What They Said: Puneet Chandok of Microsoft on AI

India is now ready for its AI century. And we at Microsoft are so proud to stand tall with India to make this happen.

India’s advantage will not just come from frontier innovation. We’ll obviously innovate like nobody else can. But it will also come from diffusion—from classrooms to boardrooms. India will rise when a billion Indians rise with AI, and that’s our vision.

We want to be copilots to India’s AI transformation, and we will make that happen. That’s what we get up every morning for and get to work.

And I think there are two macro forces at play. Geopolitics—which we’re all trying to make sense of—but also AI. We’ve got the largest infrastructure build-out happening on the planet today. Even in India, we’re building the largest hyperscale presence.

I think these two forces—geopolitics and AI—working together are making this world very unfamiliar. I call this the world rearranging itself. The world is rebooting.”

Chandok outlined the structure of his address:

“First, I’ll talk about frontier AI. Where are we headed with AI? I’ll give you my five predictions. I’ll provoke you. I might be wrong, but I’ll give you some thought-provoking ideas on where we’re headed.

Second, I’ll talk about India—what we’re doing from classrooms to boardrooms, how we’re building the tracks before the trains.

And third, I’ll talk about frontier builders—each one of you—how you’re using AI. I’ll tell you some stories we should all celebrate: how Indian businesses and builders are putting AI at the center of everything, and more importantly, how they’re doing it.”

Intelligence on Tap

“Intelligence is not a property given only to humans. For the first time in the history of humankind, we potentially have the ability to manufacture intelligence. When compute becomes cognition—when compute becomes intelligence, or log of compute becomes intelligence—that’s the pursuit of AI.

We’re trying to build machines that possess human-equivalent intellect—and in some cases, superiority.

We’re entering a new reality today.

If you want to put intelligence to work today, what do you do? You hire a team of people. You bring them together. You get a project manager. Fully loaded cost. Three months, four months—and then you see some impact.

Imagine a world where intelligence moves from this team structure to something you can buy like electricity.

Imagine a couple of years down the line—you’re sitting in the same room, there are sockets around you, and you can plug yourself in and get 50 IQ points. That’s the world we’re entering.

Intelligence will move from something scarce and expensive to something abundant—available on demand, available on tap.

Every business today is a bundle of intelligence, a bundle of expertise. Once intelligence is on tap, you’ll have to go back and rewire your business. That changes everything—how we run businesses, how we run people, how economics works.

So if we’re moving to a world of intelligence on tap, how does the world look different?

Let me give you my five predictions.”

Five Predictions

Prediction 1: Digital Colleagues

“We’re used to physical colleagues. We’re entering a phase where digital colleagues will join the workforce.

AI so far has been about answering questions and reasoning. The next wave of AI will be true collaboration. The real unlock comes when AI moves from a tool on your phone to a true teammate.

These next-generation agents are autonomous. They can goal-seek in unstructured environments, go do stuff for you, and come back and say, “Hey boss, I’ve got something useful.”

They have perception. They see what you see. They have cognition. And they have agency. They act with your permission, but not your involvement.

AI moves from tool to teammate to true digital colleague.”

Prediction 2: The New Bat Signal

“Batman fans—there used to be a Bat Signal in Gotham City. There’s another Bat Signal now.

Earlier, the question was: what cloud do I build on?

Now it’s: what model do I build on? What API do I start with?
The first dollars used to go to servers, storage, databases. Now they go to models and API calls. Whoever wins the developer’s first API call will build the next hyperscaler.”

Prediction 3: Outcome Economy

“We move from an inefficiency economy to an outcome economy.” The more time it takes, the more your lawyer makes. The more time it takes, the more your IT services provider makes. This model is broken.

AI will end billing by the hour. We’ll move to microtransactions. “Give me my forecast for next quarter.” That’ll be $20. That’s the world we’re entering.

Prediction 4: AI Factories

“Every business will need an AI factory.”

If the model knows everything, what do you know? Your knowledge, your context, your magic—that’s your advantage. Take an open-weight model. Bring your knowledge. Let it compound.

On employment, Chandok dismissed fears of job losses driven by AI. “AI will not steal jobs. AI will dissect jobs,” he said, warning that resistance to learning posed the biggest risk. “The real pink slip is not automation. The real pink slip is our refusal to learn,” he said, describing AI skilling as “guerilla warfare against irrelevance every day.”

Prediction 5: Jobs Will Be Unbundled

Will AI steal jobs? No.

AI will dissect jobs. It will unbundle them.

Every job is a bundle of tasks. AI will unbundle them, and we’ll need to bundle ourselves better.

We’re breaking the industrial-age model of learn once, earn forever. Learning used to be building a house. Now it’s pitching a tent.

The real pink slip is not automation—it’s refusal to learn.

This is guerrilla warfare against irrelevance every day.

Turning to India’s preparedness, Chandok said the country is laying the right foundation for the AI era. “AI for us in India is not just a technology shift. It’s an economic, social, generational profound shift,” he said, pointing to India’s digital highways and growing developer ecosystem as critical enablers.

Also Read: What They Said: Preeti Lobana of Google India on AI

India: Tracks Before Trains

“AI for India is not just a technology shift. It’s economic, social, generational.”

Our digital public infrastructure—the seven digital highways—are transitioning to the AI age. Aadhaar, UPI, our 22 million developers—they’re moving us to autonomous workflows, copilots, agents.

He highlighted India’s strength in technology diffusion, citing adoption across education and enterprise. “India is picking this up and diffusing like never before,” he said, noting that value creation would come from scaling AI usage across sectors.

