Uncategorized

about this morning is right -signing governance for an AI -driven world. So what we’ll try to do with a pretty excellent panel, as I’m sure you’ll agree, is talk a bit about shared computes and data initiatives that hopefully give all nations access to AI resources. We’ll look a bit at how to up -level the playing field for smaller and developing nations. And we’ll talk about collaboration in key sectors like healthcare and education and climate resilience. I’ve got a perfect panel to do that with. I’m going to introduce them all first, and then we’ll dive straight into the conversation. So unfortunately, Navreena Singh from Credo AI couldn’t be with us this morning. She’s got a meeting with the president, so she’s excused.

But we do have… What I start with, just next to me here, Bella Wilkinson, who’s a research fellow on the Digital Society Program. with the Chatham House. Next to her is Rafik Rikorian. I hope I’ve pronounced that vaguely okay, who is the Chief Technology Officer for Mozilla. Next to him, we’ve got Rajesh Nambia, who is the President of NASCOM, our sister association here in India. And last but not least, we’ve got Halak Shirastava, managed, who’s Global AI and Public Policy and Regulatory Affairs at Cohere. And for those of you who don’t know me, I’m Sabina Chofu, I’m International Policy and Strategy Lead at TechUK. So we are the sister association of NASCOM back in the UK.

So without further ado, we will start with setting a bit of a global context, and who better to do that than Isabella. So from a kind of geopolitical perspective, how realistic, I guess, is alignment on AI governance across countries with… fair to say very different strategic interests right now. And where do you see maybe multilateral institutions? I know multilateralism is not a very popular theme these days, but where do you see multilateral institutions or maybe other international players playing a role in this space? So over to you.

Thank you, Sabina. Thanks to my fellow speakers. It’s great to be here today, really keeping the energy up on the final day of the summit. We can all do it. Let me answer your question directly and then perhaps elaborate a little bit more in detail. Global consensus on how to govern AI is a no -go. It is not going to happen in this geopolitical environment. However, partial alignment on priority issue areas is possible, and it’s pragmatic to throw our weight behind these smaller gatherings that we can then scale using the multilateral format. Now, let’s take a second. Let’s take a second to sketch out the state of play. We have some great experts in the room, on the panel, so I won’t spend too long doing this.

We have been absolutely covered in really optimistic summit rhetoric, walking into Bharat Mandapam, going to side events over the course of this week. But despite the optimism outside of these walls in the background, the US -China AI race continues to accelerate to the umpteenth degree. The capabilities of advanced and the most frontier AI systems and models, the little we know about their capabilities, mind, with huge gaps in transparency, continue to advance. And global scientists only recently have issued warnings about the state of the science and the intense uncertainty surrounding these capabilities and the impact they might be having on our communities and societies. well it’s a good thing we have strong international institutions and shared values we don’t you know it’s a really difficult time for global cooperation outside of ai you know we’re seeing i would argue since the second world war an unprecedented degradation of the international organizations the shared values the rule of law that we have all held so dearly so suffice to say it’s a difficult time for global governance it’s difficult time for the global governance of ai now institutions in the past have very much been brokers mediators and scalers of consensus on tricky governance issues and some of the governance problems we’re facing today are pretty old right i mean i’ve encountered them in previous roles at chatham house and other areas of tech i’m sure the experts on our panel have come across them and the core governance puzzle that we need to figure out is this taking into account the state of geopolitics, the uncertainty around the state of the science, the market dynamics mediated by these leading labs and intensely, intensely competitive US and Chinese ARS dynamics, how on earth do we bring rivals and competitors around the same table?

How do we bring states with a nominal or a minimal alignment of interests and incentives into the same room? Now, you started by asking me about multilateralism and institutions, but maybe let’s reframe this and talk about coalitions. In other areas of governance, what we’ve seen is intense coalition building in crisis or unstable settings around a trusted mechanism, a trusted approach, perhaps in the absence of shared values and principles. And what I’m really interested in, in the context of AI, is where coalition building can develop trust around a credible governance approach, adopt a state champion, get support from associations, from builders, from leading labs themselves, and then scale it using the multilateral format. And over the past few days, I’ve been really excited by some of this splintering to scale dynamics that I’ve seen maybe in conversations on verification, on -chip hardware, risk mitigation strategies, even anonymized collection of usage data, which came out of the commitments yesterday.

Now, what’s the messaging that can drive this coalition building in the absence of trusted institutions, in the absence of shared values? I’ll get into this later in my remarks, but I think it has to be sovereignty and strategic autonomy. Resource -constrained countries. who might decide to adopt a common data governance approach, who might decide to pool resources like compute, have to also consider a degree of governance alignment, again, at this low -hanging fruit, in order to not only withstand the dynamics of the AI race, but to ensure that the collective benefits of cooperation and governance alignment massively outweigh anything they could do individually. So I think I’ll leave it there. Slightly pessimistic take. Let’s see if there’s some more optimism on the

Thank you so much, Bella. I don’t think it was that pessimistic. You did kind of, I think you made it sound very pragmatic in terms of, look, the world is not what we want it to be, and there isn’t the level of multilateral cooperation that we maybe used to have. But you have talked about coalition building, and it’s probably the best we can hope for in the world as it is, as opposed to the world as we’d like it right now. And Rajesh, can I turn to you next? For emergence. economy, obviously access to compute data and infrastructure are critical, but what do you see as some of the barriers most pressing, but also maybe opportunities for AI adoption in India and beyond?

Over to you.

First of all, thank you for having me on the panel. Pleased to be with all of you and then a few of you showed up here as well, so thank you for coming up. We wish this was the Modi inauguration last evening, very a little bit more than this crowd, but nevertheless, we’ll do with this. But you know, I believe we used to talk about digital divide for a long period of time, and I think while that had its own puts and takes, when you compare a smaller economy and smaller country with a larger one and so on, I think the AI divide is going to be much, much bigger than the digital divide which we saw, because the biggest difference is that at least in the digital divide, the you know access and so on whereas this is all about agency and then it can completely put you at a different back foot so it is such an important topic to talk about when you talk about the broader you know haves and have nots and what really goes on with the larger smaller economies and so on and I truly believe that the accessibility when you look at the broader scales it will come across multiple things starting with compute one of the largest you know piece of what we are talking about here right I think as you mentioned in terms of the race between the US and China and so on and so forth but if you leave those two countries then of course we have a big drop in terms of where the real access is going to be and I believe totally that you know the continued limited access to the broader compute facility is going to be undue putting some of these smaller countries, especially the developing ones, into a little bit of a disadvantage.

So, I think there’s a lot that can be done around it in terms of saying, you know, what is that, you know, countries can potentially do in terms of pooling and so on. But I think there is certainly an issue when it comes to compute. And, you know, not just in terms of accessibility, but also in terms of expense and so on, because at the end of the day, all of these are, even if you use the purchasing power parity, and then sort of look at what it costs for people to sort of get into the kind of level of GPUs, potentially, or GPU clusters one has to produce to even have a meaningful language model and so on.

I think that’s going to be a very different ballgame. And the second element of this whole broader issue that we’re talking about is also the data and then the organization of data, availability of data, quality of data, and so on. I think the more you get into the development, world you will find our developing world you will find that the the data itself is very siloed in many ways there are you know different state silos different department silos and so on and it gets into a point where the data which is such an important and integral part of everything to do with AI you will end up having the data which gets fed into the broader models and eventually the AI systems will necessarily not have the right representation of that population which is a huge concern I mean even especially when you you know of course India is slightly luckier in many ways in terms of us you know playing that game a little bit you know punching a little bit above our weight in some sense but but when you when you go down on the on the list of countries which do not have access to all of these I think you’re going to find it even even harder in terms of solving the data issue and the data availability data quality all of that this becomes a bigger issue and there when we talked about infrastructure gap compute gap it’s a little bit more than just the pure compute itself gpus and so on but it’s also about connectivity power uh these are the issues which uh you know we somehow take it for granted in other segments but i think you will find that power is going to be a huge uh foundation for all of that and as you know that there are multiple layers in in building any of the ai systems and one of the uh bottom most layer is going to be power and then you know what really happens to the power and if it has to be clean power then you know does it put additional tax on the on the developing world for for making sure that that power comes out clean connectivity is a huge issue even though it’s kind of broadly solved in in some sense with all the um satellite options and so on but we continue to have the kind of connectivity you need to run a truly inclusive ai system is going to be very different from those uh you know people have thought otherwise and then of course we can go on and on in terms of the the other layers of the power and then of course we can go on and on issue, the availability of skills and ensuring that you have the right skills not just to leverage AI but also to build AI, I mean there are two different type of capabilities that you need to produce in any country so these are the issues and how do you make sure that you have a broader the opportunity itself would be to sort of look at this and say are there other ways of collaborating other ways of partnering and so on, because you know these especially when you go down the line, list of countries we have close to 200 countries or so in the world and when you leave the top 5 or 10 and then you go below and then you keep going down the list, it becomes harder I mean I don’t think that everybody is going to be producing a full blown, large language model and things that they need to sort of do it for themselves at that point in time the question will be can you really partner, can you really leverage some of the common systems that can be done across these countries and so on and so on

Thank you. I mean, you’ve done a brilliant job of putting all the free problems we’ve got and then saying you’ve got a long list afterwards in terms of cooperation. But I love the touch of optimism there at the end. It’s like, you know, if you lift a country out of the room, you still have a hundred and whatever, 85 that need to figure it out. So I liked a lot of that framing. And thanks for touching on

I mean, unsurprisingly, being someone from Mozilla, I’ll probably go with the open source angle as one of the opportunities to actually align the talent, align the capabilities, and actually do shared infrastructure. I mean, maybe I’ll draw two analogies to think about, and then we can go more deep into those as it applies to AI. But for all practical purposes, every computer on the planet runs Linux. There are a few iPhones here and there on top of it. But the Linux model, I think, is a good one for all of us to think about, that every computer… Every country, every nation in the world, almost every company in the world, contributes to the single code base which has been deployed across these billions of computing devices across the planet.

And there are lots of derivative work that happens from it. So like a company like Google can then take that and make it into Android. A company like a vending machine company can deploy Linux onto a Raspberry Pi and run inside their vending machine. So I think there’s an analogy here of being able to use shared infrastructure, shared software infrastructure as a collaboration mechanism that we can all pool resources together but still have sovereignty on top of it. So we can still all be contributing to this common core but then fine -tune our way to our own particular implementations. And I think that if we take that and then marry it with a web analogy of in the early 90s of the original web, you needed to ask for permission in order to deploy a website.

And by permission I mean effectively you had to go buy yourself a Solaris box and then you had to go buy yourself you had to buy yourself a Windows NT. server, you’re trying to configure an ActiveX scenario. And the beauty of what Mozilla and Firefox did, we’re not the only ones who did it, but the beauty of what they did there is a forced openness throughout the stack that enabled anyone without permission to build whatever they wanted. And I think we need to find a similar moment. So in that world, we went from the Windows NT stack and all of IIS to the LAMP stack. And the LAMP stack has these gorgeous analogies of just like anyone can build on Linux.

When Facebook needed PHP to move faster, they did massive improvements on PHP, which then trickled down to all of us. So people can contribute in different ways across it. That’s not the world we’re currently living in AI. We’re living in this world where there are a few frontier model companies that are effectively doing governance for all of us in some way, shape, or form. And I agree with my colleague that that’s an untenable situation. I do live in San Francisco, but you don’t want four people in San Francisco. government’s decisions for the entire world that doesn’t make a lot of sense. So I do think if we can find the LAMP stack equivalent model for AI, and this is actually what I’ve turned all of Mozilla towards, of just like how do we define open standards, how do we define open interfaces so that the vibrancy of the open source community can come together and actually build solutions that work for every single person, every single community, every single government on the planet.

You can sort of build upon, you can contribute to the common base, but then build upon it and take it in a way that makes it more aligned with your country’s values or your company’s values or your individual values, and you can fine -tune your solution out of that. So I think there is an analogy here around how open source could actually provide digital sovereignty across all the different levels. Give us agency as a person, give opportunities for flexibility at a corporation level, and then give. Give countries the ability to own their version of the stack. That could actually be quite beautiful if we can actually figure out how to do that in an appropriate way.

