Teacher-AI Collaboration for Curating and Customizing Lesson Plans in Low-Resource Schools
Deepak Varuvel Dennison, Bakhtawar Ahtisham, Kavyansh Chourasia, Nirmit Arora, Rahul Singh, Rene F. Kizilcec, Akshay Nambi, Tanuja Ganu, Aditya Vashistha
TL;DR
The paper investigates how an AI-assisted, bilingual lesson-planning tool called Shiksha Copilot supports teachers in low-resource schools in Karnataka, India, focusing on collaboration between teachers, curators, and AI within a human-in-the-loop workflow. It uses a two-phase deployment (curation and distribution) and a large mixed-methods study (1043 teachers, 23 curators) to examine content quality, adaptation practices, and changes in teaching practices. Findings show reduced administrative workload and time for planning, a shift toward activity-based pedagogy, and high English content reliability but substantial linguistic edits for Kannada translations, with broader pedagogical change constrained by staffing and administrative demands. The authors argue for embedding AI tools within Communities of Practice to sustain contextually relevant content and outline design directions to advance teacher-centered EdTech in multilingual Global South settings.
Abstract
This study investigates Shiksha copilot, an AI-assisted lesson planning tool deployed in government schools across Karnataka, India. The system combined LLMs and human expertise through a structured process in which English and Kannada lesson plans were co-created by curators and AI; teachers then further customized these curated plans for their classrooms using their own expertise alongside AI support. Drawing on a large-scale mixed-methods study involving 1,043 teachers and 23 curators, we examine how educators collaborate with AI to generate context-sensitive lesson plans, assess the quality of AI-generated content, and analyze shifts in teaching practices within multilingual, low-resource environments. Our findings show that teachers used Shiksha copilot both to meet administrative documentation needs and to support their teaching. The tool eased bureaucratic workload, reduced lesson planning time, and lowered teaching-related stress, while promoting a shift toward activity-based pedagogy. However, systemic challenges such as staffing shortages and administrative demands constrained broader pedagogical change. We frame these findings through the lenses of teacher-AI collaboration and communities of practice to examine the effective integration of AI tools in teaching. Finally, we propose design directions for future teacher-centered EdTech, particularly in multilingual and Global South contexts.