Chandok cited Indian companies using AI to transform employee experience, customer engagement and business processes, including education, finance and tax services. He said these organisations were “democratizing intelligence but they’re putting the human in the center of it.”

A country becomes frontier when three things happen:

• Digital infrastructure becomes AI infrastructure
• Connectivity reaches everywhere
• Governance scales it

India has all three.

Diffusion matters more than invention alone. From classrooms to boardrooms, India is diffusing AI like never before.

Frontier Builders: AI in Action

“The ROI of AI is no longer elusive. We see it clearly in four areas: employee experience, customer engagement, process reinvention, and innovation.”

Examples bring this to life

UST is deploying 5,000 copilots across functions, preparing for an agentic future.

UST is doing 5,000 co-pilots, really thinking about HR, thinking about every function in the organization. Getting ready for an agentic future, really building this end to end. Incredible to see the impact and incredible to see what they’re building.

ClearTax is enabling 75 percent of gig workers to file taxes for the first time using AI—without exploitation.

ClearTax, one of my favorite examples. They’re doing tax filing for gig workers in this country, bringing AI into this. The impact that they’re seeing—75 percent of workers today are filing taxes through clear tax using AI without any exploitation, without any problem. 75 percent of gig workers filing tax for the first time. Incredible to see the impact already.

Genpact is reinventing finance end-to-end, redesigning processes rather than layering AI on broken ones.

Genpacked, again incredible story on end-to-end finance transformation, really looking at processes end to end—accounts payable, accounts receivable—doing it end to end and completely reimagining the process, not putting technology on an old process, looking at a process left to right and redoing it completely.

PhysicsWallah is using AI tutors to educate millions across India at an accessible price point. Questions that took hours now take minutes.

Physics walla, one of my favorite examples again, incredible to see how they’re putting AI to work for educating India across small cities in India, anybody, and the price point right now is roughly USD 50 per year, but they’re making it super accessible. 2 million students are using their AI tutor, their AI guru, to train themselves. Queries to your teacher used to take six to seven hours. Now it’s taking a minute. So students were waiting six to seven hours for the teacher to tell them what they did wrong and what they need to do better. Now it’s minutes. It’s incredible to see what they’re doing.

“The common thread? These companies are democratizing intelligence—while keeping humans at the center. This isn’t just about productivity; it’s about doing things we could never do before.”

AI must serve people—not the other way around. We’ve captured 60 such Hero stories in a book, which you’ll all receive digitally.

Lifelong Learning as Safeguard

He noted that current professionals may be the last generation to experience relatively stable, long-term careers, while future workers are likely to pursue multiple skill-based roles over time instead of following a single linear path.

Stressing the importance of lifelong learning, Chandok said automation itself is not the real threat to employment. Instead, he described resistance to upskilling as the true danger in an AI-driven economy. Comparing learning to an oxygen mask, he said professionals must continuously equip themselves to remain relevant, likening the urgency to Delhi residents’ lived experience of air pollution.

Closing Remarks

In his closing remarks, Chandok urged leaders and developers to think ambitiously. “If you’re not thinking big for India, if you’re not thinking big for India, we not doing justice to India,” he said.

He concluded by calling the present moment a defining opportunity. “India’s digital destiny is being written today and so is ours,” Chandok said. “And as I go back, right, when a billion people rise in India with AI, the nation is transforming. There’s never been a better time to build for all of us. You are the frontier. You are the future. Let’s start building.”

Also Read: What They Said: Vijay Shekhar Sharma of One 97 Communications, Paytm on AI

Abundant Intelligence, Skilling as Safeguard

Unmetered intelligence, human-supervised ‘digital colleagues’ and continuous skilling as the key safeguard amid big shifts will shape an AI-rewired future, Chandok said while outlining his top predictions, according to a PTI report dated December 22, 2025.

Chandok said AI has moved beyond hype, delivering real impact today. “…the next phase will be defined by how responsibly, inclusively, and thoughtfully we scale it,” Chandok said.

His predictions for the rearranging world and frontier AI include intelligence moving from being scarce to abundant, with compute increasingly translating directly into cognition for organisations, in an era of “unmetered intelligence”.

AI ‘agents’ will work alongside people, taking on tasks and reasoning across data, but humans will remain firmly in control, he asserts. Outcomes will be key to reshaping business model, and value creation will shift from effort and delay, to measurable outcomes. India’s digital public infrastructure enables mass AI adoption, turning national scale into a global advantage, Chandok notes, according to the report.

Roles will break into tasks, careers will become more dynamic, he reportedly said describing skilling as the primary safeguard in the AI era.

“When I reflect on the technology story of 2025, what stands out is how decisively India moved from experimenting with AI to putting it to work across core sectors of the economy. Across aviation, healthcare, financial services, and manufacturing, organisations began redesigning how they operate-whether it is Air India reimagining customer engagement, Apollo Hospitals supporting clinicians, ICICI Lombard reshaping core processes, or Asian Paints bending the curve on innovation,” he was quoted as saying in the report.

“What makes this moment especially powerful is the India opportunity. When digital public infrastructure meets AI, adoption can happen at population scale – from classrooms to boardrooms and from farms to factories. This belief underpins Microsoft’s commitment to invest USD 17.5 billion in India, to build the cloud and AI infrastructure, skills, and trust needed to turn momentum into long-term impact,” Chandok reportedly said.

“Skilling is the most essential form of resilience in the AI era, which is why Microsoft has doubled its commitment to equip 20 million people in India by 2030 with the skills needed to participate in and shape this transformation,” he said, highlighting that Jobs themselves are changing, and work is becoming more fluid, and the most durable advantage is the ability to keep learning.

This is a developing story, and more quotes and insights from Puneet Chandok will be added as they become available.