I tried to give you a dose of optimism you have given me a dose of optimism but I’m absolutely shocked you talked about open source thanks so much Ralphie and I did appreciate you brought up the standards because I’m going to talk to Halak and we’re going to go a bit into collaboration and standards here so obviously with the myriad of AI governance frameworks I’m going to turn to you on the question of where do you see potential for alignment on standards maybe some interoperability some maybe risk management framework so keep us on the hopeful path please

I am here to provide the hopeful perspective let me start out by saying that I lead global public policy at Cohere. Cohere is a Canadian AI developer we build models and we have agentic AI our solution is called North so in my role I look across the global regulatory framework that means If our startup wants to, you know, do business in a certain country, I try to understand the regulatory landscape of that country, and then I advise our company if it’s favorable or not. When we’re talking about governance and frameworks that are existing, my perspective is I think it’s not there yet, but I have a more promising view of it. I think that in certain principles, we are converging to where we need to go, and there are strong opportunities.

Technical standards is one of them. You know, there are frameworks like NIST and ISO frameworks. For startups, these are key. The reason they’re key is because they’re flexible and they’re evolving. If we just go country by country, what that’s going to do is price out smaller companies. But if we have an international framework that is evolving and flexible and, you know, we’re going to be able to do that, you know, also including industry coalition, which a lot of the model developers are a part of. But also, like, other stakeholders can be a part of as well. I think it really helps. The second thing I would say is around shared practices, around risk mitigation. So I think there’s strong opportunity there as we come together and share documents or, you know, evaluations around misuse or model capabilities or impact of models.

I think, you know, like I said, we have a way to go, but we are moving closer to that. And then the third thing I would say is interoperability of shared resources. This is key, key, key. We have a big ecosystem. So, yes, there is big tech involved, but there are smaller players. And every single day there’s new startups that are wanting to emerge and wanting to have a go -to -market strategy. And the only way this is possible is if industry and all of industry, big and small, the whole ecosystem starts sharing documents and documentation around, you know, red teaming or evals or multilingual benchmarks and things like that to come to some sort of consensus.

Thanks so much. I’m really enjoying this positive vibe we’re going with. And, you know, that combination, I think it kind of links really nicely back to what Bella was saying around coalitions, you know, build on themes, right? It’s like where do we think we have common ground and what we think we can build on. So I really, really enjoyed that contribution. Rajesh, can I turn to you next? Because I did wonder what all this stuff means for, you know, in kind of smaller and developing economies. And maybe if you have any examples of shared standards, pooled resources, any of the stuff that Halak was talking about, public -private models, or anything that you’ve seen that looks promising, that looks like it could deliver.

Thank you.

You know, as we said, the moment you look at shade models, there are multiple reasons why we want to do this. And one, of course, as we’ve talked about, the cost involved in doing some of that. I think that itself is becoming cost prohibitive and hence there may not be even an option for many of the countries but to sort of have this shade model. We also find that in the regional compute consortiums that, you know, folks can potentially create and you often see examples of where, for example, a standard data set and stuff like that being shared by, not just by, you know, even within a country. It could be between government, academia, and then industry sort of sharing the same sort of data sets, making sure that they’re able to leverage that in some sense.

Compute is clearly something which continues. continues to be the, you know, we shared resources in many of the, even in India, for example, you know, our own AI mission has created this cluster where it can be broadly leveraged by both industry, academia, and the government in terms of ensuring that they’re able to get access to the right set of GPU, set of GPU, GPU forms, and ensuring that they’re able to use that and then take it forward. So, public -private sharing of data, certainly the compute consortium, and then cloud credits, I think that’s something which sovereigns have been able to work with the hyperscalers, especially in terms of getting a lot of, you know, cloud credits for the GPUs, especially, right, because, which is needed for even if you, it’s not about building a frontier model, but it’s even to leverage the frontier model, build some reasoning models on top of it, and ensuring that you’re able to build an application which is meaningful, not that every time you need a powerful GPU, but there are occasions where you definitely you would need and then hence you know using some of those cloud credits will become a big need and then of course when you switch to regulations and so on and ensuring that how do you make sure that even having a policy is something which is shared across you know you don’t want to reinvent the wheel every single time so do you have a method by which you could leverage the existing you know look at what is out there in the world and then sort of leverage it and then try to reuse it because what you don’t want is to have this 100 versions of the same thing with a few nuances here and there so that’s something which I think companies will try and create a model as well.

Thank you so much and I’m gonna kind of turn over to

Yes. Yes. Looking forward to a truth transparency and accountability -driven world. It takes 30 years for FC files to come out in a place like America, the developed world. Is that the speed of the system till it collapses and till we start a new world? Are we resigned to that fate?

Yeah, so I can’t really see the link between the Epstein files and the… 30 years since the world was destroyed by Aaron Mulder in 2001. You don’t do the truth to come out. So you don’t have the system speed. Yes. Sure. Thank you. So on… Just to kind of build on what Rajesh was saying there on kind of also the capability. So maybe if we move into a bit of cross -border cooperation and Bella, if I can maybe turn to you just to build on those points. Because obviously what we are seeing across the developing world in particular, often it’s kind of the institutional capacity that’s a bit of an issue there. Yeah. kind of doing all the engagement and all the investments and all the, you know, you kind of still run into.

What are some of the, and I saw you were taking notes furiously, so I’m sure you have reflections on what has been said so far. But also, what are some of the resources

dependencies, figure out what they want to invest in and what dependencies they’re willing to accept, wanting to build strong institutions, again, that can mainline AI directly into public service delivery, and as you said, enable cross -border cooperation, might take a step back and figure out which foreign capabilities or foreign services they’re willing to accept at some levels of the stack and where they’d like to invest in indigenous solutions. And I mentioned open source earlier because this has come up time and time again, and I’m sure it’s going to be absolutely no surprise to our audience here today. An example which has really stuck with me and Rafi, I’d be really interested in your thoughts on this, is the Southeast Asian Languages Under One Network model, so a multilingual sea lion LLM.

And this is something we’ve called, again, in a really interesting collaboration with AI Safety Asia, open models. With local adaptation, really balancing. again inputs from open source models potentially provided by foreign providers with adaptation to a local context and so i think leaving the summit what i’m really going to be interested in is i think this connection between drawing on i guess inputs from the open source community fine -tuning and locally adapting their contributions and then perhaps doing so not only in the service of again strong robust institutions at the national level who are ai ready but also on this kind of collective cross -border level i hope that makes sense

it does and i’m gonna let rafi kind of uh fit into that as well because you’ve uh you’ve uh segwayed really nicely into into his uh part but also um if you can also touch upon feel free to react to what uh bella has said but also if you can also touch upon on on the what you’ve seen as best practice in international and cross -border collaboration maybe in healthcare climate resilience audience education anywhere you’ve seen good stories to tell please do share

i mean i do think a lot about the local fine -tuning and i think that that’s actually a really powerful concept of like we can all contribute to a core and then locally fine -tune for our values and our needs and i think that this has shown up in a bunch of different ways and i’m interested personally in all these alternative i don’t even want to call it our alternative but like other architectures that enable this to be possible because in some ways we’re kind of being being fed a regime that says it’s not possible but i think like architecturally it actually is in a bunch of different ways so i love the indigenous data model like looking at what different indigenous peoples have done around data collectives for their local areas so there’s a group of people for example in hawaii that is doing this for their genomic data because genomic data is really useful for pharmaceutical models and so like they’ve been looking for ways so that they can both monetize but also the provenance of their data as it goes through these pharmaceutical models.

So there are some professors out of UCSD starting to build actually what these data trusts could look like for Hawaiian people, and I think that that model could be replicated in lots of different parts of the world. Mozilla is actually attempting to do a bunch of this. So we’re creating something that we call the Mozilla Data Collaborative, or Collective, sorry. And what the Collective is meant to be is it’s meant to be a marketplace of ethically sourced but provenance -traced data sets so that you can bring your data. It will actually help you scrub it, clean it, et cetera, and also make sure you have the appropriate licenses on it so that people can come find the data sets that they want to train their models but make sure that attribution is given, compensation is given, et cetera.

So we’re literally in conversations with almost every radio station on the planet to try to get their recordings and their transcripts onto the marketplace, not for Mozilla to make money. In fact, we actually want the radio stations. to have a monetization path for all the data that they’re sitting on. simply have it scraped by big model providers to try to soak that into their systems. Instead, require that it be licensed, require that compensation be given. So I think there are models there. And on the computational side, I think there’s also a lot of interesting things showing up around federated learning opportunities. For those of you who don’t know what federated learning is, think of, you know, Google did this very famously when they trained their handwriting model across everyone’s Android phones.

So your handwriting is very personal and private. Your handwriting is on your device. And Google is able to train a handwriting recognition model that didn’t require them to get access to your data, because part of the training happened on your phone, and then the model wait through shipped it back up for centralized training. And I think something like that actually could be an interesting model for international collaboration of like, I can bring my data to the game, my healthcare data, my values data, my language data, but not have to release it to a different company, or sorry, a different country. Instead, allow you to do it in a different way. And I think that’s a really interesting model.

Thank you. of the training on my compute, on my infrastructure, and only ship model weights back up, and actually then create bigger models across borders and across geographies that could actually take into account different healthcare scenarios, different value systems, et cetera, in there. So I think that there are these interesting alternative architectures that we can actually start leaning into, these data trust models, these federated learning models, that actually could be massive enablers for cooperation and allow us to build these foundational things that we can then fine -tune and bring to our local context.

Thanks so much, Rafi. That fine -tuning seems to be definitely a thing in this conversation, how you kind of built for different cultures and countries. And Halak, maybe I can come to you next, because we keep talking about kind of international cooperation and coordination. But I’m wondering, how do you translate that? played that, you know, chit -chat into actual skills, capacity, capability for emerging economies. And, you know, I mean, we are in a very international AI Impact Summit. So, you know, kind of how do we bring that from we talk about governance to all this international policy actually delivering for emerging economies?

It’s a good question. I think let me start out by saying capacity building isn’t just, you know, running workshops or basically talking to regulators about, you know, this should be done. What it is is capacity building, I think, for emerging economies especially is critical because – hold on. Let me think. Okay, so emerging economies have unequal access to data, information, and technology, right? So what are we trying to solve for here in terms of capacity building? The first thing I would say is shared evidence. So what we need is we need players to help into this capacity building system with documents, results, performance, benchmarks, to lift up other players. That, I think, would be number one.

The second thing I think is key and sometimes overlooked is the value of, like, procurement policies. And I agree with Isabella. What if we had, like, an industry coalition, like a cross -border network, where they’re solving for procurement policies or procurement rules? And what this does is this brings in global players. So now what you’re doing is you’re opening up your country to different markets. The next thing I would say is, like, you know, a lot of – Let me think. Let me put it this way. So there are developers who develop the technology, and then there are deployers. They buy the technology, and they use it. So, for example, like a public sector agency. Why is it so – Thank you.

Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. you you Thank you. Thank you. economist Frank Nagel has a report recently that approximately 24 billion U .S. dollars are being wasted by not switching to open source models right now. So the economics are starting to make a lot of sense. So I think once all these stars align, it becomes almost obvious what an answer could look like for local governments around open source AI models, et cetera. So I’m really excited for that in the next 12 to 18 months.

Thank you. Rajesh?

No, I agree with both of what’s been said so far, but I also want to give a sense of I think it’s when you look at AI governance and people tend to sort of lead with regulatory regulation first. I believe that countries and especially the countries which we talked about in terms of more from an inclusion point of view, you’ve got to lead with innovation first mindset because I think regulation is required and certainly needed, but I think innovation is probably needed more in some sense. And also when you look at the AI governance, governance across all of that. we do while there could be horizontal governance which will apply to every AI systems I think the more meaningful governance that you’re going to find when you get into sectoral governance meaning when you look at the AI systems for health care and you’ll find there are the understanding of a harm in the health care segment is very different from financial services and so on so how do you get into those sectoral areas then you can have meaningful governance structure and last but not the least you need to have the right talent and people who can actually who understand all of this in both in public sector and people who are supposedly governing all of this that is something which is it’s not the talent in terms of broader AI model building and people who are building AI systems but how do you make sure that the talent in the governance space in the governments and people who are actually regulating it they don’t understand the real harm and then it’s going to be a bigger issue and especially when it comes to the you know the list of countries that we talked about always when you get deeper down the list you will find that the talent is going to be here and there and you will find that the talent is going to be here and there and you will find that be here and there and you will find that the talent is going to be here and there and you will find that the talent is going to be here and there and you will find that the talent issue in terms of understanding.

