Anirudh Suri

Full Transcript

Govts, Big Tech, and India’s Bid to Shape AI’s Future, with Amitabh Kant, Ex G-20 Sherpa, and Wilson White, VP, Public Affairs, Google

TRANSCRIPT

Anirudh Suri: A couple of days ago, a clip went viral by Ruchir Sharma. He said that India is not necessarily a winner in the AI race. Is that an accurate assessment?

Amitabh Kant: The race is never won by those who launched the race. India will be the determiner of its own fate in AI.

Anirudh Suri: Today we have a special episode where I have two of the main actors in the world today: government and big tech. Mr. Kant has been one of India’s most prominent bureaucrats. He served as India’s G20 Sherpa, and prior to that, as CEO of NITI Aayog. Wilson is Vice President of Government Affairs and Public Policy at Google. When something like Iran happens for a global big tech firm like Google, what does that trigger internally?

Wilson White: No one company and no one country can do this alone.

Anirudh Suri: What role do you see big tech firms like Google playing in India’s own AI journey?

Wilson White: Google’s own AI ambitions are not achievable without India. Our fates are intertwined. If we can solve for India, we’re really solving for access globally. India has the talent, the skills, the IT companies, and the digital public infrastructure.

Amitabh Kant: Big tech is important, but please use India to co create. Use India to drive R&D. Use India for tech diffusion.

Anirudh Suri: If you weren’t an IAS officer, what would you have been? And for those interested in the intersection of public policy and AI, what advice do you have for them?

Introduction

Hi, and welcome to today’s episode of The Great Tech Game podcast, where we bring you some of the world’s best thinkers at the intersection of technology, geopolitics, and global affairs. Today we have a special episode with two people representing two of the main actors we see in the world today: government and policy, and tech firms , especially big tech firms. I’m delighted to have both of these perspectives together on the show.

I’m excited for this conversation with you, Mr. Kant, and with you, Wilson. Mr. Kant, for those of you who might not know , though I’m sure most of you do , has been one of India’s most prominent and dynamic bureaucrats. He served as India’s G20 Sherpa for the year India held its presidency, and until last year, in 2025. Prior to that, he served as CEO of NITI Aayog.

I’m a big fan of the op eds I read from Mr. Kant, which span a wide range of topics, and where, as I like to say, he injects a dose of common sense into policy framing in India and into India’s strategic landscape. Mr. Kant has ultimately been a very inspiring figure, I think, for many of us who see a vibrant public sector as key to India becoming a Viksit Bharat by 2047.

To my left is Wilson White. Wilson is Vice President of Government Affairs and Public Policy at Google, based in Singapore. We’re excited to have you here in India. Thank you, Wilson. He’s the global head of government affairs and public policy for a few of Google’s core areas, and also handles the APAC region. Before joining Google’s public policy team, he was a patent litigation lawyer.

I’m going to start with something top of mind for everyone in the world right now: Iran.

Iran and Operational Resilience

Anirudh Suri: Geopolitics is, of course, at the core of what I cover on The Great Tech Game podcast, and the Iran crisis has shown in many ways that geopolitics can impact not just governments but also big tech firms. While many of us are familiar with how governments respond when a crisis like this hits , foreign ministries, national security advisers, various economic departments kicking into action , Wilson, I want to start with a question for you. When something like Iran happens for a global big tech firm like Google, what does that trigger internally? What kind of actions, thinking, and concerns does a company like Google focus on the moment a crisis this big and consequential is triggered?

Wilson White: It’s a good question , one of the most important we have to tackle given today’s geopolitical situation. I think today’s geopolitics requires companies like ours to go beyond normal operational security and move into operational resiliency. When a crisis like Iran happens, our focus immediately goes to protecting our personnel. When Iran happened, we immediately relocated our people to safe zones so we could continue operations. The next focus is preserving the integrity of our services and ensuring continuity. We distribute our digital infrastructure globally so that an outage in one area doesn’t take down the entire global network , that’s been a significant area of focus for us.

One thing we’ve noticed in these geopolitical crises is the information dimension. What we’ve been doing recently, and it played out again in the Iran conflict, is using specialized AI to mitigate or even neutralize disinformation, to ensure vital citizen information , like where to go for safety, or accurate maps and citizen services , reaches users during these very vulnerable periods.

