Other

Breaking disciplinary silos in education: A case study on leveraging large language models for a scalable, fully digital pedagogy

Pusuluri V, (Pusuluri) SB, Pusuluri K · 4 Sept 2026

Abstract

Abstract This article presents a comprehensive case study documenting the educational journey through Grade 1 and early Grade 2 in an Artificial Intelligence (AI) integrated, fully digital environment, bridging the gap between theoretical AI application and sophisticated pedagogy. We demonstrate through several use cases that leveraging Large Language Models (LLMs) as an "intellectual exoskeleton" makes advanced, interdisciplinary projects feasible within an early childhood curriculum, with AI acting as the crucial "connective tissue" for a highly personalized and adaptive curatorial process. Our approach promotes active student involvement moving beyond passive consumption to creation, and a “Verifier” model that identifies AI as a fallible digital assistant, shifting success in an AI-integrated environment from information generation to verification and demanding objective analysis to identify AI generated inaccuracies in text, logic, or code. We feature several visually captivating student-driven projects including programmed animations to study music and wave physics, the design of a custom 3D virtual globe for geography, visualizing complex computer science algorithms, and the authorship of custom multilingual copy-tracing books for fine motor skills. We leverage specialized, ergonomically safe, and distraction-free technology to create a permanent cloud-based digital record of lectures, student progress, and course material. Our approach successfully dissolves the traditional silos between languages, arts, and sciences, suggesting that such an integrated environment allows for the immediate practical application of theoretical knowledge. This could equip young learners with the technological command, critical thinking, and verification skills required to navigate a future defined by exponential change. While situated within an early childhood context, this study identifies foundational interaction patterns that are scalable to higher education, professional development, and lifelong learning. The relative simplicity of early childhood projects provides a transparent, cross-disciplinary lens through which the mechanics of AI-augmented synthesis and cognitive scaffolding can be rigorously examined and adapted for more complex academic tiers.