Lost in Algorithms: A Comparative Analysis of Human and AI Translation of Culture-Specific Items in Alkoni works

Abstract

This study investigates applying the artificial intelligence tools in translating culture-specific items in literary works. A heated debate has been initiated among scholars since the beginning of using AI tools in translation; the core of this debate was about the ability of AI translation tools to convey cultural nuances compared to human translators. Furthermore, techniques are usually adopted by AI translators in case of a lack of equivalence. To investigate these queries, fourteen excerpt, containing cultural elements have been chosen from three Libyan novels by Ibrahim Al-Koni, The Bleeding of the Stones, Dust Gold and Anubis, are utilized as a case study to compare the human translator's work with AI translation tools, namely, ChatGPT and DeepL. The major results of this study are the performance of AI tools almost the same as human translators. Although AI      translation tools outperformed human translators' work in some situations, they still require specialized training on culture specific items related to specific communities. A collaboration between human and AI tools results in significant translation rather than separating or substituting each other. The mixing between machine and human leads to important progression in the field of translation.