ImageCritique – Development and Experience with an AI-Based Tool for Image Critique in Teaching
Abstract
A multimodal AI chatbot named ImageCritique was developed to provide multimedia students with substantive feedback on image-based work. Built on ChatGPT, the chatbot uses task-specific criteria in the system prompt to deliver targeted feedback. The tool was tested in two courses at the University of Agder: DAT104 Multimedia Design and DAT227 Photography. Its effects were examined through a controlled pre-test in UiA's Usability Lab (n = 5) and an online survey administered after ordinary classroom use (n = 20). Results show that students consistently found ImageCritique useful and supportive of learning. A secondary effect was that teachers saved time otherwise spent writing individual feedback. The findings indicate that precisely formulated instructions are a key condition for feedback quality, and that AI-based chatbots hold genuine pedagogical potential in technical-creative disciplines.
Downloads
Downloads
Published
Issue
Section
License
Copyright (c) 2026 Haakon Sundbø, Kåre Mosgren, Tony Bergholtz, Per Arne Kvamsø
This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.
All innhold i Læring om læring er publisert med cc lisens
Navngivelse-IkkeKommersiell-Ingen bearbeidelser 4.0 Internasjonal (CC BY-NC-ND 4.0)
Du har lov til å Dele — kopiere, distribuere og spre verket i hvilket som helst medium eller format
På følgende vilkår:
Navngivelse — Du må oppgi korrekt kreditering
IkkeKommersiell — Du kan ikke benytte materialet til kommersielle formål.
IngenBearbeidelser — Dersom du remixer, bearbeider eller bygger på materialet, kan du ikke distribuere det endrede materialet.