Anirudh Suri: This is the first time, at least in recent years, that I’ve noticed a country specifically target big tech firms and infrastructure , whether data centers or other forms of infrastructure. When that happens, are you as a big tech firm relying on governments and their security infrastructure to protect you, or are you building parallel security protocols to protect your own infrastructure?

Wilson White: I think this is why we partner closely with governments , it’s a bit of both. No one company and no one country can do this alone. We partner with governments around the world to distribute our digital infrastructure globally, so a crisis doesn’t take down one company or one country. We’re very much doing this in partnership with governments.

Anirudh Suri: Any key lessons from this crisis , things Google will do differently going forward?

Wilson White: There are always learnings. One high level lesson that has really shaped our senior leadership’s mindset is that we need to think about resiliency as the key motivator behind our investments, how we build our architecture and infrastructure, where we make those investments, and how we partner with governments and other stakeholders. Resiliency needs to stay at the forefront.

Anirudh Suri: Mr. Kant, let me turn to you. When this crisis happened, from a government standpoint , I understand you’re not currently part of the government , is the response governments typically mount very different from what Wilson described, or quite similar? How did you see the Indian response this time?

Amitabh Kant: From a government perspective, it’s very important that citizens are not impacted by conflicts happening across the world. It’s important to note that a period of peace and stability since the end of World War II is being replaced by conflict , we’ve had conflict in the heart of Europe for over four years, conflict in the Middle East, and now conflict in West Asia. Second, global supply chains have essentially broken down, replaced by protectionism and tariffs. Third, efficiency in supply chains is now being replaced by resilience.

Anirudh Suri: It’s being replaced by trust.

Amitabh Kant: It’s being replaced by redundancy. So from a government perspective, it’s critical to ensure citizens’ lives aren’t impacted. From India’s perspective, it was essential to ensure the continued flow of oil and gas, and the continued supply of LPG, so people’s lives weren’t disrupted , energy payments, shipping, all of it is critical, alongside the larger goal of ensuring citizens aren’t hit by inflation. The overall macroeconomic picture is very important, and I think the Government of India has done extremely well to ensure the common citizen continues life normally and keeps enhancing production and productivity.

Anirudh Suri: The geopolitical conflict you’re referring to has been going on for years now and has become the norm rather than the exception, which has clearly pushed governments to think deeply about supply chain resilience and redundancy. But are we moving fast enough to reduce the core dependencies a country like India might have on certain regions or sectors? And, much like I asked Wilson, what are a couple of new lessons from the Iran crisis , things that weren’t as clear from, say, the experience of Ukraine or other conflicts?

Amitabh Kant: One key lesson from this is that we lived in a globalized world , you’d import, add value, and export. But the present world has demonstrated that you need to be self reliant in several areas, energy being one. You can’t be a large importer of energy, especially when India has a huge ability to become self reliant through renewables. That’s critical.

Second, you can’t depend on other countries for critical mineral processing. Critical minerals are available, but processing has been monopolized, so you need to become a champion of processing yourself. Third, you can’t be entirely dependent on semiconductors. You can be part of the supply chain, but you need your own chip capability , the world will keep growing and expanding around chips. Fourth, the future of the world has to be driven by clean energy technologies. These are the four areas India is working on.

India has driven renewable energy in a very big way , today it has about 270 gigawatts of renewable energy, and almost 50% of its electricity comes from non fossil fuel sources. It has onboarded close to twelve companies for semiconductor chip manufacturing , India was largely a chip designer before, but now twelve companies are manufacturing chips here.

Anirudh Suri: We just announced two more, yesterday.

Amitabh Kant: Yes, just yesterday. And third, India has really pushed the critical minerals mission in a big way, so processing eventually happens domestically. Fourth, I think the key is that the world will have to be driven by clean energy , you can’t sustain this much computing power otherwise. The world will need far more sophisticated software with less computing. Today, the energy consumed globally by artificial intelligence and computing is already more than the entire energy consumption of Japan.

