DIGITAL LIBRARY
AI STORY CUBES: A STUDY OF A GENERATIVE AI ICEBREAKER FOR EARLY GROUP INTERACTION AND TOOL ONBOARDING
Dublin City University (IRELAND)
About this paper:
Appears in: EDULEARN26 Proceedings
Publication year: 2026
Article: 1850
ISBN: 978-84-09-88444-5
ISSN: 2340-1117
doi: 10.21125/edulearn.2026.1850
Conference name: 18th International Conference on Education and New Learning Technologies
Dates: 29 June-1 July, 2026
Location: Palma, Spain
Abstract:
Hybrid and online-first programmes increasingly bring students together for short on-campus teaching blocks or workshops. In these settings, educators often need an activity that helps students begin working with unfamiliar peers while also introducing the collaboration tools and expectations that will be used later in the module or programme. This study examines an icebreaker intervention called AI Story Cubes, designed for postgraduate management teaching. The aim is to support early group interaction and structured discussion, while introducing a simple, low-stakes way of using generative AI as part of classroom activity design.

In the activity, students are randomly assigned to teams and provided with three short “story cubes” generated by a large language model. Each cube is aligned to a different course lens, for example across modules taught in parallel or across lenses within a single module. Teams produce a brief scenario narrative organised around a decision point, then add a short mapping that indicates where each lens was applied. Groups capture their work in a shared collaborative slide deck that can be viewed during class. This shared artefact supports a short gallery walk and a whole-class debrief focused on participation norms, teamwork routines, and agreed boundaries for using shared documents and generative AI in later learning activities.

The study uses a design-and-evaluate approach focused on practical implementation questions. First, can the activity be run within a standard class session using common institutional tools? Second, how do students describe its usefulness for getting started with peers, course ideas, and required technologies? Third, what adjustments appear necessary to support participation across group members? Data sources include short pre- and post-activity student feedback instruments and instructor facilitation notes. The study does not claim validated learning gains. Instead, it offers a replicable icebreaker design and practical guidance for instructors working in technology-mediated, time-constrained teaching settings.
Keywords:
Generative AI, icebreakers, group work, collaborative tools, curriculum integration, hybrid learning, student participation.