India’s AI moment hits a critical phase—Scale is real, value is the test

In February this year, when India hosted the AI Impact Summit -- the first of its kind, the moment signalled a clear shift in intent. The nation not only wants a seat at the global table, but also wants to shape the agenda. A report by SenseAI adds weight to that momentum. Global AI investment touched $800 billion in 2025. Venture capital nearly doubled to $226 billion. And the country is increasingly being seen as a serious hub for AI applications -- where ideas are not just built, but deployed at scale.

For long, the country was seen as a follower in the global AI race. The report pushes back on that view. It argues India is quietly becoming a key driver of the application layer -- where real commercial value gets captured.

The numbers back that claim. Around 75 per cent of Indian AI startups are building at the application layer. Nearly 80 per cent of funding is flowing into these businesses. A major chunk of startups is hitting revenue early.

Add Zee Business as a Preferred Source

Rahul Agarwalla, Managing Partner at SenseAI Ventures, explained this with a simple analogy during a WION panel. AI, he said, is not just about tools like ChatGPT. Think of it like electricity. The real value lies in the appliances built on top of it - the everyday use cases. That is where AI becomes tangible.

The broader message is harder to ignore. Juhi Bhatnagar, advisor to the Indian AI Research Organisation and founder of Forj, framed it in national terms. Over the last decade, countries have played to different strengths - China in manufacturing, Singapore in speed, India in services, the US in innovation. But AI is resetting that equation. Without innovation, there is no edge. And without sustained R&D, there is no innovation. Moving from research papers to patents, and from patents to products, is now the real journey.

The Indian AI Research Organisation, in that sense, is part of a larger push -- building sovereign capabilities, shaping policy, and strengthening the ecosystem.

And the shift is already visible on the ground. AI applications today make up roughly 75 per cent of the ecosystem. Enterprise SaaS is emerging as the largest slice, expanding from 14.1 per cent in 2024 to 25.5 per cent in 2025. Adoption inside companies is picking up pace. So is monetisation. Nearly 60 per cent of startups are already post-revenue, even at early stages.

But the picture is not one-dimensional. AI has clearly moved beyond experimentation. It is now entering execution mode across industries. Yet, the rollout remains uneven.

As Achin Sharma of MOVIN India pointed out during the same discussion, AI is solving problems - but often in parts. Tools exist. Use cases exist. But stitching them into end-to-end workflows is still a work in progress. That is where the next phase of value creation will likely sit.

Enterprise behaviour reflects growing confidence. Average AI contract sizes have jumped sharply - from $39,000 to $530,000 in just two years. That’s not experimentation money anymore. That’s deployment capital.

As deployment rises, so does a more complex question - regulation.

The conversation is no longer just about building faster. It is also about building responsibly. Data, in particular, sits at the centre of this debate.

Agarwalla flagged a core concern. If millions of Indians are using global AI platforms, the data they generate is often leaving the country. That raises a fundamental question: Who benefits from that data? Should it stay within the country's borders? Should domestic companies and policymakers have access to it?

Sharma added another layer. As India moves towards full enforcement of its Digital Personal Data Protection Act, the rules of the game could shift. The old internet adage still applies -- if the service is free, the user is often the product.

India today is among the largest user bases for AI globally, especially for generative tools. That scale comes with both opportunity and responsibility. Massive volumes of data are being created every day. How that data is stored, governed and monetised will shape the next phase of the ecosystem.

Bhatnagar pointed to a structural imbalance. India is generating data. It is building applications. But a large share of revenues and tax gains still accrues overseas, particularly in the US. In simple terms, India is contributing inputs, while value capture often happens elsewhere.

That gap is now coming into sharper focus.

The next phase of the nation's AI journey may not just be about adoption. It may be about ownership - of data, of innovation, and of economic value.

How India navigates this shift will matter. Not just for its own ambitions, but for how the broader Global South approaches the AI era.