MULTIMODAL PRAGMATIC ERRORS IN AI-MEDIATED CROSS-CULTURAL COMMUNICATION: ANALYZING CULTURAL MISMATCHES IN GESTURES AND EXPRESSIONS IN ENGLISH-UZBEK AI TRANSLATION

Abstract

This article explores multimodal pragmatic errors in artificial intelligence (AI)-mediated cross-cultural communication, focusing on the English-Uzbek language pair. It examines how AI translation systems fail to account for nonverbal elements such as gestures and facial expressions, leading to cultural mismatches and misunderstandings. Drawing on pragmatic and sociolinguistic theories, the study analyzes specific examples of gestural differences and their implications in AI tools like Google Translate or advanced multimodal AI systems. Empirical evidence from case studies demonstrates how these errors affect interpersonal interactions, business negotiations, and educational exchanges. The analysis highlights the need for culturally adaptive AI models. Recommendations include integrating multimodal data training to enhance pragmatic accuracy. This research contributes to the growing field of AI pragmatics by providing a detailed cross-cultural perspective on English and Uzbek nonverbal cues.