So you need clean energy to power this, and that’s where India has a huge advantage , its climatic conditions, having one grid across the country, its ability to keep scaling renewables, and policy conditions that have attracted private capital, venture funds, and private equity into its renewable energy sector. India is the only country in the world that has championed renewables at this scale, and computing will increasingly have to be powered by renewables and clean energy in the future.

Anirudh Suri: Did the Iran crisis change your company’s internal calculations around energy planning?

Wilson White: I couldn’t agree more with Mr. Kant. Renewable energy is a key part of this. When we talk to governments around the world about how they should think about AI readiness, there are four pillars we discuss. The first is having power first strategies , robust clean energy strategies and streamlined clean infrastructure permitting to support the massive demand for compute.

Assessing India’s AI Ecosystem

Anirudh Suri: That’s a great segue into the second theme I want to cover with both of you: an assessment of India’s AI ecosystem. India recently pulled off the India AI Impact Summit in Delhi.

Amitabh Kant: It was a very successful summit.

Anirudh Suri: A massive summit, one that in many ways announced to the world that India is serious about its AI ambitions. It brought together a great group of people. But at the same time, Mr. Kant, a couple of days ago a clip went viral on social media from Ruchir Sharma, an investor based in New York who covers India’s domestic politics as well as its economy. He suggested that, unlike the tech boom of 1999–2000, this time around, as far as AI is concerned, many foreign observers have taken the view that India is a loser in the AI race , that from a global investor’s perspective, India is not necessarily a winner. Mr. Kant, is that an accurate assessment? Why or why not?

Amitabh Kant: My view is that this is not an accurate assumption, for several reasons. First, he’s looking only at frontier models , and yes, as far as frontier models go, the US is undeniably in the lead, way ahead of every other country, with China a close second. Second is computing power , Nvidia is way ahead of others there too. But the critical thing about AI is that you can’t build frontier models without data.

Anirudh Suri: Correct.

Amitabh Kant: Twenty percent of global data is in India. That’s clear. Fourth, you can’t do this without the data centers of the future, and for data centers you need energy , renewable energy will come from India, and the data centers will follow. Fifth, and to my mind most important, it’s not just about frontier models , it’s about the diffusion of technology. In the end, the game will be won by whoever diffuses technology and deploys use cases for the betterment of society: how you transform education, improve learning outcomes, improve health outcomes, improve nutrition. Those use cases will come from India and will be built in India for the Global South. India’s huge advantage is that it has already built the digital public infrastructure model, which has demonstrated its ability to create open source, open API, globally interoperable systems.

The last point I want to make is that this AI race has only just begun. The race is never won by those who launched it. Clive Sinclair built the first computer , he didn’t win the race; he went bust.

Anirudh Suri: Yes.

Amitabh Kant: The race was won by Steve Jobs at Apple. Alta Vista built the first internet search engine , it went bust. The race was won by Google in the long run. So we’re in for a long race, and I think data, energy, and diffusion will be the deciding factors.

Wilson White: Let me come in on this, because what Mr. Kant has laid out is remarkably consistent with how we think about it. I mentioned the AI readiness framework we discuss with governments earlier. The first pillar is power first , robust clean energy strategies. The second is having AI corridors, strategic hubs that bring together land, energy, tax incentives, and regulatory sandboxes for experimentation and deployment. India is doing that , we recently announced a strategic investment in an AI hub in Visakhapatnam. That hub is part of these AI corridors and is strategically positioning India at the top of the game globally.

A third element of the framework is open public data sets. As Mr. Kant said, the data India has is vital to any AI system aiming to lead the world in solving real challenges in education, healthcare, or science. India is leading the world in having that kind of rich public data and making it open.

Lastly, one of the things India has already done , layering AI on top of frameworks like UPI or Aadhaar , really democratizes the technology. One of the things I saw in full form at the AI summit was how the country is thinking about this technology democratically: how do we give radical access to it? That diffusion point you’re making, Mr. Kant , India is leading the world in that, and every AI summit before this hadn’t put access at the center of the conversation the way India did. That’s going to be the real differentiator. Of course, companies like ours build the infrastructure and platforms, but that leaves room for countries to develop their own unique applications for solving problems for their people, and I think India’s third way is already leading the world in that regard.