Thank you. And, you know, as someone who lives in Brussels, I’ll make sure to take that message back. Halak.

Okay, so what am I most excited about, I guess, in the next 12 months? I mean, in the last few days, you’ve seen companies really, really excited about AI, but what you’ve also seen is countries very excited about AI. So what does this mean in governance? It means that the community and the participation is only going to increase. I don’t see it going backwards. And so, as technology is evolving, more players are going to have a voice in the system, and the standards and the ITU bodies or the ISO bodies, and I think because of this convergence, we are going to, as society, just, like, increase our, like, literacy of not only AI, but technology, but also bring it into whatever we’re in, if we’re in the private sector, if we’re in the public sector.

And because of that, I think we’re going to have to Yeah, I think a lot of progress will be made in the next 12 months, and you’ll see it as it converges.

Thank you so much. Thanks to all the panel. Thanks for being here, and enjoy the rest of your day. Thank you. Thank you.

from the world. We had practically every major AI player in the world participating in large numbers. We had so many startups getting the opportunity to showcase their work. Overall, the quality of discussion was phenomenal. If you look at the ministerial dialogue, the leaders plenary, the main inauguration function, the summit, the quality of participation, the quality of dialogue was phenomenal. Honorable Prime Minister Narendra Modi’s vision of Manav AI, which is AI of the humans, by the humans, for the humans. I think that was very well accepted practically every major AI player in the world. The ministerial, which we had the bilateral, which… I and my colleague Sri Jitinji had. Everywhere, practically every minister resonated with this and everybody felt happy that we have brought the discussion about responsible and ethical AI to the forefront by involving two and a half lakh students in this entire journey.

We had a Guinness World Record for that involvement of the students. We also have a lot of investment pledges I was just asking Abhishekji and Krishnanji I think the number is growing each day so it’s already crossed 250 billion dollars for the infra -related investments and about 20 billion dollars for the VC deep tech investments which have been committed by investors. This is a very important sign for us. the numbers are important but what is important is the world has confidence on India’s role in the new AI age. That’s very very important for all of us because as we have seen there is always a always a need to bring out the talent that we have, bring out the energy that we have in front of the world so that the world recognizes that.

I would also like to share with you that the action summit, the previous one had about 60 signatories in the final declaration. We have already crossed 70. There are many ministers who are here and they are discussing with us. So I think by the time we close the summit tomorrow, we have, as you know, we have extended it by one more day. We believe that it will cross 80. All the major countries have already signed. If you feel that somebody has not signed, you need not speculate on that. All the important AI, people who matter in AI, they have all signed. We will be giving you the formal number tomorrow once the summit closes. That is the way it should be done.

That’s the right way of doing things. We also had many interesting episodes where I met a very young innovator today morning and even in the previous couple of days back also. Some very young people have done so much work in AI, which was very, very encouraging. Because if the youth sees that hope in this new world, the youth has that positivity. about this technology. We also found very strong endorsement of our policy of working on all the five layers and our focus on having a sovereign bouquet of models. The models which were released, I tell you, in every bilateral that I had with the industry leaders, they are really surprised at the quality of output with such few resources, with such the kind of resources which some of the frontier labs have at their disposal, and with such frugal resources our engineers and researchers have produced such good models which is what gives huge, huge endorsement to our efforts.

I would also like to thank all the team members. All the stakeholders, right from media, from the organizers, from ITPO, and special thanks to the MEA and Ministry of Home Affairs, Delhi Police. for so much effort. They have tirelessly put in making this a grand success. Thank you everybody who participated in this. And also thanks to the youth who endorsed this, who took this so positively that whatever little effort that Congress made for trying to disrupt the summit was really, really, I mean, the youth very clearly said that this is their exhibition. It is the exhibition. This is the summit for the youth who want to make the best use of it. They don’t believe in the negative politics that Congress was trying to play.

We had some bad choices here, people coming into the exhibition, and we took immediate action against anybody who tried to demean the… demean the good work that is being done by our startups and by engineers, by our people who are working in the AI field. That is something that we are a very open -minded government. We believe in taking your feedback. We believe in working with you. We believe in the goal of Vixit Bharat, and that’s why we would like to tirelessly work with you for this goal, which our Prime Minister has given that vision for our entire country, and we have to do it together. This has to be done by all the 140 crore of our citizens who believe in this common goal of Vixit Bharat, and these are steps in that direction.

Friends, tomorrow we will be also laying the foundation for our next… …semiconductor plant here in Uttar Pradesh. I’ll invite all of you to join that ceremony also and on 28th we’ll start the commercial production from Micron facility and that will be one of the largest facilities that Micron has practically more than 10 cricket fields kind of facility it’s very large and that is going to be inaugurated on 28th so all these are steps very methodical step by step moving in the direction for creating that foundation which our Prime Minister is laying for the young generation for Vixit Bharat for all of you who are watching it on TV or social media our Prime Minister Shri Narendra Modi Ji is laying the foundation for the country, which will be a developed nation by 2047.

I’ll take questions, and like in the past, we’ll follow the first row,

Thank you, sir. First of all, please identify yourself and your organization’s name before asking the question. And as sir has said, start from the left. Yes, please.

Hi, sir. I am Nishant Ketu from ANI. My question is, how do you see that India’s role in… useful tools for day -to -day Python development and Python work? So we have developed a program which is like absolutely beginner… Even if you have zero knowledge in Python, you can join this program. And in 30 days, you will be becoming pro of Python. That also using AI. You will be not becoming a Python developer. You will be becoming a 10x Python developer. And the best part, like we have democratized this program. So this 30 -day program is just for Rs .199. Can you believe that? Prime Minister began this on the 16th when this program began. And today where we are.

What is the observation Prime Minister has given to you or indication? He has given to you. Sorry? Observation or indication that he has given to you about AI Impact Summit.

Next. Please.

Hi, this is Deepak Ajwani from Economic Times Digital Team. I have one simple ask. Have there been certain guidelines, guardrails that have been put together by all the countries that got represented yesterday on the stage? on effective, ethical, and responsible use of AI. Is there a paper that you can bring about maybe tomorrow where all of you have agreed that this is at least the first set of blueprint which can be iterative later? Thank you. Hi, sir. Shauvik from Mint. Sir, two questions. One is on the participation from big tech companies. Have there been conversations? The global tech companies. Have there been conversations with global tech companies in terms of the role that they will play in India as far as public services are concerned?

Because each of them spoke about AI and its role in public services. And secondly, the models that were launched under the AI mission, they’ve also been backed. Is there a takeaway from the summit in terms of where they go from here? Thank you. So, Oyeek from Money Control. So, just wanted to ask you, yesterday you had, you had the frontier AI commitments. So, the declaration will also come tomorrow. So the frontier AI commitment is voluntary in nature. The Delhi Declaration, I’m assuming, is non -binding. So how do we ensure that this does not remain on paper? How do we ensure implementation?

Can you repeat your question?

So I’m saying the frontier AI commitment is voluntary in nature, and the Delhi Declaration, whenever it comes, is non -binding. So how do we ensure that this does not remain on paper? Like the declarations, commitments made in the…

Anybody else on front row? Anyone? Okay, please.

Sir, Ashish from Business Standard. All of the three previous summits had a focus area when it came to the declarations. If you could share just one line on what would be our focus area when the declaration is signed.

Please. You are close? Yeah.

Hi, sir. Shubhan from the Economic Times. I understand that the declaration will be coming tomorrow, and as you mentioned, some 80 -odd countries will maybe… The list may be as high as… Now it’s 70, maybe up to 80. I just wonder… I wanted to understand, since some of the last… Some of the previous summits have seen significant difference of opinion in India, what were some of the areas where it was relatively easier to build consensus? and if possible, what were some of the areas where it took a little bit of time?

Next. Last.

Hi, sir. We look AI -ready globally, but my question would be for the last person standing in India, how far and how long it will take to reach to that one last person in India? How long will it take for AI to reach there? Very good question. Hi, sir. This is Lalit from Best Media Info. My question is, we have been seeing that traditional media sectors like TV, radio, print, they have been fighting for ad ex, for advertising revenue, while digital is scaling up. Is there any way that AI or any policy can actually help bring balance in this revenue share of advertisement? My second question would be, there has been a long -pending TRP guidelines overhaul that was formulated.

Normally, you know, it was meant to bring multi -agency system into the picture and removal of landing pages. We just want to know where or in what stage that guidelines are in, and can we expect the guidelines coming in anytime soon? Sir, I am Prashant from AsiaNet News. There were very good sessions in this summit. How do you wish to take down to the grassroots level how these sessions can help the lives of common man?

So, there are questions about where do we go from here? What will be the implementation? I’ll take all these questions one by one. I think, friends, the journey so far has been very meaningful, very methodical, starting from building the base and working through all the layers of the IC, and creating that foundational level of work. and now getting the entire world to come here, deliberate, interact with our industry. Now we’ll take the next level of our AI mission where we will be focusing and taking to a totally new level of models, a new level of common compute, new level of safety. We have so many collaborations agreed in the last few days, which is where I would like to address that point about paper versus real action.

Yes, there is lots and lots of real action, real MOUs, real understanding, which has happened in the past few days, where many of these things which concern us as well as the entire world will be working in a very collaborative manner. That is the… That is the real action which will come out of it. We will be very soon start working on the AI mission 2 .0, which will be definitely bigger than what it was in the AI mission 1 .0. Many of the goals we had set for ourselves in the mission 1 .0 are on the verge of getting completed and many of them have actually exceeded. We wanted about 10 ,000 GPUs. We have 38 ,000 already and another 20 ,000 very soon going to be launched.

We have foundational models. We were looking at two foundational models. We have a bouquet of 12 models and very multimodal, reasonably, I mean very well rated. We wanted to have an AI safety institute. We have now 12 institutes working on this in a network mode. So all these goals that we had set for ourselves are getting implemented very rapidly. So now we have to set bigger goals. And achieve them as a part of the AI mission 2 .0. our Honourable Prime Minister has always led from the front the vision Manav AI that he gave yesterday is something which everybody resonated everybody accepted in the ministerial dialogue, in the bilaterals everywhere people thought that first time they have heard a vision which is so compelling and it just cuts across every civilization every country this is meaningful for everybody every generation, every sector every country because ultimately it’s the humanity which matters the most and that’s why this vision resonated with everybody big tech participated very much in this same the participation of the startups and young innovators it was very good participation there is huge consensus on the declaration we just want to maximize the numbers we in India should be we are not going to be reading that effort.

It’s so natural to do it. And given the size of this summit which has happened, it’s natural to set a number so that the record is always here. So that’s why we are trying to maximize that in a very… In fact, Abhishek… Do a little more work. He was thinking he would take a day off. But no, he’s not going to get a day off. So do a little more work. Very important question which came is about diffusion, about the last person. How do we see the benefit? If you go there… rich countries, you will find that 5G is very patchy. It’s not very the way it is in our country. Vaisahi effort isme bhi lagayenge.

Aur isme bahut mehnat karni padhegi. And we are prepared to put that hard work, put that effort. Our Prime Minister keeps inspiring us that for us, we should not stop till the benefit reaches the last person in the society. That is our goal always. And we in BJP have always had this basic tenet as Antiyode. We believe in inclusive growth. And if you look at Honourable Prime Minister’s programs, each and every program, whether it is Jan Dhan, whether it is the Swachh Bharat, whether it is construction of toilets, whether it is Hargarh Nal Sejan, each and every program has been created and executed to bring the benefits. Thank you. We believe in inclusive growth as a basic political philosophy, and that’s why here also that same political philosophy will be reflected.

I have absolutely no doubt about it because this is a family in which inclusive growth is one of the most important tenets of our thought process. There are questions about guardrails. You might have seen the first major, first time all the big AI players came on the same stage and agreed. Voluntary is more like saying it, but of course we have discussed with them, and all of them have come to this consensus. Taking those first steps was very, very important, and I don’t want to… exaggerate this but if you ask any major policy leader in the world and I had so many meetings today each and everyone is surprised how we could pull together the entire AI industry coming forward and started openly.

