Human Touch in AI Translation: Strategies Employed by Professional Translators in Post-Editing Arabic-English AI Generated Translations of Media Texts

Abstract

Artificial intelligence has been widely used in translation industries. However, not enough studies have been conducted to investigate the appropriateness of the translations produced by ChatGPT for Arabic- English media texts and the strategies that can be used to improve such translations in case they prove inappropriate. As such, the current study aimed to investigate the extent to which the translations provided by ChatGPT for media texts from Arabic into English are accurate and to identify the strategies employed by professional translators when they post-edit or improve the translations provided by ChatGPT for media texts from Arabic into English. The findings of the study indicated that the translations produced by ChatGPT for Arabic-English media texts are not totally appropriate and need to be post-edited by human translators. The result of the study also revealed that human translators employed three main strategies to improve the translations of ChatGPT for media texts from Arabic into English, namely addition or expansion, omission or semantic minimization and play over diction. These strategies have been proven to make the translations of ChatGPT more appropriate and acceptable to the recipients of the target language text.