Dies ist eine Übersichtsseite mit Metadaten zu dieser wissenschaftlichen Arbeit. Der vollständige Artikel ist beim Verlag verfügbar.
Capability and Potential of Generative Artificial Intelligence (AI) Tools in Science Education: An Exploratory Study of ChatGPT-Powered Tools (DALL-E and Globe Explore), Midjourney and Gatekeep in Multimodal Science Education
2
Zitationen
3
Autoren
2025
Jahr
Abstract
Generative Artificial Intelligence (AI) tools, including DALL-E (GPT-powered), Midjourney, Gatekeep (GPT-powered) and Globe Explorer, can be used to generate images, videos as well as search, and group resources in science education. This study utilises an evaluation exploration approach to investigate the capability and potential of these tools in generating multimodal teaching and learning materials for basic science education. The findings illustrate that all these tools can be used to create teaching materials and learning resources of basic science concepts and knowledge points with simple and easily learnt prompts. DALL-E and Midjourney can be utilised to generate coloured images corresponding to the prompts while still constrained in adding correct labelling and texting in the images. It is surprising that Gatekeep is decent at producing proper videos to explain scientific knowledge and concepts by giving examples, audio, and animations, with transcripts and summaries on the website. Globe Explorer, as a generative AI searching engine, can be operated to search science concepts and knowledge and obtain systematic and well-structured resources. Future research is suggested to examine the use of more AI tools in materials generation and classroom practice such as teacher’s and student’s perceptions of these tools.
Ähnliche Arbeiten
Explainable Artificial Intelligence (XAI): Concepts, taxonomies, opportunities and challenges toward responsible AI
2019 · 8.292 Zit.
Stop explaining black box machine learning models for high stakes decisions and use interpretable models instead
2019 · 8.143 Zit.
High-performance medicine: the convergence of human and artificial intelligence
2018 · 7.539 Zit.
Proceedings of the 19th International Joint Conference on Artificial Intelligence
2005 · 5.776 Zit.
Peeking Inside the Black-Box: A Survey on Explainable Artificial Intelligence (XAI)
2018 · 5.452 Zit.