Leveraging Artificial Intelligence for Teacher Training: Enhancing Professional Development in India’s Foundational Education System After COVID-19

Abstract

This paper evaluates the potential for using AI to enhance teacher professional development (PD) in India's foundational grades (pre-primary to Grade 3) in the wake of COVID-19, revealing gaps in pedagogy and learning, as well as the constraints of cascade-style PD, which lack scalability, personalization, and feedback. We offer an AI & muscular PD package (adaptive micro-learning modules, lesson planning copilots, coaching chatbots, and analytics dashboards) via DIET/BRC/CRC networks. The multi-site mixed-methods design integrates quasi-experimental comparisons with design-based research cycles. The data comprise TPACK, self-efficacy, and AI attitude questionnaires, as well as classroom observation data, platform usage logs, and formative student assessments. Results indicate growth in both teachers' TPACK and observed classroom practice, with the most significant gains when coaching and analytics were combined. Adoption is driven by perceived usefulness and ease, as well as school digital readiness; early indications suggest gains in foundational literacy and numeracy without overburdening the teacher. Initial costing suggests that per-teacher expenditure is comparable to that of legacy cascade training, while allowing for more frequent feedback loops. We discuss implications for scalable, equity-aware PD with strong guardrails for privacy and bias testing and point to priorities for longer follow-up and deeper classroom analytics.