana or I’m in Jharkhand. And at the union level, the principles that I’ve got in place need to be sensitive to how it’s being experienced from the ground up. So there are many different facets of trust that we need to work on and put in place, including transparency, which AI is, you know, we are facing a challenge regarding transparency across the world. So this is an issue.

And then, you know, an approach to having redress and grievance systems and then literacy. You know, people need to be able to understand how to use this exciting technology and also protect themselves. I’ll pause.

Absolutely. Thank you for that wonderful perspective, Professor. Coming to you, Manish. Now, the Dell Technologies blueprint really calls for tighter alignment between policymakers, between industry, academia, institutional capacity. How can frameworks like this one really ensure growth that is both globally competitive for India, but also locally inclusive? Because there is a lot of regional growth. There is a lot of, you know, geography that needs to be taken into the fold of AI when we are talking about being globally competitive as well.

Thank you for that question. And, you know, just before I go in there, I would just add or maybe, you know, speak on a couple of topics. That professor just spoke about your trust, while he spoke more from a non -technical standpoint, you also brought in a technical aspect to it, you know, around the entire governance. And it’s not just non -technical in today’s world. It’s driven by data privacy and all of the things around. But equally on literacy, there’s also explainability. You know, that trust comes really inherently once you have got explainability and people are aware on what outcomes are coming. There’s the is that explainable and do they understand that? Right. So it’s a very, very interesting world that we are in.

Now, back to the blueprint that we were talking about and Vivek articulated that beautifully well across the three pillars of invest, invest in in data centric capacity, invest in energy infrastructure, invest in people, which goes back into the innovate side of it. Because like we like like we just discussed, we’ve possibly got the largest pool of engineers around AI. And that’s that’s killing on innovate part is what’s going to differentiate us, what’s going to make it really real and practically doable within the industry and would differentiate us versus other nations equally make it make it make it more democratized and ubiquitous across the nation. And lastly, it’s really about how do you continue to build in the guardrails?

How do you? Build the trust like we just discussed to ensure that there is that. the ecosystem knows that this entire process can be trusted and can be built upon. We’ve also got to remember the sustainability aspect of it, you know, and which is where, as you look at the blueprint, you will see us talk about the fact that energy efficiency, sustainability on data centers and new architectural models are becoming super important. And that’s something that NextGen under Raj Kapoor’s leadership has demonstrated in building highly sustainable, more energy efficient data centers that will allow us to use our energy resources in the best possible manner while democratizing the access of compute capacity, while democratizing the access of data center capacity to organizations and verticals of various sizes.

So that’s going to be the critical pillars around which we really believe that there’s practicality in adopting and differentiating ourselves on the AI arena. You know, as we go forward. Right.

I’m going to pick up on that data center piece and come to you, Rajgopal. Now, given the scale that India will really need for competing globally, what would it take to truly build sovereign, cost -efficient AI infrastructure that’s not just available to large enterprises, but is also very, very affordable for the long tail of innovators that we have in this country?

A. S. Rajgopal:

Yeah, if you see the data center industry, it’s been pretty concentrated in Mumbai and a little bit in Chennai. And, you know, the other markets didn’t really take off as much as they should have. So what we are trying to do is to have, I mean, our current plan is to put about 100 megawatts of data centers in about six states. And what I see is that going forward, I mean, this could be the model that can be built on where each state has got a capacity. Because these states itself have got so much of consumption that can happen primarily because. if you start looking at applying AI into elevating the quality of education in India, which will be one of the first things that will get rolled out at scale in India, and also the healthcare aspects of it and citizen services, these things can require lots of computing capacity.

So we are working with few state governments to actually see if we can bring a total transformation, you know, actually consolidate their applications, bring about a data lake, and then apply data to it, sorry, intelligence to it, and take it to masses. So the way I think it will evolve is that there will be many more regions where data centers would spring up, and then when they’re distributed, they need to be interconnected. We have very good interconnect system, and not just the telcos, but you have things like, you know, we have railway networks, we have power networks which can actually assist the good connectivity between them. When these things come into play, actually, we can have a pretty distributed, a good amount of compute in India that can actually serve this aspect.

