In India, the divide between private and public education is stark and deepening. Public education has been in slow decline for decades, sustained largely by the sheer scale of enrolment among poor and low-income families who have few viable alternatives. Despite persistently poor quality and weak accountability, the system continues to serve nearly 50 per cent of all enrolled students, from pre-primary to higher secondary levels.
Creaking infrastructure, outdated curricula, inadequately trained teachers, and an overwhelming reliance on rote memorisation define the lived reality of public education. As if these structural deficiencies were not damaging enough, the rapid rise of artificial intelligence now poses a far more profound challenge—especially when public education is compared with its private counterpart.
Over the last three to four years, AI has begun to play an increasingly central role in education. Its capacity to personalise learning, adapt content to individual needs, and accelerate cognitive development threatens to exponentially widen the learning gap between students in private and public schools. Left unaddressed, AI risks perpetuating inequality at a scale and speed never before witnessed in India’s education system.
This emerging asymmetry can be understood through four illustrative examples.
First, access to something as basic as a personal learning device now determines whether a student is included in or excluded from the AI-driven learning cycle. AI excels at adaptive learning—tailoring content, pace, and style to the individual. Without personal devices, AI-based education becomes standardised and blunt, stunting outcomes rather than enhancing them. For students from low-income families, the lack of reliable internet and limited access to devices constitute the single biggest barrier to meaningful AI-enabled learning. This stands in sharp contrast to wealthier students, who benefit from personalised learning supported by digital infrastructure both at school and at home. Students in public schools—often deprived of access in both spaces—are doubly disadvantaged.
Second, language plays a decisive role in producing uneven outcomes. Most AI models are trained primarily on English-language data, incorporating Western pedagogical norms and curricula. Students from rural and economically disadvantaged backgrounds—often with limited exposure to English—are unable to harness AI effectively. Their education takes place largely in regional languages, shaped by local cultural contexts that are poorly represented in current AI systems. As a result, they encounter poorly contextualised or even inaccurate content, pushing them further behind. Private education, where English is the primary medium of instruction, prepares students far better to use AI as a learning aid. Over time, these language-mediated advantages reinforce educational hierarchies, locking disadvantaged students out of quality higher education as they progress through the system.
Third, the role of teachers in the AI era is diverging sharply across the public-private divide. Well-funded private schools are far more likely to create environments where teachers use AI to design better lessons, refine pedagogy, and upgrade their skills—enhancing the learning experience. In contrast, public systems are more likely to deploy AI as a cost-cutting substitute rather than a tool to empower teachers. In systems already facing acute teacher shortages, AI risks becoming a justification for reduced human interaction rather than improved instruction. As a result, teacher-quality differentiation becomes even more pronounced, widening learning disparities.
Finally, AI is reshaping assessment itself. AI-enabled tools challenge rote memorisation by asking better questions, enabling personalised evaluations, and emphasising reasoning and problem-solving. By contrast, AI-absent learning continues to reward standardised answers, memory, and exam performance under time pressure. It is therefore unsurprising that AI will play a decisive role in shaping students’ socio-economic trajectories after they exit the school system—determining who accesses better universities, who is perceived as more capable, and who is prepared for an increasingly complex world.
The deployment of AI in public education is inevitable. But deployment alone is not enough. If equitable outcomes are to be achieved, the capacity of public schooling must be strengthened, and teachers must be empowered—not displaced—by AI. Without deliberate intervention, students from disadvantaged groups—whether defined by poverty, caste, region, or disability—will have little real chance at educational equity.
For a country like India, this is not a peripheral policy concern; it is a defining test of its development trajectory. When nearly half of the nation’s students—those in public schools—risk being structurally excluded from AI-enabled learning, the consequences extend far beyond classrooms. Over the coming decades, this asymmetry could erode India’s demographic dividend, lock millions into low-productivity livelihoods, and weaken the foundations of its ambition to become a developed nation. The question is no longer whether AI will transform Indian education, but whether it will do so for all, or only for the privileged few.
---
Vikas Gupta is an AI inclusion and governance advisor who works with organisations on responsible AI adoption. He writes on the governance risks of algorithmic systems and their implications for workforce equity
Comments