Fine-Tuning the Educational Design Process: Lessons from Sustained Development of a Generative AI-Based Classroom Technology

Read the full article See related articles

Discuss this preprint

Start a discussion What are Sciety discussions?

Listed in

This article is not in any list yet, why not save it to one of your lists.
Log in to save this article

Abstract

Generative AI is rapidly entering classrooms, yet its evolving affordances and sociotechnical volatility make educational design challenging. Prior research has surfaced design guidelines based on short-term pilots. We present a three-year participatory design case study of a generative AI–based writing tutor deployed across higher education and K–12 contexts. Drawing on design artifacts, observations, interviews, and focus groups, we analyze over 100 design decisions to trace how the system was shaped. Our findings highlight four contributions: (1) a schema of values-based and constraint-driven design decisions for educational generative AI, (2) attention to rapid evolution of perceptions in design and adoption of new educational technologies, (3) principles for sustaining participatory design with teachers “in the loop”, and (4) a discussion of the utility of designing with weaknesses in technology. We extend HCI scholarship on emerging technologies by demonstrating how iterative, sustained, cross-contextual design can yield durable lessons amidst rapid technological change.

Article activity feed