But you must be aware that, you know, this game is actually, I mean, if you see my context, I mean, I have four diamonds to work on. One, of course, is geopolitics. I mean, there is quite a lot that is happening. We need to ensure that we have access to the technologies that we want to bring to India. So that India actually works on the best available infrastructure. It is slightly better now, but, you know, there were restrictions before. The other aspect is the amount of money. I mean, so I think it requires multiple billion dollars of investments to do. And that should be facilitated. That should really come into the country. When we have the money piece sorted out and then we build the infrastructure, a very good, you know, a very good, you know, a very good, you know, a very good, you know, a very good, you know, a very good, you know, a very good, you know, a very good, you know, a very good, you know, But the good thing that we can leverage is open source.

I mean, when we leverage open source, we can actually combine the infrastructure and open source and bring down the cost of compute so much that it is actually palatable for the Indian citizens to use because it’s not about serving this 2 percent, 3 percent of the population which pays the income tax. It’s about serving the 90 percent others. The moment we succeed in doing this, I think the talent also, the good talent that we have, we have access to good talent in India, but we don’t have good talent. The quantity is missing. I mean, so we have good people, but, you know, you need many more good people. They’re going all over the world. We want to bring them back.

These people can come back when this money and infrastructure fall in place. Right. That would ensure that India is playing a role which is actually pretty balanced, leveraging the global technologies and leveraging local talent and actually setting the standard for the future. It’s a blueprint for all the other countries which don’t have. which may miss out this revolution and become digital colonies of the top two countries that are investing heavily in AI. So I personally believe we have a good blueprint, and the blueprint can be applied to multiple countries, and we are well on the path. And I would prefer a distributed development of data centers across the country so that we are closer to the users and we

Absolutely. Absolutely. Well, Professor, coming to you next, studies on digital competitiveness have consistently shown that institutional capacity often determines whether technology adoption really translates into economic value or not. Now, in the Indian context, what do we really need to do to really strengthen institutions and to really build the institutional muscle here? That ensures that AI drives very, very inclusive growth rather than deepening the tech divide, because there are already many divides that we are battling with.

and I can then end up with a solution to the problem. So the same thing for skill building, literacy. If I can see my ability to speak, my ability to read in multiple languages improve, you know, suddenly my trust goes up. So what is the minimum amount of institutional safeguards I need to provide that? Then I come to something like health care. When people have a much bigger chasm to cross, that’s where people have a lot of concerns about, you know, should I be putting my information into the system? How is it going to be used? Can I trust the phone where I’ve relied on a doctor or relied on, you know, a wise person in my community?

I’ve relied on my mother, you know, for maternal health care advice. So how do I cross that chasm? You know, being able to provide the foundational trust elements is going to be important. So the answer to your question, the long answer is it depends. It depends on the user. As is the case to a lot of questions about India, it depends, right?

Well, Manish, you know, globally, we are seeing nations tie AI strategy to strategic autonomy. Now, whether it comes to the semiconductor ecosystem or the supply chain, you know, the strategic autonomy is becoming extremely important for countries. For India, what are the two or three foundational capabilities that you think we need to build domestically in this decade to ensure that we are true creators of AI value and not just consumers?

Awesome. Great question, Midu. And, you know, we as a nation have proven ourselves to be phenomenal adopters of technology. You know, and the best example in my mind comes on UPI or digital payment. Ten years back, 11 years back, we were just not there. and today we are by far the largest consumers or the number of digital payments and the value of digital payments within India is multiple times of the second economy that does this, right? So that’s a great example of how we have been able to localize, democratize and proliferate the use of technology. So within that, I would really put on three hats here. The first and just, you know, inverting the pyramids, not starting with technology but starting with people.

We have really got to think away from the users to the developers. You know, it’s got to move from 1 billion users to 1 billion or 1 million or 10 million developers and that’s the skill set, that’s the IP that we are going to bring in because we’ve got that largest talent pool residing within the country. The second, and again, I heard you talk about semiconductor and supply chain. I think we have got to adopt the best that’s available. They’re globally. But equally, we’ve got to move, we’ve got to not just think about made in India. but talk about trusted in India and which is where work with organizations such as AISI, Artificial Intelligence Safety Institute to ensure that we are putting the right guardrails, putting the right governance policy and the entire institutional framework to ensure that the AI that we are building here is trusted.

