Integration of AI-Based Tools in Teaching Ethnocultural Units: A Qualitative Comparative Study Between Kazakh and English-Language Learners

Abstract

The use of AI translation tools in foreign language learning is growing due to the rapid adoption of AI technologies in educational settings. There is limited research on how AI translations influence students' understanding of ethnocultural expressions and the development of their intercultural sensitivity. This study investigates how students understand ethnocultural expressions in AI-translated texts, identifies cultural errors they find, and examines how they verify and correct these translations. A qualitative research design utilizing content analysis was employed. The study involved 20 university students divided into two groups: Kazakh (n=10) and English (n=10). Students used a variety of AI tools to translate texts that contained ethnocultural units. Semi-structured interviews were conducted post-assignments to assess students' perceptions of the cultural appropriateness of translation, identify biases, and evaluate verification strategies. Cross-case content analysis was employed to examine the collected data. The findings showed that most students recognized the literal nature of AI translations and the loss of cultural nuances, particularly in idioms, proverbs, and metaphors. Patterns in the data suggested that Kazakh participants often drew on prior cultural experience when interpreting ethnocultural expressions, whereas English-language participants more frequently referred to analytical and contextual strategies. The most common verification strategies were comparing several AI systems, back-translating, analyzing context, and consulting additional sources. Additionally, engaging with AI translations fostered a critical mindset toward automated translation tools and raised cultural awareness. Overall, AI translation supports language learning and intercultural reflection, but its limitations in conveying deep cultural meaning require pedagogical guidance. Future research should expand sample sizes, examine different educational contexts, and incorporate quantitative measures of intercultural competence.