Major major achievement and this kind of achievement shows how India can be leading the thought process. We also had Pax Silica today which is very important for us from the semiconductor industry perspective from resilient supply chain resilient value chain perspective and the fact that we are today seen whether in Europe or in Australia or in US or in Southeast Asia everywhere we are seen as a trusted country that itself speaks a lot about how our Prime Minister has conducted the foreign policy how our Prime Minister has developed that trust among the entire global every sector every moment every geography every part of the world we will Youth Congress I have already responded

next room just a second second room please we will come one by one yes please

MIBC related questions I will answer later today we will talk about AI mission we will talk about AI mission next time

Namaskar Sir Sir Sir Thank you. Thank you. Thank you. These are my two questions.

Anybody else?

Sejal Sharma from Hindustan Times.

Just a second. Yes, please. Who is asking? Second row? Yeah, please.

Congratulations on the declaration, sir. I just wanted to know, could you give us names of some of the countries that have signed the declaration already? Just a few. Not proper. Good evening, sir. I’m an independent journalist. I’m an independent journalist. So first, I want to know what are the outcomes from each of the seven working groups that were formed before the summit. Second, before the summit began, the Indian government focused on how India will lead the global south. How has that materialized during the summit? Has it materialized in form of bilateral conversations, any MOUs, any pacts being signed? And number three, today the SGI amendments are supposed to go into effect. We had all the big AI companies and big tech companies of the world here.

Has there been any discussion on that? Because the companies have been fairly critical, both off the record and on the record, about the compliance deadline as well as the three -year takedown window as well as some of the provenance -related specifics of the SGI amendments. Sir, Manas from the Times of India. Sir, has the objective of a technological framework been achieved? And how many countries are on board? And what is the reaction of the big tech? And there is that. And given the representation of the big tech companies, what is the government doing to ensure that we are not going to be the data and talent supplier?

Last in the second room, I think she’s about more sauce. Thank you.

Sir, Momita from PTI. I think everybody is curious to know the contours of the New Delhi Declaration. The focus and thrust, you know, rightly for India has been impact. The use cases and how it benefits the public. If you could just give us some color on what the New Delhi Declaration contours look like. Which are the areas where consensus has already been reached, where 70 countries are coming together and supporting those causes. And how would it benefit Indians?

Third room. Anybody in the third room? Okay. Please.

Momita from Outlook Business. Thank you. Yes. So I wanted to understand, recently French President mentioned about he urged India to be a part of the social media band for those under the age of 15. So has there been some sort of consensus that you reached with other countries about the same?

Anybody in third row? Okay. Please.

Hello, sir. Himanshu Desai from Rajasthan Patrika. Sir, so I wanted to ask, like, what role will…

Patrika se toh Hindi pe pushna chahiye. Mujhe bhi toh chance mile Hindi me jawab ne nahi ka.

Ji, bilkul. Sir, toh main yeh pushna chahta… Pachpang se padta ho Patrika. Sir, main yeh pushna chahta ho, jase aaj humne Dr. Mohan Yadav, Madhya Pradesh ke CM ki bhi briefing dekhi. So, matlab, like, state governments ka jo pura ka pura plan rahega, aur kaise government jo hai wo state governments ke saath mil kar kaam karegi? Like, agar hum specially…

State governments… Bawal chak. Bawal chak. Thank you. Thank you.

Hi, sir. Yaku Tali from DLU Hindi. Sir, my question to you is, what is the government doing about data protection? Because we are seeing OpenAI, ChatGPT and Microsoft are taking access to all the data. Yesterday, a notification was also sent in which it was said that you can now share your contacts and then reach out to your contacts. So, don’t you think that all the Indians are taking their data?

Hello, sir. Yes, right side.

Yes, among them. Anybody else on this thing? Third row? we are working with industry on that visheshkar jo colleges aur schools mein course curriculum banna hai usme industry ke inputs lagataar aare hai bhi aur jaisi wo final hota hai wo aapke saath share bhi karenge semiconductor ka bhi industry ke saath milke kia tha telecom ka bhi industry ke saath milke kia hai to iska bhi industry ke saath milke hi karenge jaisi ki ek relevant practical useful knowledge aasake industry ke liye state governments ko bahut closely isme participate karenge kyunki ultimately janjan tak paunchne ka madhiyam rajya sarkaron ke tarike se hi ho sakta hai sarvam karib karib har benchmark pe bakiyon se kharay utra hai aur khas kar jo open AI se bhi aur deep seek se bhi aur gemini ke pro model se bhi kai benchmark pe wo better hai aap chaho to uske jo unho ne drop kia on all the globally accepted parameters.

The new regulations of SGI have been accepted and everybody has told that this is a necessity of the country and many countries in the world are already talking about bringing regulations in this direction. In fact, many countries have congratulated India and have taken the first step. And in the coming time, many more countries will watermark this. And the main purpose of this is that Is it real content or synthetic content? That transparency is necessary so that you can decide for yourself whether to trust it or not. The second thing is also very important for us to know that the SGI has been accepted by the world. that the law and constitution in the society that is illegal, is also illegal in the online world.

What is illegal in the physical world is also illegal in the online world. It is a very natural constitutional mandate. So I didn’t get anyone who opposed it. If you get someone, then do reach me. The techno -legal framework is growing very fast. I have already given a statement on children’s protection. The data protection framework is very strong. In fact, I don’t want to take names, but three countries have said that they want to make their data protection framework equal to India’s data protection framework. Already, in today’s and tomorrow’s meetings, Aaj kal aur parso 3 din ki meeting mo Already 3 desho ne kaha ki Aapka template bahut achcha hai Isi tare ke kanun ko am bhi banana chahi Mythbusters ki baad Training ke saath hi chudegi Taaki kis tare se Koi bhi point ho, uska benefit kya se Abhi chalain Fourth row Left side se Anybody in fourth row Please

Namaskar, main Sandeep ho Prabhat Khabar Jharkhand se Jis tarike se AI ko lekar Shor mach raha hai Usme Sabse jada train karne ke Zorat bachcho ko hai aaj ke daur me To kya kisi tarah ka koi module Ye course start kiya jayega Schoolo me ki saath saal Aat saal, das saal ke bachcho ko AI ki training di jayega Is tarah ki koi yojana hai Koi Koi Planning hai

Next Aage Next

Thank you, sir. Arundeep from The Hindu. So just one question. You’ve had the opportunity to interact with a lot of leaders in AI and world leaders on a range of subjects. Does the government of India, after this event, believe that AGI is coming in the next two years? What is the government’s position on that, clearly? And if so, are we prepared for that as a country? And I lied, I have another question. Second question is, the next summit is going to be held in Switzerland. But given the response to this edition, is this something that we might do again in the coming future?

Okay. Ajay.

Sir, the question is, what India has done in the UBI, what we have done in the UBI, democratization waha par kya? Kya hum koshish karenge AI mein? Aur is democratization ke is open source model ko puri duniya ne kis tarah se liya hai? Wo hum samajhna chahin. Thank you, sir.

Fourth row, anybody else? Highest, please.

Hi, good evening, sir. Ashmit from CNBC TV18. Firstly, congratulations on the largest ever AI summit. I had two questions, sir. Amongst the companies that were here, there were also the likes of NVIDIA, AMD. One concern, India is going for a data center build -out, as was evident from the large commitments that we’ve seen. The cost of compute, the cost of chips is something that constantly kept coming up in the conversations. Has that been discussed? And are there any material assurances, gains for India under the PAC silica arrangement? That’s one. Second, you spoke earlier about diffusion. I just want to get a little clarity as a part of the mission 2 .0 that you made a reference to.

A lot of these applications for AI for social purposes are the ROI may not be immediately available for the developer. In such a case, is the government willing to step in under Mission 2 .0 under some form of support or viability gap funding? Okay. Can I ask? Okay, I’m Surabhi from the Economic Times, sir. Two questions. One is I wanted to understand from you that when we launched the first version of the India AI mission from that time to now, I think a lot has changed in the AI ecosystem. So what are going to be the main focus areas of the next phase of the AI mission? Secondly, I know you want to talk about the declaration tomorrow and not today, but I wanted to understand that you’ve had meetings with the biggest names of AI as far as AI leaders are concerned.

What are some of the things, discussion points that have come up? What have been some of the asks that they have made to you and you have made to them as far as their contributions to India are concerned?

Mr. Roshan, no matter how many questions you ask, once you have answered them, fifth row, anybody? It’s less than two minutes.

Good evening, Mr. Minister. This is Arunodai Mukherjee from the BBC. I just wanted to understand and draw your attention to the U .S. delegation, which was here earlier today. They have very strongly rejected calls for global governance in AI. I wanted your response to that. And doesn’t that go against… what this entire summit was all about, charting a path which is a unified path towards global governance. How would you respond to that?

Yes. Thanks. Amrit Pal.

Minister, this is Amrit Pal from DD India. The IMF chief today said that while AI could lift global growth by a percentage point and help India achieve… What is… How is the government preparing to deal with that? My question to you is, in the face of rising deep fakes and sophisticated artificial intelligence misleading information, how does the government ensure the accountability without touching ease for doing the business for startups?

Backside. Brahma Prakash. Brahma Prakash, the way Zee News say. Thank you. Next. Yes, please.

So my question is related to the declaration. I do understand that you want to talk about it tomorrow, but if you could throw some light on whether there is some sort of consensus on demarking high -risk AI, or will that be left to national governments to decide and demark it? Thank you.

Yes, please, on the left side. Please pass on.

Sir, my question is in regards to the Global South. Since this was the first summit to be held in a Global South country, we saw significant representation. Africa, there was an Africa air village. So my question is, can we see, you know, Global South priorities in terms of how AI should be developed? Will they be reflected in the joint statement? And what, according to you, are the major takeaways for the broader Global South? And since Prime Minister Modi also, you know, he championed the inclusion of African Union during the G20 summit. Thank you.

Anyone else? Yes, please. Back side, one person is left.

Hi, sir. Jatin Grover from Mint. A couple of questions. I wanted to understand any discussions with the participating nations, maybe to create a G20 -like group, so that help us creating some sort of a binding agreements with the nations on the AI declarations. that’s one and till a few minutes back at the ATL conference you talked about having a legal framework to address the cyber crime basically arising from AI can you please elaborate more on that what kind of legal framework the government is looking at thank you

anybody else wants question otherwise we are closing ok ok

hi sir from the economic times this is about the 12 foundation models that the India AI mission is backing we launched 3 of them do you have visibility on when the rest of the 9 will be launched and also have you finalized the terms of the agreements with these companies on how much the government of India will be getting in terms of equity etc

last question at the back side

good evening sir I am Shreyas Bharadwaj from IIM Indore and IIT Indore I am an independent journalist but also a student of masterclasses of science and data science and management thank you for letting me speak so my question would be very open to any questions what has the government learnt in two aspects one, itte bade viman ko chalane ke liye bohat kuch challenges aayonge AI impact summit 2026 se government ne sapse zyada kya learning sikiye as a learner as a lifelong learner number two, tech me government ne sapse badi sikya liye is poore summit se that’s my question, thank you sir

ok, thank you very much UPSC me se kam the questions ma sabse pehle packed silica lunga that’s very important because see we are trying to create the complete ecosystem develop the complete ecosystem of semiconductor industry in our country to get the ecosystem it’s very important that all the major players, the major countries where the ecosystem currently resides should also support and encourage our journey That’s why it’s very, very important that we had the packed silica sign today. From all the discussions that we had, it very clearly emerged that the world looks at India as clearly a trusted partner for semiconductor supply chain, which means the way semiconductor industry will grow in our country in the coming years, that looks like a very important, it will emerge as a major sector.

It’s a very important sector that is very visible. Very clearly, it was evident from the discussions. Same thing will apply to… Do you know in 2026, the highest -paid people in industry are not MBAs or fancy degree holders. They are agentic… By now, I recall two meetings in which people are looking at reducing, power consumption at least 50%. and reducing cost significantly. Sometimes even some people even said that a fraction of the cost of the current chips. So that kind of innovation is happening. And India will be a big beneficiary of that innovation because we are starting our design and semiconductor journey at a point where we can use all the benefits that we know about AI and optimize our design of chips according to the new age.

We are not bound by the legacy of the past. We can actually make a new beginning, which is what we have challenged our startups in the Semicon 2 .0 where we want to have a series of deep tech startups designing chips. I’ve spoken about the next steps. I’ve spoken about education, democratization, diffusion, ROI. Yes, I believe that ROI will come from the application. Most of the enterprise… use cases which are visible here in large number. I think I read one story from one of the digital versions of one of the big channels where this point was also very clearly brought out that while people are mostly focusing on the consumer facing applications but the large number of enterprise solution providers who are participating in this exhibition, that’s very important both from the jobs perspective, from the IT industry’s health perspective and from the direction that India will be taking as a major player in the AI world going forward.

