Satya Nadella's 2025 India AI Tour: A 3-City pitch amid India's AI ambitions

Nadella's three-day visit to India also highlights Microsoft's employee-centric approach, with Nadella engaging teams in cities where the company has major hubs.

Microsoft Chairman and CEO Satya Nadella's three-day visit to India this month wasn't just corporate PR, the itinerary mirrored India's urban diversity: Delhi for governance, Bengaluru for R&D, and Mumbai for business scaling.

It was carefully crafted as a response to India's digital ambitions, talent pool, and policy environment, amid fierce competition from rivals like Google and Amazon.

The city-run also highlights Microsoft's employee-centric approach, with Nadella engaging teams in cities where the company has major hubs.

Nadella's hype (India can win the AI race by adopting the fastest) positions the country as a counterweight to China in global tech.

This follows Google's $15 billion and Amazon's $35 billion pledges, turning India into an AI battleground that could create millions of jobs and boost GDP by 15-20 percent via AI, as per McKinsey estimates.

For everyday Indians, it means cheaper, localised AI tools in education and healthcare.

Also, read: Amazon upstages Microsoft’s India bet with $35-billion pledge

Nadella kicked off in India’s capital, Delhi, where he spoke the language of systems architecture and national capacity.

He unveiled a technology stack grounded in sovereign cloud options, local data processing, and contextual AI layers. The framing was not about Microsoft’s Copilot, but about India’s Copilot, built, trained, and governed locally.

Also, read: India’s ‘virtuous cycle’ for tech involving policy, market is “magic”, says Nadella

In India’s tech capital, Bengaluru, the tone shifted toward developers and software architecture.

Nadella highlighted the idea of an AI-driven software development lifecycle, where engineering starts from evaluation, not specification. Tools like Foundry, Copilot Studio, and GitHub Agent Board formed the backbone. Here, Microsoft was not pitching infrastructure but positioning a platform for innovation.

Also, read: Satya Nadella puts Agentic AI front and centre with Microsoft's latest model

In India’s financial capital, Mumbai, his tone was different again.

Nadella called data the most strategic asset in the AI era. He emphasised operational impact, unveiling MahaCrimeOS, a police-facing AI anti-cybercrime system that now lives in 23 stations and is scaling to 1,100.

He also highlighted that the stakes of AI are not just in code or infrastructure, but in service delivery and public trust.

“Nadella’s comment on technology and data sovereignty hits home as it is critically essential for India to build these technologies and products indigenously for becoming a 'Technology Product Nation', which will bring technology sovereignty and create a pathway for ‘Viksit Bharat 2047’,” Satya Gupta, President of VLSI Society of India and also a Member of the National Committee on Electronics Manufacturing, told Moneycontrol.

Also, read: Global giants Amazon, Google, Microsoft make multi-billion dollar AI bet on India

Nadella’s core pitch across each city ended with how the technology giant is seeing the next phase of growth in India through sovereign data handling.

On cybersecurity, for instance, Nadella in Delhi said, digital sovereignty and global intelligence need to be balanced. “When it comes to cybersecurity, it is an intelligence game,” he said, cautioning that “you don’t really want to fight a cyber war without global intelligence.”

Also, read: DoT notifies cybersecurity rules for telcos to tighten security 

Microsoft is currently rolling out in-country data processing for Microsoft 360 Copilot across India, Japan, Australia, and the UK.

The plan is to extend these controls across 15 countries in 2026, including Canada, Germany, Italy, Malaysia, Poland, South Africa, Spain, Sweden, Switzerland, the United Arab Emirates, and the United States.

This feature predominantly caters to the demands of governments and highly regulated industries.

Also, read: Quantum will break encryption, and AI might hand it the key

During the last leg of his tour in Mumbai, when Nadella finally addressed the conglomerates and large businesses (read: potential big ticket customers), he emphasised on leveraging existing data sets within organisations.

His message was concise: data is a company’s most strategic asset.

And Nadella explains how. Every organisation stores data that includes the knowledge of people, their relationships with other employees, their work artifacts – spreadsheets, documents, conversations in email, meetings in teams.

This information, he believes, will decide the quality of the AI models built pertaining to the organisation.

There will be a direct correlation between the performance of AI models and how good the data is being used to feed it. “In some sense, models are becoming a commodity,” he said.

Also, read: AI models are becoming commodity, data one of the most strategic assets, says Nadella

In Delhi, Nadella announced Microsoft would train 20 million (2 crore) Indians in AI by 2030.

Already, 5.6 million (56 lakh) people have received AI skills training since January 2025. This mass skilling effort includes public sector integration, such as an MoU signed with the Labour Ministry to bring AI to e-Shram and the National Career Service.

These platforms serve over a whopping 31 crore informal workers!

Therefore, this is not just training, it is a layer for economic mobility. For example, job matching, resume building, and upskilling will now happen through multilingual AI interfaces.

For Nadella, this was not about futuristic scenarios but about building an AI-native labour economy.

In all three Indian cities, Nadella’s keynote highlighted Microsoft’s full-stack Agentic AI platform, where developers no longer write software line-by-line but orchestrate AI agents to do the work.

Microsoft Foundry now hosts more than 11,000 models, with GitHub evolving into an agent workspace.

Copilot Studio allows prompt-based agent design.

The good old Microsoft Excel, Word, PowerPoint, and Teams now run on agent modes that handle task execution, insight generation, and decision support.

Nadella and his colleagues demonstrated these deployments.

The Chief Executive highlighted how, at Apollo Hospitals, clinicians use AI agents for diagnostics and treatment workflows, and ONGC uses them for upstream oil analysis.

These examples show how Indian companies are reorganising around AI and not just integrating it.

Nadella put it plainly: “The classic SDLC has to change to a new form of AI-driven SDLC.”

Also, read: Gen AI improves software engineering productivity by 70%

However, the AI-driven SDLC wasn’t completely unknown as IT giants have deployed it for some time now. For example, around 30 percent of Cognizant’s code is now written by AI.

What was new in Mumbai was the showcase of MahaCrimeOS, built with Microsoft and CyberEye. The system was piloted in Nagpur, which cut case processing time from months to weeks.

This is what Nadella meant by real-world AI, the technology is no longer abstract but is embedded in the daily operations of healthcare, policing, energy, and employment.

Moreover, the company's enterprise push gained explosive momentum through strategic partnerships with four IT powerhouses: Cognizant, Infosys, TCS, and Wipro.

Each is committed to deploying over 50,000 Microsoft Copilot licenses, totalling more than 200,000 across the board. This massive rollout, announced amid Nadella's $17.5 billion investment pledge, marks a new benchmark for agentic AI adoption.

The larger picture is that the IT sector is evolving into a global AI powerhouse by embedding these licenses, driving Agentic AI into core operations for massive productivity gains and a competitive edge.

According to tech advisory firm Greyhound Research, Nadella’s visit marks a strategic shift, with Microsoft moving from courting India to embedding itself deeply within its institutional and cloud infrastructure.

“Hyperscalers are no longer betting on India’s future,” said Sanchit Vir Gogia, CEO and Chief Analyst of Greyhound Research. “They are fighting over its present.”

Also, read: Cognizant calls time on the IT era, bets future on AI built delivery

Nevertheless, India’s ambitions face structural headwinds.

Skilling 20 million people is a start, but expertise in model development, safety alignment, and governance remains limited.

This is because AI engineers and domain experts are still in short supply, and this is compounded by cybersecurity vulnerabilities.

Also, read: Indian GCCs retrain staff as AI engineers to combat talent shortage