And lastly, goes back to the same thing. And maybe I’ll use the same example. You know, we had the UPI of money. We need to have UPI of AI. Where we are, we are building that at scale using the data sources that we have. We have the largest ones and some of the initiatives that the government has taken, India AI mission. But equally think about AI Kosh. There are more than 7000 data sets that are now available to organizations of all sizes. Use that to ensure that we are developing for the country at the population scale through academia, through private sector, through startups, through MSMEs, all coming together. And that really represents. Be quiet. a consistent API layer that’s bringing theoretically, maybe even all of the data center and the compute capacity that we are creating as a part of AI mission to be one single layer that can be consumed by anybody and everybody across the nation to start to innovate on that, to start to develop on that.

So going back, I know a long answer, but if I were to summarize three things there, UPI of money to UPI of AI. You know, made in India, transitioning to trusted in India. And, you know, from a developer users to from a billion users to maybe a million or 10 million developers.

Made in India, but made for the world.

Absolutely.

All right. So I’m going to ask each one of you for a few concise takeaways today. Now, the blueprint that we are talking about that, you know, Dell Technologies has just unveiled talks about agile, trusted AI governance with sectoral baselines, testbeds, strong institutional coordination. Yet globalization, only what we’ve seen is that speed often beats cost. caution. And we are seeing, you know, some of the scary stuff coming out with AI as well. A lot of the experiments, a lot of, you know, agentic AI experiments that people are doing across the world, some of them call for caution. Now, in the Indian context, how do we stay globally competitive while also operationalizing very, very stringent safeguards?

And where should we really draw the line between the speed of innovation or innovation velocity and regulatory discipline? So, Rajgopal, if we can start with you,

A. S. Rajgopal:

please. If you take the birth of Gen AI, I mean, it actually, I don’t think none of those rules were actually followed. And, you know, it was built on every data that was available in whatever form. Personally, you know, in most places, I’m actually trying to tell people, just it’s not about ignoring the risk factors of it, but I think the regulation should not curtail the innovation in the thinking that you know we should be restrictive about whatever we are working with and so one of the most important things is i think we should have less regulation in this place because i think overall the benefit you should look at ai like a utility i mean you will have more good with ai than with bad yes there are things that can be handled as we go along if you see what we do in cloud today we haven’t been able to sort out our security and you know data protection postures even today it’s an evolving journey and i think we will continuously catch up with the bad factors around the ai adoption and and that’s a journey it cannot start or stop or it can be implemented at a point in time so we should keep looking at those aspects and keep putting whether regulations or technology interventions to ensure that we handle the problem and we should keep looking at those aspects and keep putting whether regulations those pieces but we should go fast forward with implementation and adoption of And I see a lot of Indian enterprises really being reluctant in terms of adopting it, especially the larger ones.

But if you see in India, I think government will set the pace and the startups and the MSMEs will actually catch on from there. And the large enterprises will actually struggle to catch up with the amount of innovation that’s happening in these

Where does that reluctance come from? Like, what are the top three reasons large organizations are reluctant? One is, of course, the fact that it’s not easy to adopt and transform a large organization. And perhaps startups and SMEs have the benefit of the agility and the small scale that they’re at. What else?

A. S. Rajgopal:

So I think the first issue is not about security and all that. I mean, about these aspects. But most importantly, I think a lot of people are struggling to imagine where to apply AI. And I think the moment we can understand that, you will start seeing that. that the benefits far outweigh the negative aspects of it. So I think that’s the first thing that people should look at, is to not just look at leveraging the Gen AI in its chatbot form, but actually really look at where you can deploy. I talked about that deduplication piece. We are working on more than 150 projects, and not all of them are bot -based. So that imagination is what is important, and once that imagination comes, the benefits will outweigh the negative aspects of whatever we

All right. Professor, final takeaway from you on speed versus caution.

