This paper examines the implications of AI translation technologies for foreign language education, emphasizing their potential to alter motivations for language learning fundamentally. English has served as the lingua franca, facilitating communication across language barriers in a globalizing world. However, AI-powered translation tools allow seamless, near-instant communication between speakers of different languages, potentially diminishing the practical necessity of learning foreign languages like English. Drawing on cognitive linguistics, the paper argues that foreign language education offers more than just communication skills—it cultivates new ways of construing the world. Language shapes thought processes and cultural perspectives, as demonstrated by differences in linguistic construal between Japanese and English. Through language learning, learners not only recognize these differences but also develop metacognitive abilities to reflect on their ways of construing the world. While AI translation enables accessibility and convenience, it risks obscuring subtle cultural and cognitive distinctions embedded in languages, i.e., invisible culture. Foreign language education, therefore, remains indispensable for fostering critical thinking, cultural empathy, and the ability to challenge the illusion of complete mutual understanding that AI translation might create. This paper stresses that language education in the AI era plays a crucial role in visualizing internal diversity of different language speakers.

This study aims at clarifying a part of students’ learning process by using conflict resolution tasks to achieve proactive and interactive learning as it appears in the national teaching guidelines. Students tried to mediate the two different claims by interacting with others in conflict resolution tasks, which included various communication skills; claiming their own ideas, accepting other’s opinions, and negotiating. In this study, I had students write what they felt and learnt by conflict resolution tasks. I collected the data from 23 students and analyzed them using M-GTA. The result of the analysis was divided into two factors: proactive learning and interactive learning. Having students who experienced claiming their opinions to the opponent was likely to lead to proactive learning. On the other hand, students who assumed a mediator role tended to experience interactive learning. Through the conflict resolution tasks, we have concluded that students were considering their learning more deeply.