Integrating Artificial Intelligence into High School STEM through Plant Identification

The integration of artificial intelligence (AI) into secondary education has emerged as a promising approach to enhance student engagement, promote interdisciplinary learning, and develop essential digital competencies. This study investigates the effectiveness of a short-term, AI-integrated learning activity aimed at improving high school students’ understanding of image processing and deep learning through plant identification using local Thai herbs. A pre-experimental one-group pretest-posttest design was employed with 26 purposively selected students who had limited exposure to coding or AI concepts. The intervention included hands-on experience with Python and Google Colab, using a pre-trained VGG16 model to classify four morphologically similar plant species. Conceptual understanding was assessed using a 15-item test and analysed through paired samples t-tests and Wilcoxon signed-rank tests. Results indicated a statistically significant improvement in post-test scores ($\mathbf{p}\lt.001, \mathbf{d}=0.92$), with students demonstrating enhanced comprehension of AI-related tools and techniques. The findings suggest that even shortduration workshops, when grounded in culturally relevant content and authentic coding practice, can effectively foster AI literacy in secondary education. Implications and limitations for curriculum design and further research are discussed.