Yes, so if you think about speed, I always like to use the analogy of a car and a road. you can think about the speed that you can build into the car, the velocity of the Ferrari. And a lot of the conversations that have happened, not necessarily in this room, but in other rooms, is about the Ferraris, whether I’m talking about agentic AI or AI optimized for certain applications and the technical aspects, you know, really, really important. Now, if you take the Ferrari and you bring it into the Indian context, maybe it’s a Maruti or something else that I need to be talking about. But then the question is, what’s the road on which this Ferrari is going?

If it’s a dirt road full of potholes, even a Ferrari is not going to go very fast. So much of our conversation here is about that dirt road. And what are the things, what are the potholes that we need to fix? There’s one elephant on the table that we did not address, and I’m just going to leave it at that, which is when we talk about trust, there’s a whole bunch of things you can do from an institutional standpoint to build trust, transparency, explainability, and so on. There’s a huge issue that we need to think about, which is what is going to be the impact on jobs? What’s going to be the impact on jobs?

This is the youngest major country in the world. It’s also one of the least employed country in the world. And now with AI coming in, is that going to help boost jobs or is it going to take jobs away? If we don’t fix that problem, get ahead of that, all the trust we are talking about, all the institutions you build could come down. So part of the policy infrastructure here is to figure out what is the post AI jobs picture.

Absolutely. Manish, final word to you.

So, you know, I honestly don’t think that these are opposing forces. Agility versus security. And, you know, particularly in this in this side of technology, you cannot have them act as opposing. It’s really about building the frameworks that are going to take both of them together. This is fast evolving as a technology, but equally as as as institutions will have to be faster than that in evolving. I think the government has done a phenomenal job in building some of the frameworks around that, you know, and the and the institutions, the ASI as one example. on the privacy side, DPDP or DEPA, all of those acts being there, are good frameworks to start with. And I’ll just index back on the question that you had asked Raj Gopal earlier on what is the hesitation from enterprises in adopting.

I don’t think it’s necessarily about security. You know, it’s really about saying how many of those have real use cases? While the real use cases exist, how many of them are able to monetize or are able to see them scale from experimentation or pilots into production? And I think that’s a job that we as industry folks who understand the technology, who are innovating in this space, really need to bring to the table so that we can bring this to the fore across the nation and enterprises and organizations of all sizes and academia and public. I think that’s where this will get practical, but equally these are not opposing

Right. Well, thank you, gentlemen, for that absolutely incredible conversation. You know, the takeaway is clear that investing, innovating and, of course, innovating by expanding skills pipelines and accelerating AI deployment is going to be key to India’s sovereign AI infrastructure. And appreciate you joining us here and taking the time today. We are also very delighted to now be joined by Honorable Shri Jayant Chaudhary Ji, Minister of State for Education and Minister of Skill Development. Huge round of applause. We are going to have him up here shortly. Thank you, gentlemen. Thank you very much for joining us. So if we can have you up here for a quick photo op and we will then continue the conversation.

Thank you. Thank you. Thank you, everyone. Thank you. Thank you, gentlemen. Manish, if I can please request you to felicitate our speakers. Mr. Rajgopal. Let’s have a huge round of applause for our panelists here today. Professor Bhaskar Chakravarti. thank you so much for joining us here today if you all can just get off the stage for two minutes we are getting it ready for our next conversation thank you so much well ladies and gentlemen time to move on now if India’s AI ambition is to translate into real economic growth it’s obviously not going to be any one entity’s job it is not going to be driven only by the government or by the industry alone it will be driven by partnership India has the talent the digital backbone and the momentum the real question though is how do we scale AI responsibly securely and inclusively so our next fireside chat conversation will explore what is the role of AI in the development of the future what a powerful public -private model regarding AI could really look like.

