LINGUOPRAGMATIC CRITERIA FOR ASSESSING THE QUALITY OF ARTIFICIAL INTELLIGENCE-GENERATED TRANSLATIONS
Keywords:
Keywords: artificial intelligence, translation quality assessment, linguopragmatics, pragmatic equivalence, post-editing, machine translation.Abstract
The rapid advancement of artificial intelligence (AI) has significantly transformed contemporary translation practices by enabling fast and accessible multilingual communication. Neural machine translation systems such as ChatGPT, Google Translate, and DeepL have achieved remarkable progress in producing grammatically accurate and semantically coherent translations. Nevertheless, these systems still experience considerable difficulties in preserving pragmatic meaning, communicative intention, discourse coherence, and culture-specific elements. Consequently, evaluating AI-generated translations requires criteria that extend beyond lexical and grammatical accuracy. This paper examines linguopragmatic criteria for assessing the quality of artificial intelligence-generated translations. Special attention is paid to pragmatic equivalence, communicative intention, discourse coherence, cultural adaptation, and contextual appropriateness as essential indicators of translation quality. The study argues that linguopragmatic evaluation provides a more comprehensive framework for assessing AI-generated translations and contributes to improving post-editing practices and translation quality.
