By Rohit Vyas, Director of Solution Engineering, Confluent India
In early 2026, the AI conversation in India has matured. In fact, it has expanded multi-fold recently with our global presence, the AI Summit, and big announcements. It gives us a signal that we have moved beyond the spectacle of chatbots and generative demos to something far more consequential: infrastructure. The national debate is no longer about whether AI is transformative, but about whether India can build the data backbone required to lead in it.
We are not merely a large digital population but are also becoming the world’s most significant data factory, with nearly 960 million active Indian internet users. India contributes close to one-fifth of global data creation, and the average smartphone user consumes roughly 36 GB of data every month.
Every UPI transaction. Every OTT stream. Every e-commerce checkout. Every telemedicine session. Every GPS ping. Every farm-tech sensor reading. Data is not arriving in files anymore. It is arriving in flows per second, per millisecond.
With 5G subscriptions accelerating and expected to cross a billion by the end of this decade, the velocity of data creation is compounding. We are not just producing more data. We are producing data faster.
But raw data does not equal intelligence. As we enter what I call India’s AI Decade, one truth is becoming unavoidable in boardrooms and policy corridors alike: AI supremacy will not be decided by who owns the most GPUs. It will be decided by who builds the most intelligent data infrastructure.
The data centre moment and its strategic meaning
This year’s Union Budget reinforced what many of us have sensed for some time: data infrastructure is now national infrastructure.
The tax incentives extended to global cloud providers operating India-based data centres until 2047 are not just fiscal measures. They act as strategic signals highlighting India’s outlook to anchor compute locally, showing that sovereignty matters and that digital capacity is economic capacity.
Our data centre capacity is projected to triple by 2030, crossing 8 GW. Investments are pouring in from hyperscalers, private capital, and state governments. The IndiaAI Mission continues to expand compute access, and the DPDP Act has formalised accountability and governance frameworks. But according to me, here is the uncomfortable truth: Data centres are vaults. They are necessary. But vaults alone do not create value.
If the data inside remains fragmented or locked in silos, then capacity becomes underutilised capital. Infrastructure without intelligence is nothing but real estate. The next phase of India’s AI story will not be written by those who build the tallest server stacks; it will be written by the ones building the smartest data pipelines.
Our data reality: Volume, velocity, and variety
Our network exchanges routinely handle traffic between 740 and 980 gigabits per second. Nearly 83% of our mobile subscriptions are 4G, with 5G adoption rising rapidly. Data is no longer episodic. It is continuous.
Yet in many enterprises, data architecture still reflects a different era, one built around nightly batch jobs, warehousing, and retrospective reporting. That architecture was designed for predictability. But we now operate in a state of unpredictability.
Fraud detection in fintech cannot wait for morning reconciliation. Supply chain optimisation during festive surges cannot rely on yesterday’s dashboards. AI copilots serving millions of users cannot depend on outdated context.
The Caveat: Investment without stack modernisation
The current wave of investment in data centres is necessary, but it is not sufficient. If we channel our capital into storage without parallel investment in modern technology stacks, we risk building high-capacity infrastructure running on legacy architectures.
The shift we must consciously make is from data-at-rest to data-in-motion.
Real-time data pipelines, event-driven architectures that stream, process, filter, and govern data as it flows, must become foundational. Not experimental. Not peripheral.
Because AI models do not operate on static datasets anymore. They operate on streams of behavioural, transactional, and operational signals. An AI system trained on yesterday’s data is misaligned by design.
Why real-time architecture matters for sovereign AI
When we talk about Sovereign AI, we often focus on local LLMs, GPU clusters, and compute capacity. But sovereignty has an architectural dimension that is less discussed.
Sovereignty means control over how data flows. It means portability across cloud environments. It means governance embedded in pipelines. It means visibility at every hop.
Therefore, efficient stream processing reduces unnecessary storage and compute cycles. That has energy implications. India’s AI ambition cannot come at the cost of unsustainable energy consumption. Intelligent filtering, processing only what matters in motion, is more than just a technical upgrade.
What data centres must mean for India
So what should data centre investment mean for us? It should not mean passive storage expansion. It should not mean compute capacity without orchestration.
It should mean building intelligent, real-time digital nervous systems across sectors.
Banking. Telecom. Healthcare. Retail. Public digital infrastructure. Each of these sectors generates enormous streams of data. The opportunity is not in accumulating it, but in activating it.
The real economic dividend will come when enterprises can connect every event, from a factory floor to a mobile screen, into a continuous decision loop.
That is how we move from reactive to predictive, from predictive to autonomous, and from autonomous to adaptive.
Therefore, we must architect for velocity, not just volume. We have the demographic advantage. We already have scale. We already have policy momentum and investment appetite. Now we must match that with architectural discipline.
Because in this decade, the winners will not be those who own the most data. They will be those who can move the right data, at the right moment, with trust and intelligence built in. India’s future is not stored in racks. It is flowing every second.