And for this, I’m delighted to welcome two very eminent leaders who are instrumental in shaping the journey, both from policy and industry perspectives. We have, of course, Honorable Shri Jayant Chaudhary Ji, Minister of State for Education and Minister of Skill Development and Entrepreneurship, Independent Charge, Government of India. And we have Dr. Vivek Mohindra, Special Advisor to the Vice Chairman and COO, Dell Technologies Global. If I can please have both of you up here for a quick conversation. Thank you so much. Thank you very much. Thank you, gentlemen. Well, it’s quite clear that public -private partnership is going to be critical to AI scaling and adoption in India. You know, Chaudhary, if I can start with you, how can PPP models really, how can PPP accelerate?

large scale AI infrastructure? What have been some of the on ground experiences you’ve seen so far? And of course, the government has been moving at breakneck speed when it comes to deploying more technology, sort of, you know, giving a more fillip to innovation in India. How are you really ensuring trust, resilience and long term national competitiveness as AI becomes very mainstream with this event in India?

In the Indian context, as the audience is aware, we had a lot of catching up to do. And it’s fair to say that a lot of what we are seeing around us in AI has been facilitated by creating an ecosystem in a short span of time. Perhaps we may enjoy a second mover advantage with regards to this technology. And that has come about only because of a strong top down emphasis and push. the only reason why this event is happening here in India is because the leadership at the top understood very quickly the value and the potential of this new technology that we should not view it as a disruption but view it as an opportunity to leapfrog legacy problems deficit problems and provide access and equity to our citizens and dignity to our workers and that’s why prime minister you know the last event in France shared that leadership space every opportunity he gets he talks about skilling about young people about the potential for AI and the enunciation in manner of that this technology needs to be human -centric I think that has given us a real emphasis and a push for the academia for our industry for our vibrant startups systems you know to really think about what they are doing in this space I think that is the background to the event that we are all witnessing.

Thousands and thousands of people, casual visitors, apart from those who are already entrenched in technology. And the message that goes out is that one billion strong young people in a developing country are already thinking about what AI means to them and what they can do in this space. Not just be consumers, but also be producers and innovators and thinkers and creators. Now, PPP in this domain for me, and when you think about Manav being human centric, citizen centric, the P that really matters is the people. And in that context, it’s important that you have the broad architecture which is open. This is something that India has stood for from the beginning. When there was a lot of debate about what should be the policy that enables AI.

But there was also a lot of fear around AI about trust factors, about privacy, data, sovereignty, multiple issues about the human interface, the augmented human worker, what this means for education, for the future of jobs. A lot of those issues were being discussed and debated. And India said that, yes, it’s good to have a strategy. And out of that strategy and experiences will evolve a robust policy. It is essential to have guardrails. But it is a starting point. Currently, we don’t want to infringe upon the possibility of innovation. And India took that approach. And we had open access to whatever compute, you know, India AI mission was set up with a target of 18 ,000 GPUs.

And in a short span, they’ve surpassed it. It’s about 38 ,000. And a roadmap is by the end of this year, it’s going to cross one lakh, threefold. now think about it all of this compute facility that has been created is a model of ppp it has to be housed in educational institutions so that real research can happen in our premier educational institutions this is a great time when academia is more important than it perhaps ever was in the indian context in the indian context academia was partly separated from industry and the real economy in our minds but now every indian citizen is realizing the value of research and innovation every family is saying that no this is important we must value it and every educational institution is saying that we are not divorced from the market and needs of our community and society and nationhood building so that engagement with nationhood building and concept of technology is deeply immersed thanks to the efforts of india ai mission and here i’ll just you know leave one data point to you what is the cost of this compute facility it’s being provided for startups, for researchers at 65 rupees for an hour.

You pay 300 rupees for a couple of hours PVR cinema ticket. So it’s probably the world’s cheapest compute facility which is open. We are celebrating Sarvam. Let’s not focus on for profit not for profit. Because everything has to be for people. If you look at Sarvam, that’s also in my mind a PPP. It’s been incubated by IIT Madras. It has been supported by AI mission. So that’s another example because you’re right. Government cannot invest in from data to energy to compute to innovation. It really has to come from our citizens, our researchers, our technologists. It’s a collective mission.

