By Sandeep Dutta
When AI moves from the lab to the lives of a billion people, the stakes change. In India, AI is solving problems never solved before, at a scale never attempted. The opportunity is immense, but so is the responsibility to ensure that this technology is trusted, inclusive and built to deliver lasting value.
India’s approach to AI is shaped by its unique combination of scale, diversity and digital infrastructure. Announced at the India AI Impact Summit in New Delhi, the MANAV framework laid out five pillars for AI governance: moral & ethical systems, accountable governance, national sovereignty, accessible & inclusive design, and verifiable & safe AI. These are not abstract ideals. They map directly to the choices India’s enterprises and policymakers are making today.
India’s digital public infrastructure (DPI) offers a useful starting point. Over the past decade, it has created a nationwide trust layer, making verified identity, real-time payments and digitised credentials accessible at population scale. AI is now building the intelligence layer on top of this foundation, turning access into insight, automation and more personalised experiences.
Healthcare offers one example of what this can mean. PMJAY, the world’s largest health insurance scheme, serves over 550 million beneficiaries. Today, AI validates 1.2 million monthly claims, many built on unstructured medical records that once took 48 hours each to process. Processing time has been brought down to under 24 hours. Similarly, Digi Yatra, through facial recognition, has transformed the travel experience across more than 100 million journeys since the app was launched.
These examples illustrate what happens when AI meets India’s digital public infrastructure: trust, speed and governance at population scale. It is a powerful combination, and one shaped by the country’s particular needs.
Across industries, organisations are moving from reactive decision-making to predictive intelligence. A bank is underwriting loans in seconds. A manufacturer is reducing equipment diagnosis from seven hours to under 10 minutes, shifting from reactive troubleshooting to real-time predictive intelligence. A retailer is hyper-personalising experiences in real time. These are solutions born from a uniquely Indian context, where scale and diversity demand approaches that can work across a wide range of users and operating environments.
The impact is clear. But sustaining it requires more than ambition. It requires the right architecture of choice, governance and talent. Organisations come to AI from different starting points, and the ability to choose between models and architectures is central to building systems that fit their needs. An open AI ecosystem gives businesses the freedom to select the right tool for each task, rather than being locked into a single approach. It also creates room for innovation, allowing enterprises to adapt as technologies evolve and their requirements change.
This includes sovereign models built for India-specific needs, making initiatives such as the IndiaAI Mission vital. Sovereign cloud adds another dimension, ensuring that sensitive workloads across government, financial services and regulated industries can run with data residency and inferencing within India. Together, these capabilities can help build an AI ecosystem that is trusted, inclusive and responsive to the country’s requirements.
Yet technology scales only when governance is embedded into architecture. Security gives organisations the confidence to adopt new technologies at scale. Cloud computing demonstrated the importance of building security into infrastructure from the start. AI must follow the same principle.
Consider the scale of digital payments in India. In August 2026 alone, UPI recorded 24.51 billion transactions valued at `29.82 trillion. As AI becomes part of similarly critical systems, governance and security must be embedded at every layer — infrastructure, models, applications and data.
In practice, this means building trust by design. Organisations need visibility into where AI is operating, permissions that define its authority, continuous monitoring of its behaviour and clear accountability for its outcomes. The objective is not to slow innovation, but to make it dependable enough for widespread use.
None of this will matter without investment in the people who will build, deploy and govern AI. That investment must begin at the education level. AI literacy needs to be woven into school and university curricula, rather than treated as a niche specialisation reserved for engineering graduates.
Organisations, too, must approach AI skilling as a continuous capability-building effort, not a one-time exercise. It must span every level of the workforce, helping employees understand how to use these tools responsibly and creatively. The opportunity is not simply to adopt AI, but to build a generation of AI-ready talent that can apply it at scale.
This also means ensuring that the benefits extend beyond India’s major metros to tier-2, tier-3 cities and beyond. A country as diverse as India cannot afford to build an AI economy that serves only a narrow section of its population.
India is no longer merely adopting AI. It is shaping how the technology can be built and deployed for a billion people, and potentially for the world. The ingredients are here: talent, infrastructure, market diversity and sovereign compute.
The next phase will depend on how these strengths come together. If India can combine its digital foundations with open choices, strong governance, secure architecture, and sustained investment in people, it has the potential to demonstrate not just AI at scale, but AI done right.
The writer is President, AWS India & South Asia.
Disclaimer: The views expressed are the author’s own and do not reflect the official policy or position of Financial Express.