India’s AI curriculum broadens access, but risks widening the employability gap. India’s Economic Survey 2024-25 flagged the tension directly. Chief Economic Advisor V. Anantha Nageswaran noted that while technology eventually creates more jobs than it displaces, the critical period lies in between. That interval requires supporting institutions, changed academic curricula, and changed workplace practices. India’s AI education policy, rolled out through CBSE from 2026-27, is one such institutional response. Its design choices carry specific implications for who benefits from the AI economy and how soon. The National Education Policy 2020 identifies AI, data science, and related fields as essential for future employment and economic growth. The Ministry of Education frames the CBSE curriculum for Classes 3 to 12 as building a pipeline of AI literate students capable of contributing to technology-driven sectors. The curriculum emphasises computational thinking as a precursor to AI capability, not as early technical specialisation. Students begin with logical reasoning and pattern recognition. Depth is deferred. This is a deliberate choice. And it signals a specific theory of labour. India’s AI curriculum is explicitly cross-sectoral. Programmes like YUVAi, under which students apply AI to agriculture, health, transport, and rural development, link school education to real-world problem solving across domains. The objective is not to produce AI engineers alone. It is to produce workers who apply AI thinking across a mixed economy with a large informal sector. China’s model runs on a different logic. Its Ministry of Education released guidelines in May 2025 establishing a tiered system where primary students learn AI through voice recognition and image classification, junior high students examine machine learning processes, and senior secondary students design and refine algorithmic models. The China Academy of Information and Communications Technology reported that China’s AI sector crossed 900 billion yuan in 2024, a 24 percent year-on-year increase. Education is organised to feed that industrial machine. India delays specialisation by design. China accelerates it. Changjiang Securities forecasts China’s AI education market will reach 160 billion yuan by 2027, aligned with a coordinated industrial policy framework where labour demand is anticipated and training pipelines are built to match. Embedded Inequality DIKSHA (Digital Infrastructure for Knowledge Sharing) uses AI to improve accessibility, including for visually impaired learners. But ASER (Annual Status of Education Report) 2024, which surveyed 649,491 children across 605 rural districts, reveals what inclusion looks like at the ground level. Only 43.5 percent of government schools have computers for teaching, against 70.9 percent in private unaided schools. Among 14 to 16-year-olds, 89 percent report smartphones at home but only 57 percent use them for education. Over 44 percent of Class 5 students in government schools still cannot read a Class 2-level text. Gender compounds the gap. ASER 2024 found that 36.2 percent of boys aged 14 to 16 personally own a smartphone, against 26.9 percent of girls, a 9.3 percentage point gap consistent across all states. Girls from conservative rural households face reduced digital access at home, which means AI curriculum delivered in school cannot be reinforced at home. The policy’s inclusion intent and its likely outcomes are not the same thing. China’s centralised system reduces implementation disparity by design. Standardisation ensures more uniform exposure across schools, and the Ministry’s coordinated mechanism synchronises curriculum, teacher training, and infrastructure nationally. But China’s model subordinates individual flexibility to national industrial priorities. Skills built for current strategic sectors may not adapt when technological trajectories shift. India’s curriculum incorporates ethical reasoning alongside technical exposure. Students are expected to understand how AI affects society, not only how to operate AI systems. This shapes a different kind of worker over time, one who questions systems rather than only operating within them. China’s model places greater emphasis on application and optimization within defined systems. The May 2025 guidelines prohibit students from submitting AI-generated content as academic work and mandate critical thinking about AI outputs. But the curriculum’s stated goal is technological self-reliance and industrial competitiveness. Ethical training is supplementary to that objective, not co-equal with it. India’s approach builds contextually adaptable workers suited to a diverse economy with a large service sector, an informal workforce, and developmental challenges that require cross-domain problem solving. China builds functionally specialised workers aligned with a coordinated industrial state. The question for India is whether its broad-base strategy produces workers with enough technical depth to compete for high-value segments of the AI economy, and whether the infrastructure to support that depth exists outside the country’s top-tier urban schools. Execution Gap Teacher training is the most acute pressure point. India has approximately 96 lakh school teachers as of UDISE+ (Unified District Information System for Education Plus) 2024-25. NISHTHA’s (National Initiative for School Head and Teachers Holistic Advancement) grade-specific modules are the planned delivery mechanism. But AI pedagogy is not yet integrated into B.Ed. curricula. Teachers trained through short-term workshops face students who will not see the benefit if that training does not translate into sustained classroom practice. The infrastructure deficit compounds this. A school without functional electricity cannot run AI-enabled tools. A student without home internet access cannot practise outside school hours. The 18.5 percentage point internet access gap between government and private schools is not a technology problem. It is a resource allocation problem that requires sustained public investment, not curriculum design. If India addresses these gaps, its broad-base model holds genuine promise. A large, demographically young workforce applying AI across agriculture, health, logistics, and finance could produce development outcomes that specialised industrial pipelines do not reach. If those gaps remain, AI education will expand awareness without shifting labour outcomes. Students will learn what AI is. They will not learn to use it at a level that changes what work they access. The curriculum is set. The system that must deliver it is under-resourced in precisely the schools that serve the students who need it most. (The writer is an independent public policy researcher who writes on political economy, climate, and the ethics of everyday systems. Views personal.)
Comments