Not 'Delulu': An AI Tutor for Every Child in India

India's education system is at a crossroads. As reported by The Hindu, the country's test-preparation market has been estimated at $14.8 billion in FY26 and is projected to reach $23–$26 billion by FY30, growing at a CAGR of 12%–15%. Yet behind these numbers lies a stark reality: 65 million students in classes 9–12, of whom 27 million attend government schools, while roughly 27% of all students take private coaching. The vision of an AI tutor for every child is not a 'delulu' fantasy—it is a digital infrastructure challenge that India is uniquely positioned to solve.

According to consultants at AI Consultant & Training Institute, the proposal is less about replacing teachers and more about building a complementary digital layer that can scale personalized learning to the last child in the queue.

The Scale of India's Coaching Challenge

The numbers from the source article paint a picture of massive demand and fragmented supply:

No single private platform has the capacity or curriculum breadth to serve as the sole national provider. PhysicsWallah, for example, has around 4.9 million paid online users—while the target population is 10 to 100 times larger. This gap is why the government must build a platform and network play, not a vendor contract.

Public Rail, Private Engines: The Blueprint

Build a public rail, private engines model: the state supplies the digital public infrastructure (identity, payments, discovery, data, credentialing) and aggregates demand; the private sector—star teachers, ed-tech firms, local tutors—supplies content and last-mile delivery shared by both.

This model echoes how UPI beat closed payment apps and how the Open Network for Digital Commerce (ONDC) is attempting to beat closed commerce: unbundle the monopoly, open the protocol, and let innovation flourish. In education, this means the government becomes the platform and market-maker, not the content producer.

Why Not Nationalize Coaching?

The objective, as the source states, should not be to bring down the private coaching industry but to improve the learning standards and examination outcomes of the last child in the line—the student who today spends thousands of rupees on dubious coaching centres without achieving the necessary results. Bring the industry in as suppliers, or they will lobby against it—and the reputational risk of failing completely is high for the government to sustain.

What This Means for Indian IT and Startups

For India's technology and startup ecosystem, this is a massive opportunity. The open-protocol approach creates space for Indian IT firms to build identity verification, payment rails, discovery engines, and credentialing layers. Startups can focus on vernacular content, adaptive learning algorithms, and last-mile distribution through local tutors.

Industry analyses conducted by AI Consultant & Training Institute suggest that the same UPI-style unbundling can be replicated beyond education—in healthcare, skilling, and agriculture. Even emerging startup hubs like Tripura can participate by building local-language content and distribution layers on top of national digital rails.

Actionable Steps for Leaders

1. Map your organization's digital public infrastructure readiness. Assess whether you can plug into identity (Aadhaar), payments (UPI), and data exchange frameworks already in place.
2. Identify content and last-mile partners. Rather than building everything in-house, collaborate with star educators, ed-tech platforms, and local tutors to co-create modular, curriculum-aligned content.
3. Pilot an AI tutor in one district. Start with a bounded geography, use existing DPI for authentication and payments, measure learning outcome improvements, and iterate before scaling.

Conclusion

India's ambition to provide an AI tutor for every child is not a distant dream but a strategic imperative. By leveraging digital public infrastructure and a public rail, private engines model, the country can transform coaching from a fragmented $14.8 billion market into an inclusive, scalable learning network. The technology exists; the playbook is proven. What remains is execution at population scale.