Yes, we have a comprehensive plan. Every sector, as we have maintained right from day one, every sector will be benefited by this. On cyber security, so many sessions have happened. We just inaugurated one research institute between Zscaler and Airtel, and many more such initiatives are going to happen in the coming time frame. The declaration will, when the text comes out, you’ll be able to see the contours of the declaration. Global South, of course, participated in large numbers and very interested in collaborating with us, and that level of trust is there. When the next models will be launched, we’ll keep sharing with you as we progress on that. We had committed one, we have done three.

So it’s like a good, it’s a journey which we’ll keep sharing with you. Learnings, many. One was very surprising learnings that when so many good things are happening, how one small thing can be highlighted so much, it’s a personal learning for me. It was also a learning for me that… It’s a learning for me. people who are in politics, they don’t even, some of the people are opposition, they don’t even understand what the youth today wants and they try to create things which really, I mean, it’s really sad in a way and funny in another way and unko kaun samja sakta hai, I don’t know. Many learnings are there. Here, we’ll use these learnings to improve all the future and this was a very large scale.

As I said, already five lakh plus visitors have already, we were just doing the estimate, I think actual number is about six, but we are just being very conservative, everything which is measured is what we would like to share with you. That kind of participation is there and in the end, I’d like to request MEA to because they have been very important partner for us. your role has been stellar. Delhi Police also I would like to thank. All the security participants who were present throughout this and all the friends of media, you played a very constructive role. A big round of applause for the media. Thank you, friends. Kandeer.

Thank you, sir. It has been a pleasure for us in the Ministry of External Affairs to work along with METI as Team India to put our best foot forward for the world. This event has been a success, may I say a grand success. We have heard world leaders who are here. We had 20 world leaders who attended this AI summit. In addition, we had 45 delegations represented at ministerial level from across the world. We also had 100 countries represented. Thank you.

Good evening, distinguished guests. Welcome to the session on powering AI. As AI scales at speed, so does its infrastructure demands. Data centers are facing unprecedented power and cooling requirements. A single large AI training run can consume as much electricity as thousands of homes use in a year. This raises critical questions like how do we plan for rapidly rising and uncertainty energy demand? Can edge computing reduce the load, or is centralization inevitable? To address these critical issues, we are joined by our exceptional panelists. Mr. Vineet Mittal, Chairman of Avada Group. Sir, I request you to please come on stage. Mr. Natham Blom, Vice President, Cooling Chambers. Professor Raghav Chandra, Professor at IIM Calcutta, Founder and CEO of Consult and former Chairman of NHAI and Secretary to Government of India.

Moderating this important conversation is Mr. Ashish Khanna, Director General of the International Solar Alliance. Mr. Ashish Khanna, Director General of the International Solar Alliance. Thank you panelists for being here with that I now request Mr. Khanna to please take the discussion forward

Good evening everyone not easy being the last panel especially when we are probably starting at the time that we are supposed to end but we hope and we will try and make it more interesting for all of you we are here to talk about Powering AI the format will be that I will begin in terms of framing some of the issues at heart and also tell you a little bit about what International Solar Alliance is going to do then I will hand over to each of the esteemed panelists to make an opening kind of a statement of what’s their vision on this question of Powering AI for about 5 minutes each and then I will ask them one question each and then I will ask the question and then I will ask the question on some of the specific issues for which they are probably an expert on.

And finally, if there is any time left, we will see if any audience member wants to ask a question. Let me start off by saying, why is International Solar Alliance in this session and in this AI Impact Summit? We are here primarily for two reasons. The first reason is, the world has done 1000 gigawatt of solar doubled in just last two years, what was done in last 25 years. Almost 40 % of that is decentralized, which means it’s either solar rooftop or pump or others. That figure is only 15 or 20 % in India and obviously very low in a lot of developing countries. And a distribution company often does not like decentralized solar because it impacts the distribution. It impacts the distribution system and finances.

But the right amount of digitization and AI can actually help them absorb it and reduce the cost of the system as a whole. And therefore, India’s ability to more than double decentralize renewable energy, but in general, world over, will require AI. That’s issue number one, for which actually I will say that we launched a global AI mission for energy in the AI Impact Summit. We call it AI for Energy. The session is going to talk about energy for meeting AI demands. But let’s first talk about AI for Energy. Why? Because there are some elements that the world has not seen, which is, if some of you were part of some sessions earlier, can consumers trade power based on what rooftop and batteries do you have, P2P trading, that requires certain digital enablement of the trade of millions of consumers, producers and consumers, that right now needs a lot of regulatory evolvement.

An IT architecture, so that each distribution company in India, but for that, that matter anywhere in the world will know what will it make ready to actually trade that power. all. It’s about jobs. Today, a lot of AI engineers do not understand energy. Energy engineers do not understand AI. We at International Solar Alliance, which is now 125 country member body, headquartered in India, is creating an ISI Academy to train people to bring together AI and energy skills engineers together. This intersection of energy and AI will be the fundamental shift over next five years, the way Amazon changed retail. This is what is going to happen, we believe, in renewable energy. Third, is about innovation ecosystem. We are in the AI Summit.

A lot of startups are having fundamentally disruptive ideas on both decentralized renewable energy, as well as the way you manage generation, transmission, and others. The fourth is about financing. How will all this financing and de -risking be done? Because not all places have a lot of venture capital or commercial loans and equity possible. We are in the process of creating a new industry. We are in the process of creating a new industry. We are in the process of creating a new industry. We are in the process of creating a new industry. We are in the process of creating a new industry. We are in the process of creating a new industry. And finally, there’s a global dimension where International Solar Alliance is involved.

What are going to be the interoperable standards? Because the world is not united on how all of this will be done. So that’s a lot about AI for energy. But there’s also an equally important element of energy for AI. The world’s largest sources of increase in electricity consumption right now are only two. Data centers and cooling. Some of it is going to happen through electrification of cars, EVs as well. Now, 70 % of all data center demand today is US and China. But in times to come, it’s increasing by almost more than 50%. A lot of it is going to happen in developing countries. And we’ll hear some of that in addition to global elements. A lot of that is also having a lot of innovation that will renewable energy provide that energy.

Can 24 by 7 solar and storage provide cost -competitive energy to some of these data centers, whether they’re small or hyperscale, hyperscale being above 100 megawatt? What’s happening on innovation on cooling? We will hear some of the experts on the private sector who are trying to come out with a lot of innovation on that and what happens on the ecosystem. Obviously, today’s data centers are consuming a grid equal to Spain right now, and it’s going to double every three years. So this is a very important segment. Without further ado, I’m actually going to go probably to the esteemed panel. I’m going to request Mr. Raghav Chandra. Sir, you have been part of the government and now teaching. When you look at this big…

element of powering AI, how do you see it?

Thank you, Ashish. You’ve done a fantastic job in a short time period covering the larger macro issues connected with this sector. Friends, as we gather here in a nation racing towards digital sovereignty and sustainable growth, I want to emphasize and putting on my academic professorial hat, the single greatest constraint on AI’s future, which is not algorithms, not chips, but it is energy for AI -based data centers. And, you know, I’m going to mention a few such instances. In late 2024, Mark Zuckerberg made a confession that stunned his employees. A nesting colony of bees had torpedoed. Meta’s plans to open the world’s first nuclear -powered AI data center. That single environmental snag exposed their deeper vulnerability that Meta’s AI strategy depended on a single resource that it did not control and command, which is electricity.

Power outages and energy shortages have increasingly disrupted major tech companies’ operations, particularly as AI -driven data center demands strains global grids. There has been another very famous incident of March 29, 2025. A sudden loss of utility power in Google Cloud’s Columbus, Ohio unit triggered a critical failure in the uninterruptible power supply UPS batteries that created a major havoc for society. Several hours. This caused a cascading outage of six hours, in fact. Over 20 services were hit. Various customers experienced degraded performance or total unavailability, affecting cloud -dependent apps and websites globally. No direct apology was, of course, issued, but the event underscored energy reliability in a big way in an era of AI growth. In September 2019, utility power failed at one data center in Amazon Web Services, AWS’s North Virginia zone.

Backup generators activated but ran out of fuel after about an hour due to faulty automated refueling systems exacerbating the blackout. And it affected about 7 .5 % of the volume of… Apps and databases and some customers lost data permanently. Backups weren’t in place. Services like Slack and Netflix saw major ripples. And this has happened not only with Google Cloud or with Amazon AWS, but it has happened with companies such as Microsoft’s Azure, which suffered a major setback in 2018. It has affected TikTok, that’s ByteDance’s new USDS joint venture, causing widespread system failures. And what it underscores is the need for ensuring that there is suitable energy availability for data centers and that there is suitable backup for data centers.

Otherwise, you will not be able to have high powered. So energy guzzling, AI based data centers, which are the basic, basic. unit for AI to be implemented across the board for simplifying and for making and achieving our goals of ensuring that we have AI which is responsible, ethical, efficient, and which can do our job effectively. There is one county in the U .S. which my friends here on the dais would be aware of, of Ludon County, Virginia, just outside Washington, D .C., where data centers now outnumber people in density. And this 40 -square -kilometer area of computer server farms is Christend, the data center capital of the world. It hosts about 200 operational facilities, and another 100 or so are coming up.

Their peak draw is nearly 3 gigawatts. That’s enough to power a small country. Over 70 percent of the global Internet traffic passes. This is the clear area. What brought Ludon and its implications to the world’s notice was the massive outage at Amazon, causing tripping of crucial banking services and various social media companies. In Ireland, their data centers consume already one -fifth of the nation’s electricity, more than all the urban homes combined. Data centers traditionally began as largely in -house centers for proprietary computing data storage. They have since evolved, and today they are largely remote facilities or networks of facilities owned by cloud service providers, housing virtualized infrastructure for the shared use of multiple companies and customers. They need tons of electricity.

With all the power -hungry hardware and cooling systems, a data center today uses, higher -density racks, and whereas earlier… the data center typically used something like 150 to 300 watts of electricity per square foot. Today these higher density racks can consume as much as 100 kilowatts per cabinet which equates to 10 ,000 watts per square foot. And therefore a data center power problem can have global ramifications for the company. AI is supercharging data center boom that will recharge global energy systems. Global data center electricity consumption today is 415 terawatt hours. That’s about 1 .5 % of the world’s total consumption of electricity. And by 2030 it’s predicted to be nearly 945 terawatts or 3 % of the total consumption. So AI is not a side story. It’s the main driver with accelerated servers growing 30 % annually.

in the United States, which is the current epicenter. Data centers use 176 terawatts in 2023, or 4 .4 % of the national electricity. Projections are staggering. That’s like adding, in fact, the entire power demand of countries like Australia or Spain. So, you know, when we look at powering AI, we have to look not just at the upstream issues of creating the requisite demand, of creating the requisite power supply, but the other factors which come into play are the downstream effects and the hidden costs of progress. Environmental, if we rely only on fossil fuels to bridge the gap, emissions soar, so you have the debate between thermal and between renewable. which my colleague here will talk about. Big tech’s scope to emissions are already up 30 to 50 % since 2020.

Globally, data centers could claim 40 % of new fossil generation if clean supply lags. And so, while on the one hand, AI can help to accelerate decarbonization through optimal strategies and with intelligent working, but at the same time, the very fact that they are power guzzlers, they have an environmental issues which is inherent, and therefore there is a need for choosing a virtuous path. They have economic and social costs, while power prices are spiking. For instance, in the US, in and around areas which have data centers, the power cost has gone up significantly. In fact, wholesale electricity has jumped 200 to 250%. . in five years in certain areas, and the households are feeling that pinch. There is an issue of reliability.

Grids weren’t built for this. Voltage swings in Virginia have already tripped dozens of centers. In a warming world with rising AC loads, blackouts aren’t theoretical. They’re a governance failure waiting to happen. You have the equity issue. Who bears the burden? Communities near data centers face noise, heat, and land -use conflicts. In developing nations such as India, the digital divide widens if energy access for AI crowds out basic needs. So there’s a need for ingenuity when we’re dealing with this issue, and efficiency has to be the best weapon for dealing with the larger social, environmental, and other issues connected with this. And, of course, you know, in India, a lot is happening about which we’ll talk about.