Anirudh Suri: I want to come back to the role tech firms play in helping countries build both their AI infrastructure and their broader ecosystems. But before that, Mr. Kant , the strengths are clear: a lot of data, since India is a digitally connected country with a massive number of financial transactions through UPI and heavy internet usage across sectors, plus the renewable energy potential you mentioned. But where are the gaps? Clearly there are gaps, since we’re not yet at the cutting edge of diffusion or innovation. Where does India need to move quickly to realize its full potential?

Amitabh Kant: Another point I want to make, and this is critical: the world is going to run on trust. It will operate within a democratic framework, and it will favor countries with pro innovation regulatory policies. India’s policy on data and AI is extremely progressive , it’s not a backward looking policy, it’s very pro innovation. In the long run, India will be a great beneficiary of technology diffusion. But India needs to raise its game on computing power. We’re working through the AI Mission with roughly 3,000 GPUs, but look at the scale of the transformation taking place , we need much more computing power for research institutions and startups.

The second thing is that we need to build a more vibrant, dynamic R&D ecosystem.

Anirudh Suri: Absolutely.

Amitabh Kant: Why does this happen in the United States? Because at the university level, students work closely with their professors, and there’s a huge flow of venture capital.

Anirudh Suri: That’s true.

Amitabh Kant: And that’s how great startups get built. India needs to build a very vibrant deep tech ecosystem, which is why the government is now putting about one lakh crore into the R&D ecosystem through public private partnership. It’s starting a deep tech fund of funds to attract a lot of venture capital into this space.

Anirudh Suri: Yes.

Amitabh Kant: That will really drive deep tech innovation in a much bigger way. India has moved up the ladder on patents , we now register patents much faster than the US or Japan. But those patents need to be commercialized in the long run. Building this dynamic R&D ecosystem is critical to building the intelligent society of tomorrow. India has the talent, and what the Stanford AI Index shows is that in the last of its nine years, India actually became an absorber of technology. It used to send talent out to the rest of the world, but last year a lot of AI talent came back to India to drive many of the 2,700 global capability centers here. Imagine , the world’s AI models are being driven out of India’s GCCs, powered largely by talent that’s returned home. That’s unparalleled.

Anirudh Suri: Helped a little by Google, maybe?

Wilson White: We see it a lot , our biggest footprint outside the US is right here in India, because the talent is here. The talent piece isn’t questionable at all. I’ll add one more thing to what Mr. Kant said, which I agree with entirely: AI and leadership alignment matter. AI can’t be relegated to IT departments alone , it needs a top down, whole of government approach, and I’ve seen that here in India. It’s a nationwide effort, from the top down, treating AI as a national strategy , not just in terms of compute power, but also skilling and preparing the workforce for an AI powered economy. These priorities already exist in India, and I think those investments will pay off exponentially.

Anirudh Suri: Wilson, let me also draw on your perspective running APAC policy for Google to get some lessons from what other countries in the region are doing in AI. We’ve spoken at length about India’s strengths and ambitions , but looking across APAC, are there things Malaysia, Taiwan, Japan, Korea, or Singapore are doing that India could learn from and incorporate into its own AI strategy?

Wilson White: Yes, and a number of these elements already exist in India. Let me call out a few things that, from our perspective, have been very helpful in powering AI ecosystems, some of which are already being done here. First, as Mr. Kant mentioned, having a pro innovation AI regulatory framework , the new frameworks India has put forward are very forward looking and pro innovation, and that has to be the baseline. Then, when you get into implementing regulations in specific sectors, you have to stay true to that pro innovation approach.

Where we’re seeing other countries in the region do well is in revisiting legacy legal regimes that may need updating for an AI world. Mr. Kant, you mentioned patents and intellectual property , that’s an area with a long history of legal frameworks that may need rethinking in terms of the societal trade offs we want AI’s benefits to deliver. Copyright is another example , some countries are exploring exceptions to copyright law to allow local investment in AI and local models that deliver benefits in specific ways for their citizens.

Anirudh Suri: Your background as a patent litigation attorney isn’t going to waste, then.

Wilson White: It’s being put to use in some ways , I probably should have studied more copyright law, honestly. Other countries are also building central repositories of public data. Access to high quality data is one of the most important things you can have to power an AI ecosystem, and some countries have opened up their public data sets, centralized them, and let the whole ecosystem , universities, the private sector, government agencies , leverage that data to solve problems in key sectors.