Absolutely. Well, Dr. Mohindra getting your point of view from perspective now, if we look at PPP as far as job enablement is concerned, because that’s the big concern that citizens have that what’s going to happen to our jobs. And of course, skilling is part of that journey. How can Dell partner with, you know, Ministry of Skill Development and Entrepreneurship? What are you all doing from a future skill labs perspective? How are you accelerating AI apprenticeships, so that basically the jobs also move beyond the metros, because tier two, tier three towns is where, you know, we are looking at a lot of talent sitting there, but we are looking at a lack of access of sorts when it comes to skilling?

Yeah, I think that’s a great question. And I think, Honorable Minister, good to see you again. It reminds me of a discussion we had back in last time we met in October 2024. I know we missed each other and often. And we covered very similar ground. I think at the heart of it, it does come down to, and I commend India on the progress it’s made in making access to all these GPUs available. It is an industry -academia partnership, working closely with the minister here. And our view is when you think about skilling, there are three different levels. You have to think about schooling. You have to think about college level. And you have to think about employment.

People entering the workforce are employed today. And you think about delivery of this through online, in -person, incubation. So those are the two big dimensions. And from our perspective, we are very excited to partner on extending it to Tier 2, Tier 3, working closely with the minister and other institutions in India. And the core of it, having accessibility to this GPU at such an amazing price point, really unlocks the potential. And I think as Minister and I, we have to be very, very, very, very, very, very, very, very, very, very, very, very, very, very, very, very, very, very, very, very, very, very, very, very, very, very, very, very, very, very, very, very, very, very, very, very, very, very, very, very, very, very, very, very, very, very, very, very, very, very, very, very, very, very, very, very, very, very, very, very, very, very, very, very, very, very, very,

Right. Well, finally, to both of you, now, as we embed AI into all our critical sectors, whether it’s BFSI or telecom or agriculture, healthcare, education, now governance has to move from intent to very, very strong operational safeguards. What does Zero Trust AI architecture then practically look like at a national level? Minister, if we can start with you.

Well, Zero Trust, it’s an interesting terminology. For me, the way I look at it, it means that you have to be able to verify each and every protocol in your design. And in India, we generate a lot of data. Everything, and Indian citizens are quite open about access. globally privacy has been a major concern and sometimes it becomes also an impediment for governance because those data points aren’t being collected, analyzed, researched. In the Indian context, citizens are okay about sharing their data. And I’m saying this with the knowledge that, you know, crores of upper IDs have been formed, created using consent. But we have not received any blowback. In a way, from students’ families saying that, why are you?

Because they understand that if we are able to, with technology, customize and tailor our experience in the classroom for every student, where no student can get left behind, what that means for the employability, for the knowledge acquisition for that student, for the quality of their educational experience is immense. So I think, but once you’re collecting the data, there is a lot of… effort that needs to be put in so that that trust is maintained from zero trust to 100 % trust in the public mind. So I think our data sets need to be segmented. There are protocols within Government of India. In education, we’re thinking of creating a complete AI stack, which means the anonymized data sets will be made available for researchers for creating the value for the layers of innovation, for enabling startups to engage with that data that Government of India and the citizens have shared.

Similarly, in skill, we have Skill India Digital Hub, which is also looking at creating those data sets, which can then really help us unleash the next wave of innovation and requirement that we have in skilling. So I think once you have that system design in place, it can be achieved. The prime minister spoke about a label for content. That it should be verifiable and legal. this technology and consumer awareness is a big aspect of it and how we engage with these tools and how we understand what the outcome of our engagement, how true is it? How is it verifiable? Where are these AI models trained? Is there any bias in that data set? All that knowledge needs to be out there for the consumers.

I feel also that there needs to be an audit trail for our new AI models. Maybe in the future you could have CAG come out with an audit report of all the AI models. So it’s a brave new future, but it’s a balance. For partnership at scale, you need architecture with trust.

Absolutely. Final 30 seconds to you, Dr. Mohindra.

I think the minister said it very eloquently. I would extend the notion to zero trust should extend to start with data, go into AI models. the usability, the cybersecurity elements, and the identity and access management. Those would be the ways I would extend it. And practically, it means beyond the governance framework, having things like a national risk registry, observability, being able to report whenever there is an infraction and auditability. But, Minister, you said it very eloquently, and I think our AI blueprint has more details that would be worth looking at.