Research Summary: What a teacher cannot delegate: a pedagogy-first framework for integrating conversational AI agents across the simulation cycle

A framework published in Frontiers in Education outlines the integration of conversational AI agents across the simulation-based learning cycle, which comprises preparation, enactment, and debriefing. The core principle establishes that educator evaluation of pedagogical requirements—rather than raw technological capability—must govern how, where, and if conversational AI is deployed, reinforcing that AI serves to extend rather than replace instructional staff.

Why it matters

This framework provides a model for shifting AI adoption from technology-driven deployment to pedagogy-led integration. It ensures that investments in conversational technologies directly reinforce learning outcomes without undermining educator authority. By treating AI as an augmenting tool across specific instructional phases, institutions can scale complex learning simulations while maintaining educational quality.

Key insights

• Simulation-based learning effectiveness hinges on fulfilling the distinct pedagogical demands of preparation, enactment, and debriefing.
• Technology deployment should be determined by the educator's pedagogical assessment rather than available AI capabilities.
• Conversational AI agents should be tailored with specific epistemic stances and course knowledge corresponding to each phase of the simulation cycle.
• AI tools are positioned to extend educator capacity rather than substitute for fundamental teaching roles.

Source

Frontiers in Education — https://www.frontiersin.org/articles/10.3389/feduc.2026.1961346