Last, some countries in the region, particularly in Asia, have leveraged regulatory sandboxes much more aggressively. We don’t fully know how these things will play out as we deploy AI solutions, and without the ability to experiment, we won’t learn what works, what doesn’t, and what guardrails are needed. A healthy use of regulatory sandboxes lets us do what we say at Google: be bold and responsible. We can push the boundaries of this technology and solve big challenges, but do it responsibly, so people benefit while staying safe.

The Role of Big Tech in India’s AI Journey

Anirudh Suri: One of the reasons I was excited for this conversation is that a country’s AI strategy clearly can’t be built by government alone. AI is largely driven by private tech firms today, global ones at that , though one hopes startups from around the world, including India, will become the big AI and tech firms of the future. In the meantime, Mr. Kant, what role do you see big tech firms like Google, and others, playing in India’s own AI journey right now? What are they doing well, and what would you like to see them do that they aren’t doing yet?

Amitabh Kant: And I’ll be taking notes, by the way.

Wilson White: It was intended for you to hear.

Amitabh Kant: Good feedback, then. First of all, I’m a great believer in big tech firms being critical players in driving innovation. If India’s ambition is to become a 35 plus trillion dollar economy by the time it turns 100, in 2047, it’s very important to bring in the best technology companies , they’ll bring technology, capital, and the best talent from around the world, and they’ll drive innovation in India. That’s important.

From an AI perspective, it’s very important that we don’t isolate ourselves , we need to be part of the global system. And it’s important to understand that when we talk about sovereign AI, sovereign AI does not mean isolationism.

Anirudh Suri: That’s true.

Amitabh Kant: There’s no such thing as being fully sovereign in an AI world. Even the United States depends on ASML for lithography machines. The rest of the world, including China, depends on Nvidia’s GPUs. Much of the world depends on Qatari gas for heating. It’s an interdependent, globalized world , we all depend on each other to drive innovation. Big tech is important, but big tech coming to India shouldn’t mean using Indian data only to drive innovation elsewhere. Please do not make India a colony of the rest of the world , that would be reverse colonization. Use India to co create. Use India to drive R&D. Use India for tech diffusion. Use India to transform India’s learning outcomes and health outcomes. Use India to change the lives of Indian citizens.

When you do that, you’ll be creating models not just for India but for the next four and a half billion people worldwide who will move from poverty to the middle class over the next decade. Use the power of AI to genuinely change lives across the world, to lift people above the poverty line , that’s what AI must do. That’s the critical role big tech companies like Google must play, and they’ve done it with DeepMind’s research in pharmaceuticals and biosimilars. Similarly, when they apply the power of AI in a country like India, they’re transforming the lives of 1.5 billion people here, and through that, the lives of the next four and a half to five billion people in the world. That’s what Google must do in India.

Anirudh Suri: That’s a great call to action from Mr. Kant about the range of things a company like Google, or tech firms broadly, can do. If you were to hear that call and do an honest self assessment, are there clear areas where Google needs to do more, and areas where it’s already doing a lot?

Wilson White: I couldn’t agree more with what Mr. Kant said , his vision is 100% aligned with our company’s vision, both globally and for India specifically. It’s core to our mission to organize the world’s information and make it universally accessible and useful, and that access piece is exactly how India is leading the world. It’s not just the 1.5 billion people here , India is the oracle for the entire Global South, for the next five billion people who will come online in a digital, AI powered economy. We see India as the leader in that. If we can solve for India, we’re really solving for access globally, and that’s been our commitment. The investments we’re making are designed to make us a strategic partner to the country in that journey.

We think we have a lot to offer , a full stack approach to AI, some of which will be useful to India on its own journey, and we want to partner with the country on that. But India will be the determiner of its own fate in AI. India has a reservoir of talent and component parts like renewable energy that set it apart, and we’re very grateful for the opportunity to be part of that journey.

Anirudh Suri: Let me push a little further on that, Wilson. With your investment in the strategic hub in Visakhapatnam, is enough being done on some of the other pieces Mr. Kant mentioned , for instance, setting up much bigger R&D labs? As far as I understand, there isn’t a DeepMind equivalent initiative out of India yet, but correct me if I’m wrong.

