India's AI Education Push Risks Automating Social Hierarchies
New national AI curricula for primary students arrive amid warnings that algorithmic bias and unequal access could codify a technological caste system.
India is integrating artificial intelligence into its primary education system, but experts warn the move could inadvertently codify a new social hierarchy. The Central Board of Secondary Education (CBSE) will introduce a Computational Thinking and Artificial Intelligence curriculum for students in Classes 3–8 starting in the 2026–27 academic year.
While the initiative aims to democratize technical knowledge, the rollout occurs against a backdrop of systemic inequality. There is a growing risk that AI will create a tiered society: those who design and control intelligent systems, those who merely use them, and those who are simply classified, monitored, or replaced by them.
The Algorithmic Divide
The risk is not merely one of access, but of embedded bias. A study (arXiv:2601.14506v1) revealed that AI systems produced simpler educational explanations for profiles associated with regional-language schooling or marginalized castes. The research suggests that the AI treated these social markers as indicators of lower cognitive ability, effectively automating a form of intellectual discrimination.
This digital disparity is further reflected in the workforce. A research paper titled "The Privilege of Exposure: Caste and Generative AI in India’s Graduate Labour Market" (arXiv:2606.13314) utilized PLFS 2025 data to analyze the intersection of generative AI and caste. The findings indicate that caste remains a primary determinant of occupational attainment, with graduates from marginalized backgrounds less likely to occupy roles that benefit from generative AI compared to upper-caste graduates in the same districts.
Structural Implications
India's history of caste-based discrimination has long been tied to the control of knowledge and occupation. The current push for AI integration threatens to mirror these patterns. A significant gap persists between affluent English-medium schools and under-resourced rural institutions regarding high-speed internet access, personal hardware, and quality instruction. Without intervention, this divide risks evolving into a "technological caste" system.
If AI literacy remains concentrated among the already privileged, the technology may do more than fail to bridge social gaps; it could invisibly encode prejudices into critical infrastructure. This includes the automation of bias in job recruitment, credit scoring, and educational support, further marginalizing Dalit and tribal communities.
The Path Forward
As the 2026–27 academic year approaches, the focus shifts to whether the CBSE curriculum can be implemented equitably. The primary concern remains whether the state can provide the necessary infrastructure to rural schools to prevent AI from becoming a tool for further exclusion. Observers will be watching to see if the government introduces safeguards to prevent algorithmic bias from mirroring the country's historical social stratifications.