Subtitling Legal Expressions in English Series into Arabic by Netflix, Machine, and Artificial Intelligence

Ruba Alkhatib1 & Ahmad S. Haider2

https://doi.org/10.62271/pjc.16.4.513.528

Abstract
Artificial intelligence (AI) and machine translation (MT) have revolutionized translation. However, assessing AI and MT of audiovisual (AV) content has not yet gained much interest.  This study examines the strategies used to translate legal expressions by Netflix, Google Translate (GT), ChatGPT (GPT), and Gemini (GEM) in four English Netflix series. It also classifies the collected legal terms thematically. The results showed that the majority of English legal terms are related to criminal law. Using an eclectic approach of translation strategies, the findings showed that each translator (Human, MT, and AI) employs unique strategies, with paraphrasing being the most commonly used strategy (33.1%), followed by literal (26%) and cultural substitution (25.3%). The analysis also showed some cases of mistranslation, with Netflix showing the least and GT demonstrating the most errors. The results showed that AI and MT systems need further improvement in specialized fields such as legal. The study concludes that human translators follow subtitling requirements more precisely than machine translation and artificial intelligence systems do when it comes to legal terminology.

Keywords: Audio-visual Translation (AVT); Arabic; English; Series; Legal Expressions; Netflix; Machine Translation; Google Translate; Artificial Intelligence; ChatGPT; Gemini