Wilson White: We’ve actually done that with DeepMind here in India.

Anirudh Suri: So are Google’s future R&D plans keeping India at the core, or is that still unclear?

Wilson White: Google’s own AI ambitions are not achievable without India , our fates are intertwined, which is why we’re making investments in frontier research here. Google DeepMind has set up shop in India, and that operation is continuing to grow, with ongoing investment. But it goes beyond that , it’s about thinking through the Indian reality, including the multiple languages that exist here, and how we use AI to surface languages beyond English or Hindi so everyone has access to this technology. That learning comes from partnering closely with the Government of India, understanding those contexts, and bringing them to life.

Sovereign AI vs. Global Interoperability

Anirudh Suri: Let me move to a different piece. We’ve discussed the AI capacity India is building and the role tech firms play in that, but at the core of it all is the geopolitics of AI. Countries and companies realize this is central to their competitive advantage, and they want to protect it. The US does this vis à vis China, and Google will want to do it vis à vis a Meta, an Amazon, or a Microsoft. Mr. Kant, in a world where the geopolitics of AI will clearly determine what a country or company can or can’t do, are you worried about access to frontier models being a concern for a country like India, even as India remains an important piece of Google’s, or any large tech company’s, overall AI strategy?

Amitabh Kant: India has the talent, the skills, the IT companies, and the digital public infrastructure to create better and better models and application layers with less computing power over time. So the challenge for the world will be that AI gets commoditized in the years to come , the frontier models will get commoditized.

Anirudh Suri: Absolutely.

Amitabh Kant: We shouldn’t be in a position where our own data is used to build models that are then sold back to us as products at high cost.

Anirudh Suri: Right.

Amitabh Kant: It’s important for India to build very good application layers. You can’t be a champion of everything in the world. India’s challenge is to become a champion of good applications and good use cases, but also, within the overall AI stack, to be a champion of some of the nodes in that layer. India can’t be a champion of frontier models or of computing power, but it can be a champion of energy, a champion of talent, and a champion of diffusion.

Anirudh Suri: Yes.

Amitabh Kant: So there are three areas where India must become a champion: diffusing AI technology through use cases that transform citizens’ lives, becoming a champion of talent , since the latest Stanford AI Index shows India is absorbing talent from the rest of the world , and demonstrating that it can do AI with less computing power but better software.

Anirudh Suri: Is Mr. Kant right that frontier AI models will get commoditized? Because if that’s true, it raises questions about the massive investments big tech firms like yours are making, and the returns you’d expect from them.

Wilson White: I agree with most of what Mr. Kant said , we need to find one thing to disagree on.

Anirudh Suri: I’m trying to get you two to agree on something, not disagree.

Wilson White: But that combination of talent and building on top of, for example, our open source models like Gemma , which is a frontier model that’s been open sourced , India’s huge reservoir of talent will take that much further than a frontier model alone could. India specific applications can easily be built on top of that.

What I think is worth underscoring, a point Mr. Kant made, is the need for global interoperability, because the conversation around sovereignty is very loud right now, and understandably so, given the unpredictability in the geopolitical space. Every country is asking how to protect its citizens, its talent, and its economic prosperity , those are the right questions. The answer, though, isn’t to isolate, as Mr. Kant said, and that’s worth underscoring, because achieving true global access to the benefits of this technology requires global interoperability. It requires countries to work with countries, and companies to work with companies.

Anirudh Suri: Let me move to something we’ve touched on, Mr. Kant, but haven’t explicitly addressed: is there an inherent tension between what Wilson described , global interoperability and an open diffusion model , and a sovereign AI approach? As India tries to build itself as the “use case capital” for the world, and as it builds its own sovereign AI models, like Sarvam, which was showcased well at the India AI Summit, we also clearly want to adopt global AI models to benefit farmers, students, and patients in India. Will that fast adoption of global AI models, open source or proprietary, ultimately make our own sovereign AI push meaningless? Is there an inherent tension there, or can these things move in parallel? China, for instance, has taken a different approach , even as the US and its big tech firms develop frontier models, China isn’t pushing to adopt those models to benefit its own citizens; it’s building its own in parallel. India isn’t taking that approach. Do you see an inherent contradiction here?

