Exploring Thai Students’ Strategies in Using AI Chatbots for Vietnamese Vocabulary Learning

Generative artificial intelligence (GenAI) tools have become common resources for vocabulary learning among foreign language learners, with AI chatbots attracting particular interest for their interactive and responsive features. Despite this trend, research on AI-assisted vocabulary learning in less commonly taught languages remains limited. This study investigated the vocabulary learning strategies (VLS) that Thai university students employed when using AI chatbots to learn Vietnamese and examined their perceptions of these tools. Both Thai and Vietnamese are tonal languages, but their tonal systems differ substantially, creating learning challenges that text-based AI may not effectively address. An explanatory sequential mixed-methods design was employed, involving seventy students at a university in Northeastern Thailand. The results showed that reference and translation strategies were the most frequently used, whereas production strategies were underutilized. Strategy use varied significantly across academic year groups but stabilized from Year 3 onward. Moreover, students perceived AI chatbots as effective for meaning and contextual learning but less so for tonal development. The findings point to prompt engineering as a potential new dimension in VLS frameworks and support a blended learning approach, combining AI tools with teacher-led pronunciation instruction. As an exploratory study, these findings should be interpreted with caution and confirmed through larger-scale research.