Evaluating the Feasibility of Using Generative AI for Educational Research within the Context of Vietnam’s 2018 General Education Curriculum

This study examines the feasibility of using generative AI tools in educational research within the context of Vietnam’s 2018 General Education Curriculum. The research evaluates three AI-powered tools – ChatGPT, Gemini, and Copilot - amid growing interest in AI's integration into academic fields, particularly in education. The focus is on their strengths across specific areas of educational research: curriculum development, implementation requirements, and evaluation and assessment. The tools' performance is assessed based on five criteria: accuracy, comprehensiveness, logical clarity, relevance, and currency of information. ChatGPT performs effectively in global citizenship education (curriculum development) while Gemini excels in history assessment standards (evaluation and assessment). Copilot shows promise but struggles with accuracy in certain domains. Despite variations in performance, all tools demonstrate potential in improving research processes, especially in tasks where absolute precision is not critical. However, accuracy remains a significant challenge across all platforms. The findings suggest that AI tools can greatly enhance academic work when used with proper verification and structured commands, underscoring their practical applications and future potential in transforming research methodologies.