But there is, indeed, a moment of great… happiness that AI is powering us, but there is also need to be concerned about whether we will be able to power AI effectively and whether we will be able to effectively and efficiently manage the downstream effects of powering that AI effectively. Thank you.

Thank you so much, Raghavji, for the different elements of sustainability risks for the society. Nathan, your opening statement especially from the cooling perspective.

that keeps these northern Virginia, as an example, data centers from adapting to more efficient and effective technologies. But when you’re starting with new builds, with white space technologies, you have the opportunity to actually build for the future instead of build for the past. And so that, to me, is the most important element as to how we’re going to solve powering AI in the future.

Thank you, Nathan. I’m sure you educated a lot of us in terms of what’s really happening on the cooling side on the innovation. Vineet, over to you, that from one of the leading renewable energy developers, how do you see?

Good evening, everyone. So I see AI as one of the biggest opportunity. For the renewable sector, historically, people believe that renewable is intermittent, which it is. It is difficult to predict when the sun shines and wind blows. So we needed the technology which can help us intermittent power dispatchable at 15 minutes interval so that the grid can operate in a stable environment. So what AI has helped that with the help of a lot of climatic data, which your weather department collects, company like renewable companies are collecting, defense department collects. And then you can get real time data from low earth orbit satellites. If you use all of them in the right way, you are able to predict using AI that what would be my generation like.

And then you go a step further and you can schedule and dispatch that power like a conventional thermal power would do. So that makes AI for energy and energy for AI. And that empowers the grid to have always on. Clean power. which is the uniqueness India offers. So let me tell you, friends, when India started adding solar and wind some 15 -16 years ago, we didn’t even have 5 megawatt of operational asset. And this year alone, India is going to add 50 ,000 megawatt of solar and wind capacity, making us the second largest green energy player besides China. And what it gives power to India is that, like the previous panelists were saying, in the U .S., in Malaysia, even in Ireland, which used to be the data center capital, every country started charging some surcharge on powering the data center.

But the reality of life is that there are not going to be 50 megawatt or 100 megawatt data center. Now we are talking about 500. 500 megawatt, gigawatt data center because the… compute requires so much of eating as Nathan has explained. Without impacting the society and affordably if you have to do, India is the place. And the reason I say that we are blessed with abundance of sun, wind and water. So using the pumped storage because of our geography, we are actually getting a natural ability to do storage. And largely in most of the states, sun and wind are complementary in nature. So what happens is using sun and wind alone, you can generate 14 to 18 hours of power and then you complement it with pumped storage and battery.

And if you combine with the AI and you build your AI stack properly, you are looking for round the clock green power. So India is the perfect location India is adding 50 gigawatt It’s not competing with the normal consumer. India has a lot of very good policy where using green power, they are able to move even farming activity from the night shift to the day shift. So and our per capita power consumption is one of the lowest in the world. We are less than 1500 kilowatt hour per year per capita. So if India has to become a Vixit Bharat, you can’t become Vixit without data. And data is the new oil. And unfortunately, what is happening today is that we are we have 1 .4 billion people and out of which a billion people are connected.

And we are one of the cheapest data connectivity package in the world. So we are. The largest user of YouTube in the world, almost 700 million. user on the YouTube is from India and we are the largest content creating economy whether you take Insta, whether you take YouTube, you take any social media. It’s a repeat story even on WhatsApp we are more than half billion user. And all of this data as previous speaker was saying resides in some other countries. So why should we generate so much of data and the data should reside in any other country because probably earlier we didn’t focus to use all this abundance of energy and power that data center and now today the scenario is it makes economic sense in US.

Now you cannot get any power before 2030. All even the gas machines are sold out. So if you look at the grid waiting time in the US is typically 7 to 8 years. Permitting you can get during that time but if the world has to adopt TI at a massive scale India offers that opportunity where we can set up multiple gigawatt data center. We can provide them green power using solar, wind and storage. And we actually have a very unique situation. Unlike US or Europe, India has a single grid. You can insert the power in Rajasthan and can pick up in Mumbai in real time basis. And India has invested heavily into the grid. And we continue to grow that national grid where the whole country is connected.

So the best location for solar, best location for wind, best location for pumped storage and battery can bring power to the data centers in Mumbai, Chennai, which are already connected. So the latency does not become the bottom line and it becomes the ideal choice. What is needed is probably more data sovereignty type of act. Indian user content has to be located in India by certain time frame and so that developers can plan for the grid they can plan for the large data center capacity and can bring that to light so it’s one of the greatest opportunity for India Indian ecosystem is purely geared up for that and on top of that we have if you look at even in the AI probably more than 25 % of talent resides in India and that talent currently is working for other countries so they will be based in India, work for India and provide services and intelligence to the rest of the world and that’s the way moving forward

Great so let’s have a little bit of a discussion and I do hope we get time for one or two questions so there was a little bit of questions that, Raghavji, you talked about, but a lot of optimism on both sides. I will ask a question combining two elements, which are important. One of it relates to the whole policy and regulatory landscape. Is India, or for that matter, developing world’s policy and regulatory landscape conducive for promoting data centers? I think, Vineet, you talked about the importance of data, the policy and regulatory landscape related to data sovereignty. Even Africa, I remember, was thinking of having a legislation like Europe, where the data for that particular continent or that particular country should be within that region.

But then there is also a policy and regulatory landscape for discovering price of power for data centers. India believes it’s very competitive. U .S. is struggling with the cost of providing power. Power probably is a limiting factor rather than the Nvidia chips. So that’s where the U .S. is. The second element is on innovation. I think you spoke about it, Nathan, but we’d like to hear what would an innovation landscape for cooling look like? Is it a lot of startups? Is it a lot of some of the larger companies doing some process efficiency? What would this innovation landscape look like? I want to request each of you to think about and say what would a policy and regulatory landscape change and an innovation landscape change accelerate both the speed and the cost of what meeting the demands of data centers look like?

Raghav.

So in the Indian context, you know, the stakes for us, as Vineet mentioned, with all the opportunity and the resources that are available to us in terms of land, in terms of water, in terms of the skilled manpower, the opportunities are enormous. And the data center capacity is all set to explode. Today, it consumes about one gigawatt of power today. We are expecting it to reach about eight or nine gigawatts by 2030. And it’s continuously growing. We have ambitious states like Andhra Pradesh, which can effectively be called the data cities or data states for the country. We have a coal -dominated grid, which India has fortunately allowed to continue in a very pragmatic way. We have rising cooling needs from extreme heat.

And as Nathan mentioned, that some of our states can have a power usage efficiency or effectiveness, which can be extraordinary. Because of all the heat, whereas ideally it should be. one which is the perfect index and we also have a net zero ambition of ensuring that we have complete renewable focus non -fossil fuel based energy dependency to reach by the year 2070 which is our global commitment which is I think again a very bold and generous commitment of India but the biggest issue that I find in this entire landscape if you ask me is about the ease of doing business in India and I am not being skeptical but having been an administrator who has been a managing director of the state industrial development corporation the state investment corporation the managing director of the road corporation the urban development principal secretary chairman of the national highway authority and various other such positions Now when I sit back and I’m on six company boards, I realize that the biggest bottleneck in India today is the lack of synergy between the states and the center, between the departments of the government and essentially between the states and the center.

And if India has to move forward to achieve this huge target that it has set for itself for ensuring that we become the data center country for the world, that we exploit our entire human resources, that we exploit our land resources, the solar energy that we have, we must have, you know, apart from the regulatory schemes, et cetera, and for the regulatory. On the regulatory side, much is being done. For instance, in the latest budget, we are all aware that how the finance. Minister announced the scheme for ensuring tax exemption for data centers that are set up in India with foreign collaboration for the foreign component part of the investment and for their revenues. Lots happening on the renewable energy front.

Lots happening on the various data centers that are being set up. However, lots needs to be done in terms of getting synchronized coordination, ensuring that the best technologies are brought in. One of the points which Nathan made was about leapfrogging and ensuring that India should capture technologies which other nations have faultily or by mistake adopted. We can certainly skip that and go on to the best technology. Water in the days to come is going to be a very, very big and critical issue. Thank you for India. And therefore, using liquid coolants and solutions such as that for cooling are going to be extremely important. And this has to be realized not only by the central government, by the states, and by everybody who is working in the field that they must facilitate ensuring that these things are adopted in a positive manner.

I had an example of a foreign company which the other day was talking to me. And they had signed an MOU with a particular state government for a huge amount of data centers to be established there. And they said that, you know, we are struggling. We’ve made eight presentations and we haven’t been able to move forward on that. That’s the kind of thing which with the best intentions and with our prime minister being so proactive that we should really have proactive chief ministers, everyone getting down to business, and using the large number of experts who are available all around to explain to them the best technology and moving beyond perhaps even L1 to be able to get the best configurations on the ground to ensure that we are not only efficient but we are effective.

Great. So a lot of potential but work to be done on ease of doing business, center state coordination and also from innovation a big potential for Indian companies to innovate on cooling, liquid cooling especially given water constraints. Nathan, over to you.

Yeah, I’ll comment on that innovation because that is innovation is the foundation that the IT industry is built upon and it’s built upon the idea that any one individual or small group of individuals can create an idea that changes the entire multi billion dollar industry itself and those who don’t innovate end up falling off the map. You know, you don’t talk anymore about AOL or Ask Jeeves or companies like that, and maybe we’ll say the same thing about Meta or Microsoft or, you know, Google or Amazon someday. Who knows? Because that’s the nature of the industry. And so as we look into the future, I think the innovation is going to require these smaller companies who are able to take risks and think bigger, especially around these cooling technologies.

And that’s what’s already happening today is we’re seeing people who are thinking outside the box of what we’ve normally considered to be advanced cooling technologies. Today, when we talk about advanced cooling that’s being deployed, what we’re really talking about is moving from that air -cooled ecosystem to just a simple liquid cooling ecosystem, which was developed in the 1960s for the Apollo space mission in the United States by IBM. And it’s been used for all of those years, including in the 1960s. If you’re a gamer at home, it’s been used in those large desktop gaming systems. And so this is an old and proven technology. You basically use… ethylene or propylene glycol mixed with water and you pump it through a pipe and it touches a cold plate on top of the hot chip and it captures it in liquid and moves away.

And that is a very simple and easy way of capturing heat, but it has limits. And what we’re facing is the limit that that liquid, as it leaves the chip, is getting so hot that you then have to have some coolant, some way to cool it back down. And that uses an incredible amount of electricity to cool that water back down, to use chillers on the roof of your data center to chill that water back down. And so the delta between the heated water, glycol, and the chilled water has to continue to get bigger and bigger and bigger, which means you have to cool that water lower and lower and lower using more electricity, so you eliminate the efficiency.

There’s now technologies emerging, and this is what my company is focused on, that is very similar to the way we cool. You can cool air in an air conditioner or in your car or in your refrigerator, and it’s called a two -phase technology, and basically what that means, instead of pumping liquid around… and it’s staying liquid, it actually, the liquid boils and vaporizes and that change of phase that is from a liquid to a gas is 10 to 20 times more effective and efficient at capturing heat. And that technology, though, is being spearheaded by small companies and those small companies will get bought up by large companies and they’ll be adopted into the ecosystem. And so expect to see that.

Expect to see the same basic use of refrigeration or refrigerants that we have today and we’ve been using for a long time, but using them specifically within the IT load of a data center ecosystem. That allows us to get those PUEs, that utilization efficiency ratio, not 1 .5 but 1 .05. You see that, I mean, that’s a massive step function increase in efficiency, which means the power generation doesn’t, doesn’t have to be strained nearly as much. And so I think that’s where the innovation is really going to come in the next three to five years.

great I think on the lighter note I’m always it’s baffled but amusing that the gaming industry was the start of GPU and now the cooling as well it’s fascinating how gaming industry is responsible for the AI revolution but lot of space for small companies if you have on the innovation side Vineet what do you think?

Gaming and best actually because the large batteries requires the same amount of cooling so the way I see innovation happening across the board is when the knowledge and cross industry expertise starts fertilizing and for that to happen you have to start creating local ecosystem see we can’t be sitting on the fence and be solving and innovating consistently that you are doing in theory but when you are building large data center of gigawatt scale you can find solution and use those skill because the similar challenge comes when you design the clean room. So how do you combine the expertise of building a clean room of millions of square foot with the expertise which is required for cooling the batteries and the expertise which is required for power usage efficiency in the data center.