Amitabh Kant: Two things. First, it’s very important for India not to close itself off from the world.

Anirudh Suri: Correct.

Amitabh Kant: Innovation is happening in one part of the world, and India must benefit from that innovation to transform lives. Second, AI will not be fit for purpose if it isn’t multilingual. It won’t be fit for purpose without use cases for education, health, and nutrition. It won’t be fit for purpose if it isn’t low bandwidth. It won’t be fit for purpose if we don’t make it open source, open API, and globally interoperable. That’s very important. We’ve done this with digital public infrastructure , our belief and our learning from DPI is that technology benefits citizens, and we should do the same with AI. We should build DPI for artificial intelligence. That’s really critical, and in the long run the world must benefit from it. The power and innovation of AI must eventually lift vast segments of the population above the poverty line , that’s the bottom line.

Wilson White: Absolutely.

Anirudh Suri: Wilson, is there an inherent tension between countries wanting to build sovereign AI stacks and what Google or other tech firms have in mind?

Wilson White: No, I think it’s important for governments to think through their own national AI strategy , they have issues they need to solve for their citizens, and that instinct toward sovereignty is understandable. How it plays out, though, is where I agree completely with Mr. Kant: a sovereign approach to AI shouldn’t be isolationist. We pride ourselves on offering the full stack, from chips to infrastructure to platforms to applications, so we have a lot to offer. A country may pull some of that from Google, build other parts natively, or pull other parts from different private sector players. It’s up to the country to decide which aspects of that stack fit into their national strategy , that’s the sovereign piece: the country decides where it sources each part of that stack.

Anirudh Suri: Correct.

Amitabh Kant: And if I may add, I think one of the key pieces of the word “sovereign” has to be the freedom of decision making.

Anirudh Suri: Correct.

Amitabh Kant: The autonomy of decision making for a country is very critical , even if you’re dependent on ASML for certain things, or the US, or tomorrow perhaps Singapore or Taiwan for others. The autonomy to choose which models, which infrastructure stack, and what energy sources to rely on , that will be critical for countries.

Wilson White: The applications have to be built.

Amitabh Kant: Which applications, which node of the value chain a country wants to specialize in , that decision has to be taken by India, for India.

Rapid Fire

Anirudh Suri: Let’s start the rapid fire round. Mr. Kant, first with you: UPI or AI , which will have a bigger impact on India’s GDP by 2030?

Amitabh Kant: AI, for sure.

Anirudh Suri: Sovereign AI , is it a must have or a nice to have for a middle power?

Amitabh Kant: Sovereign AI that isn’t isolationist, and that’s used to transform citizens’ lives.

Anirudh Suri: Compute, data, energy, or talent , which is the scarcest resource for India right now?

Amitabh Kant: Computing power.

Anirudh Suri: On regulation , does a first mover advantage exist, or is a wait and see approach better?

Amitabh Kant: Pro innovation regulation is the way.

Anirudh Suri: One thing India’s tech strategy needs to change immediately?

Amitabh Kant: We should become a product nation and focus on R&D.

Anirudh Suri: Biggest dependency India needs to remove , tech or energy?

Amitabh Kant: Push for renewable energy, become a champion of it , that’ll help remove your tech dependency. Target 1,500 gigawatts of clean renewable energy by 2030.

Anirudh Suri: Wow , that’s another great call to action. One thing you’d like India’s private sector to do differently?

Amitabh Kant: Become far more innovative, far more R&D focused. Spend more on R&D , that’s critical. Every Indian company must become more R&D oriented.

Anirudh Suri: One thing you’d like government or the public sector to do differently?

Amitabh Kant: Create a very large deep tech fund , a minimum of two lakh crore , to attract venture capital to support the startup movement.

Anirudh Suri: Institutionally, does India need a tech envoy or an AI envoy, the way it has a G20 Sherpa or a climate envoy?

Amitabh Kant: No. Every CEO in India, every minister, must be an ambassador for AI , not just one person, but each of them a champion of it. This is a general purpose technology; it must get embedded in every sector of India.

Anirudh Suri: Moving to some personal questions. If you weren’t an IAS officer, what would you have been?

Amitabh Kant: I would have become a startup founder.

