Abstract:
Artificial intelligence (AI) has rapidly entered school education, yet evidence-based accounts of how
it can be integrated into primary reading lessons in pedagogically responsible ways remain limited.
This study reports a classroom-based case study of an AI-integrated Grade 3 reading lesson, “My
Favorite Book,” implemented in a technology-rich primary classroom in Ho Chi Minh City, Vietnam.
The study was guided by the 2018 Vietnamese General Education Curriculum and informed by
scholarship on AI literacy, multimodal learning, and formative assessment. Data sources included
curriculum and lesson-plan analysis, classroom observation, students’ oral responses, digital learning
products, self-assessment forms, and rubric-based performance data from 20 students. Descriptive
statistics were used to summarize frequencies and percentages across four performance levels, while
qualitative evidence was used to interpret the pedagogical roles of AI across lesson stages. The findings
suggest that AI-supported tools strengthened four aspects of reading instruction: activation of prior
reading experience, multimodal access to the text, differentiated classroom participation, and the
collection of richer formative assessment evidence. Most student outcomes clustered at the Good and
Very Good levels, especially in prediction, oral sharing, comprehension responses, and creative
extension tasks. At the same time, the study indicates that AI did not function as an autonomous
instructional agent; its educational value depended on careful teacher orchestration, explicit curricular
alignment, and selective use of digital tools. The article contributes a theoretically grounded and
practice-sensitive model for integrating AI into primary reading instruction while also clarifying the
methodological limits of single-classroom evidence. |
Keywords:
artificial intelligence, Grade 3, multimodal reading, formative assessment, primary
education
|