How do you combine those skills and build the solution which is good for India where the humidity in some of the cities where optical fibers are terminating through the sea is large and how do you balance it out. So you have to use the external environmental data also to customize your PUE efficiency. So we see that efficiency is possible at all levels whether the ceiling height should be 6 meters or 8 meters. How close to India. So India is in that sense is fortunate that we are building those expertise locally without being building those expertise locally. building those 100 gigawatt off data center. In Morgan Stanley did a study. There is a $4 million opportunity cost for the power.

So they are saying the battle for the AI is no more compute. And it’s no more intelligence. It’s the power. Power is the biggest challenge. And there is a lot of innovation which is happening on power sector in India. You gave a good example of P2P trading using AI. And the policy in India is quite open on open access. So when I give power to the grid and I’m taking it out, I’m getting the power in the real -time basis, which is very few countries are able to do globally. And we have to account for on the monthly basis. So that gives a flexibility to the data center, which you always want. clean power and they want 24 by 7 365 days reliable power that is what is available in India and I agree with Raghavji ease of doing business is not similar across the states in the country but that’s why government of India is doing stack ranking of the state so today you can’t be just dependent on one state like look at Maharashtra the kind of support they are providing today if you want to build data center is amazing like permitting land everything is fairly streamlined and on top of that they incentivize so I think government has got it not every state is on the same page that if you have to become a developed nation your data is the biggest enabler if you have to win any kind of manufacturing battle data is the biggest enabler like if you look at today even our financial data most of the software companies whether it is Oracle or SAP or Microsoft they want the data to be on the cloud and those data even your financial data now even SAP you can’t do ECC everything goes on HANA on rise which is on the cloud so you buy the space either from AWS or Microsoft because they have only partnership with those two so even the 40 ,000 odd companies which are on these ERP softwares in India where are their data going and so the opportunity wise I think India because of its own need will innovate consistently ease of doing business is a challenge and that’s where there is an opportunity to continuously work with the government on transparently on your challenges and suggesting a solution which is not benefiting one voice is a And then third is understanding the nuances of how the application layer is working across the industry and educating government also that why should they have the data localization initiative.

And I see all of this getting combined and India becoming probably the third largest country where the AI adoption and data center would be one of the enabling block for the future growth. Thank you.

So a lot of optimism. I did promise one question. I have one question space only. So please go ahead. If you can identify yourself and have a brief question.

My name is Umesh Prasad Singh and I’m an associate member of Indian Institute of Public Administration. Sir, my question is directly to you. In your paper you have mentioned about global ramification. that particular aspect of global ramifications are of both types that is positive and negative. With respect to that, will you just have the clarification on that note? I wanted to know on that.

When I said global ramifications, I’m talking of both essentially the downstream effects of focusing on data centers and the implications it has on the fact that it will have an implication for the environment because they are power guzzling, as Nathan mentioned, that today earlier we had data centers which were full of CPUs, today they are full of GPUs and you’re going into all kinds of even more complex computing units so because of the storage, the networking, they are becoming far more complex. So it’s going to have an impact on the environment because of the heat. That is generated intrinsically because of the data center, because of the environment that will be affected, because when you’re consuming coal to produce that power, you’re using water, that same water which could feed millions of people and pay the, you know, today we are not able to feed enough people for, provide adequate drinking water 24 by 7 to all our cities, yet you would have water effectively being used for the cooling of the data centers.

You will have social issues, because people today already for thermal power plants, they are creating issues where they find that their land, especially in the scheduled areas, et cetera, is being consumed for coal mining, so there are issues connected with that. Likewise, there are all kinds of social and environmental issues that are likely to happen. There are issues on the side of… You know, whether we, you know, what other implications it can have… So all these are things which are not essentially just localized, though they are local problems, but they will affect global companies which can have the benefit of India is that we can leapfrog in terms of technology. And hopefully, as one of the speakers earlier in the previous session mentioned, chips are also becoming more and more efficient.

So, you know, as they become, computing becomes more efficient, the chips become more efficient. So you will require a lesser amount of energy. If we can leapfrog, adopt the best technologies in terms of design and infrastructure, that again will be a great saving. So today, no nation is an island. Everyone is connected. And anything which impacts one nation affects the entire thing because data centers, if they are located here, as I mentioned, the case of the Ludon County outage, it affected billions of people all across. So it has a global ramification. while you have to think of your own benefit you have to keep an eye also on the impact whatever you are doing has on all across the nations and which is why when the Prime Minister talks of Manav, it is the human being who is at the center of it and the human being is not just you it is the larger mankind and the larger human community

Thank you, unfortunately we do not have time for any more questions but it’s pretty late I’m ending without summarizing but it’s pretty apparent huge optimism on the power of India and developing countries to meet the demand for AI both through solar storage, innovation on liquid cooling and of course the ecosystem with ease of doing business please join me in giving a big round of applause to all of them and thank you for staying very late thank you everyone for joining Thank you. Thank you.

economy. Saibal Chakraborty, Managing Director and Senior Partner, Boston Consulting Group. The moderator, C .V. Madhukar, Chief Executive Officer of CoDevelop. And Mr. Sangbi Kim, Vice President for Digital, World Bank, who will be joining us in a bit. Thank you. right I will let the moderator take it forward

thank you so much thank you Arjun and thank you all to the panelists the last session of the last day is a bit challenging so we will try and keep this focused and at the end if you have any questions we have time it would be great to have any questions if you want to have any discussions I think we have heard a lot about AI and DPI in the last four days I don’t want to belabor the point I think what we can celebrate for sure in India is that in terms of DPI and the thinking that we take to the world and to our own problems is amongst the best in the world.

And as we look at the journey on AI, which is just beginning for most of the world, what I see is if I look at the US, for instance, there is one spectrum of conversation, which is AGI and beyond. And on the other end, there is a despondency and worry and concern about jobs and privacy and a whole bunch of other important considerations. I think where India is, is somewhat different. India is saying, look, there will be a Chinese model, there will be an American model, there will be a whole bunch of other innovations that are going on. What do we do to embrace and use all of this for our benefit? I think that optimism and sense of can do and must do, I think is very exciting, and I think it’s been palpable in the last four or five days.

So to discuss the power of AI and DPI and AI in DPI, we have a wonderful panel as Arjun has introduced already. I will start with Dr. Hans, if you don’t mind. I think you represent a national government and you are living through these choices on a day -to -day basis. As a pragmatic government that has to think about sovereignty, inclusion, safeguards, what are the main considerations that are on top of your mind?

Thanks for the question. You’re right, I think those three words are very key. When you’re talking from a government perspective and therefore national, inclusion, integrity or the safety of the citizens. and also safeguards. I’ll just run through these three very quickly. Why inclusion? When it’s government and you introduce a new capability, you have to be sure that that capability does not increase divides, that it actually reduces divides. So you have to be very sensitive on the inclusion angle. And, for instance, AI, together with DPI, can stretch inclusion through its voice -first capabilities, we talked about cloud -first, API -first, et cetera. Now we can really seriously talk about voice -first and also the translation capabilities.

And accessibility has to be also broadened in terms of multi -modality and also, where necessary, include a human in the loop in the service delivery cycle as well. So some of the inclusion dimensions. Then more to integrity, which is a tougher one, And I think that’s a really good point. And I think that’s a really good point. An important point when you’re looking at integrity is to start from the premise that AI will not redefine DPI. AI would, or at least where we look at it from now, maybe I’ll be wrong in six months from today, but the DPI foundations must be in place first. DPI should be mature, and your approach to implementing DPI should be mature, and then you apply AI as a scaffolding on top of that foundation to accelerate your build and delivery.

So what are those foundations? Clean data and data maturity, maturity of your data architectures, clean registers, for instance, also reliable APIs, APIs which are not susceptible to cyber attacks, et cetera, and also the institutional capacity. The institutional capacity behind the DPI delivery. So these are foundations which should be in place. Now, on top of that, you apply AI and there are several unique features of AI which would deliver you super experience to the citizen. And I’ll come to that in a short time. The last one would be the safeguards, bias detection, and also the augmentation to consent. Because when you have AI systems in place, your consent could be AI generated as well. So you need to be careful about the augmentation you require, explainability, and human in the loop as well in terms of your safeguards.

So we need to be conscious that bias, opacity, at scale would mean harm at scale as well. So everything AI scales. So we need to be conscious that bias, opacity, at scale would mean harm at scale as well in terms of your safeguards. So we need to be conscious that bias, opacity, at scale would mean harm at scale as well in terms of your safeguards. Sovereignty is all about capability. It’s not isolation. it’s about building the capability in terms of being neutral and having a neutral capability in terms of vendors, in terms of cloud, in terms of technologies so that you have control, that the state has control over those building blocks and can choose across technology delivery parts as well as core technologies also data classification, protection, privacy basics, so I think if you go back to the foundations of protecting your citizens and their data you can’t go wrong and AI will only be an accelerator finally the experience angle, because a government would be very keen to deliver super experience and where does AI come in here I believe there are some specific strengths of AI when combined with DPIs and that is in terms of in particular API orchestration picking the right API to call at the right time also the fact that AI enables unconstrained scenarios so while linear methodologies would be constrained to say 4 or 5 scenarios when a citizen needs some help AI can do a billion scenarios and each scenario could be implemented using a unique set of API calls and unique set of DPI access and this can be orchestrated using AI so can AI make a big difference a leap forward in service delivery?

I believe so can governments and the sovereign use it or should they? Definitely but we need to be conscious of those 4 dimensions that I just mentioned of inclusion, integrity, safeguards, sovereignty all of these are important and we are still coming together to deliver a perfect experience

Thank you, Dr. Hans. I will give a bit of a breather to Sangbu. Thank you, Sangbu. Great to have you. I’ll go to Robert first and then… Robert, as you at UNDP have led a very important work on safeguards, worked on the Global Digital Compact, engaged a number of countries on sustainable development goals and how AI and DPI can be an accelerant to all of those outcomes if you want. From your vantage point as you look at the AI revolution that’s unfolding upon us, what captures, what’s top of mind for UNDP?

Yeah, no. So I think that the reason we have been so excited about digital public infrastructure as an approach overall is that it really does bring some very particular characteristics. And one of those, maybe the most important, is the population scale. and so it is something that can reach so many people so quickly if you get it right. We also have been learning that if you don’t get it right, then you can have problems and challenges at scale and so in the DPI space, one of the things that concern us the most is how do we ensure that as countries are building their DPI, how do we make sure that we are putting the safeguards in place?

And this is work that has been supported generously by Codevelop and Gates Foundation and others and has led over the last year and a half or so to the creation of a universal DPI safeguards framework which we’re now implementing or supporting a number of countries at national level to implement. But… But when we talk about that, what does it actually mean? And Dr. Hans referred to some of these important things. And so we are talking about the DPI as a whole. And so we are talking about the DPI as a whole. And so we are talking about the DPI as a whole. And it’s about what we’re learning is that the earlier in the process that you can start discussing the safeguards, the better off you’ll be down the road in terms of inclusion.

So if efficiency is your only metric, then you will probably rush ahead and leave people out. But if inclusion is your driving KPI, then you really need to make sure that you’re sitting down at the beginning and planning and designing with people in mind. When it comes to AI, then, it’s basically the same thing. And we need to be really careful, as Dr. Hans was saying, that we are putting the inclusion aspects, the safeguard aspects, at the center of our planning from the very beginning. And so that means, and you referred to a couple of these, but, you know, a multilingual platform, multimodal that can support people with disabilities, making sure that you are correcting for people.

So that’s one of the things that we’re doing. And I think that’s a really good thing. And I think that’s one of the things that we’re doing. And I think that’s one of the things that we’re doing. And I think that’s one of the things that we’re doing. And I think that’s one of the things that we’re doing. And I think that’s one of the things that we’re doing. And I think that’s one of the things that we’re doing. And I think that’s one of the things that we’re doing. And I think that’s one of the things that we’re doing. And I think that’s one of the things that we’re doing. and then of course making sure that data sets are there’s bias detection or you’ve got some understanding of your accuracy representation all of those kinds of things because when you’re going to layer AI into DPI as an accelerator well then you want to be sure that you’re on the right track and that people are considered from the very beginning with that and so I think that’s we see all the

Thank you Rob Can I go to Sangbu now as you see the evolution of DPI plus AI from the World Bank standpoint one of the things that we’ve often looked at is is having open data sets that can train air engines would be an important way of advancing the benefits of AI but as you know governments build silos one silo after the other how are you seeing this from the World Bank’s standpoint what is the next journey for the next 3 -5 years looking like in terms of getting countries to become more AI ready any thoughts on that would be great

DPI has a lot of aspects and characteristics this is one of the very productive and efficient way to ensure the interoperability even though it cannot ensure everything but we make some more effort to clarify some more interoperability capability so but it is basically what it looked like. So that’s why in AEI era, in order to prevent some siloed approach, DPI can play a very significant role, I would say. If you think about the AEI era, what is the relationship between DPI and AEI? If I just compare the DPI from the previous version of DPI, I would say DPI is more helpful for AEI era compared to the previous mobile era in many reasons. If you look back on our history of computing, we started from the computer and PC and evolved to mobile and evolved to AEI.

