Prompt design and the reconfigured teacher: Enabling insight and creativity in GenAI-mediated classrooms

Abstract

This article examines the transformative impact of Artificial Intelligence (AI) on the teacher's role in contemporary learning environments, focusing on a shift from knowledge transmission to the orchestration of cognitive ecosystems through pedagogical prompt design. The study adopts an exploratory mixed-methods design conducted in three ninth-grade classrooms at a public secondary school in Colombia, where an educational escape room generated using generative AI and a structured pedagogical prompt was implemented to stimulate emergent learning processes. Quantitative analysis focused on the frequency of “insight moments,” defined as audibly articulated ideas that restructured the collective cognitive field, and the complexity of students’ final creative responses. The intervention identified 32 insight moments and revealed two divergent pathways of cognitive emergence: one characterized by frequent individual insights and another by slower collaborative synthesis leading to high-complexity outcomes. These findings indicate that high frequencies of insight do not consistently predict deeper conceptual integration at the group level. Qualitative observations suggest that the teacher’s role evolved into a “Higgs-like” configurator, an invisible architect who shapes the learning field through prompt design rather than direct instruction. The study concludes that pedagogical prompting constitutes a sophisticated form of instructional planning that enables spontaneous idea generation and transdisciplinary cognition in AI-mediated classrooms.