Author: Xolliyeva, Sarvinoz Ikramovna;

Annotation: This article provides a scientific analysis of the complexities of the translation process of the lexicon related to the field of artificial intelligence (AI) in English and Uzbek, especially the linguistic inconsistencies arising due to new technologies. AI terms are formed at the intersection of human perception, computational systems and statistical models, and unlike traditional language units, they represent a multi-layered semantic structure, abstract concepts and cognitive processes. Therefore, their direct translation from English to Uzbek often leads to semantic distortion, narrowing of meaning or excessive expressiveness. The article highlights the differences between the conceptual content of terms such as “reinforcement learning”, “multimodal system”, “self-attention mechanism”, “parameter tuning” and “bias mitigation” in English and their use in Uzbek through practical examples. Such cases show that, in addition to semantic compatibility, conceptual compatibility is also a key criterion in the translation of AI lexicon.

Keywords: AI terminology, Translation strategies, Linguistic adaptation, Semantic incompatibility, Conceptual divergence, Multimodal systems, Technical discourse, Acquisition process, Cognitive compatibility.

Pages in journal: 396 - 403