The article explores the potential of generative artificial intelligence for creating illustrative content in next-generation school textbooks. Based on a comparative analysis of three models (DALL·E 3, Adobe Firefly, ChatGPT), their effectiveness in performing tasks of varying complexity – from simple object images to complex scientific diagrams – was evaluated. The results demonstrate the high suitability of AI for generating entry-level visual content, but also highlight limitations in accuracy and stylistic consistency when producing educational illustrations for middle and high school. The study concludes that human expert involvement is necessary during the stages of image verification and editing, and outlines promising directions for further research.
Paper
The full text of this publication is not hosted on 44B due to licensing.
Read it at OpenAlex