Kate Sangwon LEE* and KHOO Eng Tat
Engineering Design & Innovation Centre
*katelee@nus.edu.sg
Lee, K. S., & Khoo, E. T. (2023). Incorporating generative AI in project-based learning: Case study of how students utilise generative AI in tech-enabled projects [Paper presentation]. In Higher Education Campus Conference (HECC) 2023, 7 December, National University of Singapore. https://blog.nus.edu.sg/hecc2023proceedings/incorporating-generative-ai-in-project-based-learning-case-study-of-how-students-utilise-generative-ai-in-tech-enabled-projects/
SUB-THEME
AI and Education
KEYWORDS
Generative AI, technology-enabled project, project-based learning, interdisciplinary learning
CATEGORY
Paper Presentation
ABSTRACT
As generative artificial intelligence (AI), such as Chat GPT and Midjourney, continues to permeate various industries, we have witnessed a recent surge in its adoption within project-based learning in education. (Gimpel et al., 2023; Su & Yang, 2023). However, as this technology is rapidly evolving and new services are introduced by various software platforms, understanding the appropriate software services and how they could be utilised in the students’ projects are challenging. This paper presents three case studies (under the module EG3301R “Ideas to Proof- of-Concept,” offered by the Innovation & Design Programme) that highlight how students identified design opportunities where utilisation of generative AI technology could enhance and improve the effectiveness of the learning process.
CONTEXT
Generative AI usually refers to AI systems that generate new content, including images, texts, music, and synthetic data (Cooper, 2023; Gimpel et al., 2023). One of the most representative services of generative AI is ChatGPT, a conversational service that uses large language models to interact with users (Gimpel et al., 2023).
Project-based learning is student-centred, context-specific, and inquiry-based learning where students can be actively involved in the learning process by interacting with other students and teachers within real-world practices (Kokotsaki et al., 2016). EG3301R is a project-based module that guides students to learn how to develop technology-enabled product ideas to address defined problems, and generate and evaluate concept designs by building prototypes and performing user testing.
CASE STUDIES
This paper introduces the three projects which utilised generative AI technology in their development process. The first project, “the Dentistry-geriatric patients’ communication training with VR service,” adopted Midjourney to generate geriatric patient characters and D-ID to create animation (see Figures 1 and 2).
The second project is an interprofessional education training service in healthcare and used Inworld to create various types of patient characters (Figure 3).
The third project involved developing a Korean language training AI chatbot that can help Korean learners practice diverse conjugation by adopting ChatGPT to generate various sentences (Figure 4).
CHALLENGES AND IMPLICATIONS
The use of generative AI can be challenging due to its novelty and students’ lack of experience. Thus, supervisors should introduce available services and help them scrutinise possible opportunities to adopt the most appropriate generative AI technology from the market. To facilitate this process, it would be helpful to establish a database of previous cases and share it with students to spread knowledge. Generative AI services can simplify recurring tasks in students’ technology- enabled projects, such as creating various characters and scenarios, as shown in Table 1. Supervisors should closely observe their concept design and development process and advise on how to effectively incorporate generative AI technologies. Educators can further encourage the use of generative AI tools by sharing case studies and promoting their integration in students’ technology projects.
Table 1
Three projects, their objectives to use generative AI and used services
| Project | Objectives to use generative AI services | Used generative AI services |
| Dentistry-geriatric patients’ communication training | Various characters, emotions, and animations generation | Midjourney, D-ID |
| Interprofessional education training service in healthcare | Various patient characters and scenario generation | Inworld |
| Edubot | Sentences and questions generation, character generation | ChatGPT, D-ID |
REFERENCES
Cooper, G. (2023). Examining science education in ChatGPT: An exploratory study of generative artificial intelligence. Journal of Science Education and Technology, 32(3), 444-52. https://doi.org/10.1007/s10956-023-10039-y
Gimpel, H., Hall, K., Decker, S., Eymann, T., Lämmermann, L., Mädche, A., Röglinger, M., Ruiner, C., Schoch, M., & Schoop, M. (2023). Unlocking the power of generative AI models and systems such as GPT-4 and ChatGPT for higher education: A guide for students and lecturers. Hohenheim Discussion Papers in Business, Economics and Social Sciences No. 02-2023. http://hdl.handle.net/10419/270970
Kokotsaki, D., Menzies, V., & Wiggins, A. (2016). Improving Schools, 19(3), 267-77. https://doi.org/10.1177/1365480216659733
Su, J., & Yang, W. (2023). Unlocking the power of ChatGPT: A framework for applying generative AI in education. ECNU Review of Education, 20965311231168423. https://doi.org/10.1177/20965311231168423