The trend is that it is from the very supplier -centric approach through the user -centric approach. We are evolving from the supplier mindset through the user mindset. The DPI is exactly just some of the really, really important tools to ensure that user -centricity. Because without identifying some good tools of users and some interoperability, it cannot really be achieved to fully support the user -customized things. So to me, the DPI and AI will be really important relationships. On the other hand, one opportunity… I also see through this trend… maybe DPI can be also very well upgraded quickly and efficiently upgraded through AI enabled technology so we used to collect all the data in a very manual way and then try very hard to streamline the governance of data sometimes very manually and sometimes by some intervention by the programmer but now we can see some old medical way so that we can quickly streamline all the DPI platforms so this is really a good opportunity for all of us from the World Bank perspective we really expect to see some more progress of this space

Wonderful Saibal as a senior partner at Boston Consulting Group you have a bird’s eye view of a lot of early thinking on AI led innovation around the world, especially in the private sector. And I’m also sure you’ve been observing closely the India trajectory of DPI in the last decade or so. What lessons from the DPI journey in India can we take to the AI era that might also propel private sector innovation to levels beyond if we just didn’t think about DPI as core infrastructure for AI?

So I think, firstly, India’s journey in DPIs has been a fascinating one. It makes me immensely proud that whichever country I go to, and I do work with quite a few South Asian and Southeast Asian countries, India is almost always seen as a benchmark in DPI, and now increasingly AI on top of DPI. I think, so maybe I’ll just answer your question in two parts, right? I mean, when we were building DPIs, starting with Aadhaar, you know, and then, you know, moving on to UPI, et cetera, the idea was always to build open population scale software, which can then trigger innovation, right? And that’s exactly what has happened over the last decade, right? The amount of innovation that has been built on these DPIs is mind -blowing.

I mean, India is now a country of 120 unicorns, and every unicorn, some way or the other, leverages the DPIs. So then, coming to the second part of your question, I mean, what lessons, right? So if you look at the way India is now thinking about AI, and as BCG, we have been privileged to be part of two of the very leading companies in the world, right? So we have done several seminal efforts, one with India AI Mission to build AI Coach, India’s national AI platform. and also with the state of Telangana to build the equivalent for the state of Telangana. Both of these happened last year. I think we are taking a very, very similar ethos, right?

So if you think about India, as I mentioned, there are 120 unicorns. There is no dearth of VC funding in India at all. However, 90 % of that VC funding actually goes into fintech and e -commerce. Very little goes into climate and sustainability. Very little goes into education. Very little goes into MSME relevant topics. So there is a gap, right? So what these platforms are trying to do? And then similarly, access to data within private sector, there is good quality data. But the biggest source of data in India is the government. The access to government data is still at a very nascent state. Quality of data and access to data. And then, of course, the biggest thing in AI, which is compute, right?

I mean, access to compute is what makes or breaks a startup. So the way in India, the way I see it, the way we have started thinking about AI platforms, and I’ll use the word platform, it treats AI as a shared public infrastructure. Just like DPI was a shared public infrastructure, it treats now AI as a shared public infrastructure. If you look at India AI mission, more than 38 ,000 GPUs are now available at, you know, less than rupees 60 per hour, which is less than a dollar per hour. So if you are a startup, very early stages, working in your garage, suddenly GPUs have become a bit more affordable for you, right? That’s genuine shared public infrastructure.

Government data, it’s early days, it’s very early days, but government data is being provided access to through platforms like AI Coach, or in Telangana through the TGDX. to solve the financing problem. How do you channel financing into socially sensitive sectors? The central government and the states are thinking of building fund of funds, which actually then, you know, encourage VCs to co -invest, focusing on those socially sensitive sectors where they would normally not invest in. So if you think about it, the ethos is very similar to when we were building DPIs. How do I create shared capabilities centrally, which then can trigger an entire new wave of startups, you know, and therefore a market ecosystem, just like the DPIs did.

So that’s how I see the journey.

That’s great insight. I think it seems from what you say, there’s a beginning of a new innovation cycle for the private sector cycle, and we’re looking forward to what comes out. I just wanted to, we have about 10 minutes left. have a somewhat common forward -looking question to all panelists. As the economists say, in the long run, we’re all dead. But what is long run in the AI ecosystem? Is it five years? Is it three years? Because it’s so hard to predict everything. Every day there’s something new happening. So I don’t know, Dr. Hans, if you’re okay going first on this forward -looking question. The question I would ask of you is, as a relatively smaller island nation, how do you expect to leverage this wave of AI innovation over the next three years, maybe five years?

What steps are you anticipating? How are you preparing yourselves to leverage this power that we have to advance our development outcomes?

So there’s a plus and a minus of being small. Let’s start with some of the challenges. Being small, you’re on the wrong side of the AI divide unless you’re economically in a very powerful position. Say for example, Singapore is an outlier, a small country with a lot of economic power and therefore attracts investors, attracts talent and the sighting of business. Country, one of the challenges we have in Sri Lanka is one, getting that minimum level of sovereign AI infrastructure in place, having the ecosystems around it, retaining our talent. Sri Lanka has very good talent but retaining and developing the talent for Sri Lanka is a challenge but one that we are confident that we can deliver.

So future of AI I think will depend on the market. We’ll also depend on the people. We’ll also depend on the trust in us. The institutions like the data protection, the institutions as well as the laws, and Sri Lanka is very mature on this front. So on the trust side, a smaller country can execute precisely with laser -sharp focus and therefore has a strength. Talent side, again, it’s something to focus on. Now when it comes to the marriage with DPI, I think it falls onto the positives of being a small country because the ability to implement modular systems in a neat and flexible way, in a way that these blocks themselves will evolve, on the confidence that you have a strong trust environment, gives you the ability to build in AI where, like I mentioned earlier, AI sits.

On top of a solid DPI, a mature DPI frame. So I feel the future, AI future will of course be very close, can add that extra piece of experience, lower cost, faster, and more flexible, meaning that it can address multiple scenarios through digital twin and other such AI constructs to deliver citizens a very customized, I’m from the service industry in the past, so I use the word customized, but citizen -specific experience. We’ve been tracking the learning about the focus of the new government and their leadership and the presence of AI in the world. We’re working with the government’s leadership, making big advances on DPA and AI, and looking forward to much exciting stuff in Sri Lanka in the next two to three years.

Thank you for those comments, Dr. Hans. Rob, could I come to you and think about, you know, you’ve gone through the process over the last couple of years with DPI safeguards. You have the Global Digital Compact. I know there’s a lot of work you’re doing on AI safeguards. Moving away from safeguards, I wanted to see how you are envisioning the developmental role of UNDP leveraging AI in the next three years. I guess three years is long term, but anything you can say that would be helpful for us.

No, absolutely. So I think there’s a couple levels. So there is an internal to the organization level, which is how do we ensure that UNDP itself has capabilities for leveraging AI to maximum effect. And so there’s a kind of a base level of work that we’ve done internally to the organization, upskilling programs, investing in some, you know, making sure foundation model capabilities are available, working in some SLMs, et cetera, et cetera. Then there’s the layer of working across the kind of verticals that UNDP has, whether it’s environmental action, governance programs, energy, et cetera, et cetera. And so how do we embed AI solutions and thinking and approaches into those verticals? And then there’s the picture of how are we going to support our country partners?

And as you said, we’re engaged in quite a few countries already on AI transformation support, and it’s kind of looking at ecosystem pieces. Do countries have that mix of elements that Dr. Hans was referring to? Do you have the compute accessibility? Do you have talent? Do you have the data available? And so on and so forth. but the one thing that we’ve announced in this summit, during this summit is an exciting partnership with Xstep and a number of other players and we’re one of those players on something we’re calling 100 Pathways or Diffusion Pathways and that is looking kind of like more of a use case driven approach and finding over the next few years 100 different pathways to scaling responsible use of AI along different use cases and it’s something that we’re really excited about because I think we need that to complement the ecosystem support

That’s so exciting to hear because I think there will be a lot of iteration discovery and innovation to discover those 100 pathways to actually add value to people on the ground looking forward to what comes out of that Thank you if you were to look at the last decade of how development banks have funded and looked at digitization, and now if you look at the three years ahead of you, what might the World Bank and MDBs do differently to be ready, help countries become more ready for the AI era that you haven’t been doing in the last decade or so? Any thoughts would be great.

So good news is that from the Internet network point of view, the coverage is pretty good. So even in sub -Saharan Africa, more than 90 % of area of sub -Saharan Africa is covered by three -plus generation mobile tower. But the issue is that we are really struggling with lack of demand. How are we going to fully utilize this by creating some more value and profit? So now we are modifying some approach to really think about creating demand through government program, through developing some use cases. That’s why we are just keep highlighting the importance of small AI. Small AI is not really a small thing. So it is really about how we can really change the lives of our people.

So our approach is a little tweaking to the user -centric and demand -driven things. That’s our approach.

That’s great. And I think MDBs and the relationships that they have with countries can make a big difference in how the evolution and the benefits of AI will come to people. So looking forward to what comes out. Saibal, I know I started off by bucketing you as a private sector guy. But I also know you’ve been thinking deeply about government policy that enables or doesn’t enable innovation and growth. So as you look at the next few years, even from private sector innovation lens, what government policies might propel the innovation ecosystem? to serve the underserved populations around the world. Any quick thoughts, Saibal?

So, see, I think, you know, the government has a very tricky balancing act, right? I mean, when we were going through this entire experience of building AI Kosh, there’s obviously, for very good reasons, a lot of sensitivity around sovereign data, what data to expose, and if exposed to whom, right? Equally, without sharing of data, I mean, the reason why AI has taken off in a big way in some sectors is because internet came into picture 30 years back, and internet has been pumping, you know, billions and billions and gigabytes of data, right? So AI has something to chew upon. Now, if we have to do the same with government data, then that data, you know, needs to be exposed in a controlled manner.

So my sense is that from a policy standpoint, how do you actually provide that access to data? I mean, walking that tightrope where valuable data is made available to the innovators while not compromising on sovereignty or safety. I think that is one of the policy areas the government has to look at. Specifically for a country like India or similar countries which operate in a federated model, the center can do only so much. The real action, as we know, in a country like India happens at the state level and we have 30 plus of them combining states and union territories. So at the state level also, similar policies and institutions have to be set up. So Telangana, for example, has set up a Section 8 public sector undertaking to drive AI, right?

That creates the kind of focus and the agility that you will need to keep pace with this technology and do some real work at the grassroots. So my suggestion would be create those. Accountable institutions. who can anchor and drive AI and amend policies around data to make sure that people get access, the innovators get access while not compromising on sovereignty. It’s not an easy thing to do, but yeah, those would be my words.

Thank you, Saibal. I think the role of institutions, both for safeguards, but also to enable the innovation ecosystem haven’t been more important than it’s now more important than ever before. I think it’s hard to summarize this conversation, but I will say that I think we’re at the cusp of something extremely important and something very potent in some ways and can unlock a lot of opportunity for innovation. billions of people. Especially, I think, the segment of the population that was left out of the digital revolution because voice was not the predominant way of interacting. I think AI opens up that window and hopefully will drive much more widespread adoption and usage by common people around the world.

Looking forward to the next few years and thank you very much to our panelists for a wonderful discussion. Thank you all. Thank you.

Thank you so much to all the speakers. We will just have one memento being given by the organizing team. We made it. Thank you so much for being a part of the India AI Impact Summit. Just to tell you that the expo will be open tomorrow. People still want to come in and people are still not tired, but we are done for today and the sessions are done. Thank you so much. Thank you. Thank you.