Anirudh Suri: You would have been great at that. Favorite activity, outside of all the hats you wear , where would we find you when you’re not on a podcast, in a government meeting, or on a board?

Amitabh Kant: On the golf course.

Anirudh Suri: The golf course , and the mountains too, I believe?

Amitabh Kant: Yes, I trek, and I’d prefer to be there.

Anirudh Suri: Advice for Gen Z , should they learn to vibe code with AI today, or focus on communication and negotiation, which I think you’re a champion of?

Amitabh Kant: Use AI to do the coding , you don’t need to be a big techie. Today’s models let you do that. Focus instead on the softer skills: communication, negotiation.

Anirudh Suri: Let’s move to you, Wilson. Open source or proprietary , which wins the battle for the Global South?

Wilson White: Open source.

Anirudh Suri: The biggest advantage Google has over its peers in AI , and don’t just say Sundar Pichai.

Wilson White: Well, that is the number one answer , Sundar Pichai is such a visionary leader, that’s a real differentiator. But I’d also say the fact that we’ve been investing in AI for a very long time, and as a result we have the full stack. That separates us from our peers.

Anirudh Suri: We’ve focused a lot on AI today , is there a focus area for Google beyond AI right now?

Wilson White: AI is it. AI is our focus.

Anirudh Suri: Compute, data, or energy , which is the scarcest resource for Google today?

Wilson White: I’d actually say memory , not those three. The demand on compute is creating demand for memory, and we have a real supply problem with storage capacity. A lot of our consumer products, the things people use day to day, will see rising costs because of the memory supply problem. It’s an area we need to pay close attention to.

Anirudh Suri: Silicon Valley, Bangalore, Beijing, or Singapore , where do you expect the biggest killer app for the next billion users to come from?

Wilson White: The question is whether it’ll even be a “killer app” the way we think of apps today , I think AI is going to change that. It may be an AI agent that solves these big challenges, and I’d bet that’s more likely to come from Bangalore than the other cities you mentioned.

Anirudh Suri: One thing India is doing right in its tech strategy, and one thing it could do differently?

Wilson White: India is doing a lot of things right , the investments are right, the focus on renewable energy is right, the focus on creating economic and strategic hubs that attract investment is right, and the regulatory approach is moving in the right direction. The one thing I’d add is that continued focus on regulatory predictability and certainty will help. These are big investments being made, and that level of predictability will encourage even more.

Anirudh Suri: Some personal questions. Since you spend so much time in APAC, favorite city in Asia for food?

Wilson White: Food wise , wow, I wasn’t expecting that. And I’m sitting in India right now, so I need to be smart about this. Indian food across the board is incredibly diverse, like the languages here. So Indian food at large, I’d say.

Amitabh Kant: Try some good Kerala cuisine , appam and stew.

Wilson White: Stew, okay , that’s my favorite.

Amitabh Kant: And fish moilee.

Wilson White: Fish moilee , incredible food. Outside of the Punjabi, North Indian food you find all over the world, I think it’s Kerala food you have to try.

Amitabh Kant: And you have places in Delhi where you can get good versions of it too.

Anirudh Suri: If you weren’t at Google, Wilson, where would you be?

Wilson White: I’d be advising startups.

Anirudh Suri: Final question , advice for Gen Z or younger people watching this about careers in public policy and AI, something you’re doing for those interested in that intersection.

Wilson White: I think this is the tip of the spear in terms of how we help companies and countries navigate a very uncertain future. AI ultimately will be used for good, and those pursuing that intersection of tech and policy will be the ones making sure that ends up being true.

Anirudh Suri: Perfect. This has been great , you’ve both been very generous with your time and insights. I can’t thank you enough. It’s been wonderful to have both a tech firm’s perspective and a government policy perspective in the same conversation. Mr. Kant, Wilson, thank you for sharing how governments and the private sector can work together. I think the more tech firms and governments can focus together on solving the problems of the common person on the street , the farmer, the student, the patient, as Mr. Kant always says , the better off we’ll all be, and the more innovative and strategically interdependent we’ll become. Thank you both for your time, and I hope this was as useful for you as it was for our viewers.

Amitabh Kant: Thank you.

Wilson White: Thank you.