A Comparative Study of Accuracy in Human vs. AI Translation of Legal Documents into Arabic

This study investigates the comparative accuracy of human and AI-generated translations of legal documents into Arabic, focusing specifically on the performance of ChatGPT against human translations. This study employs a comparative research design, where a corpus of words 20,000 words from legal texts, including contracts and agreements, translated by both AI and professional human translators. The research aimed to assess three primary dimensions: correct legal terminology usage, clarity of expression, and adherence to the Arabic legal framework. Through a structured evaluation process, key findings revealed that human translations significantly outperformed AI-generated ver- sions in all assessed criteria. Human translators demonstrated superior mastery of legal terminology and clarity in complex legal constructs, as well as adherence to formal legal standards and cultural differences inherent in Arabic legal contexts. While AI tools like ChatGPT show promise in producing contextually relevant translations for simpler texts, they often fall short in capturing the precise legal terminology and complex constructs required for effective legal communication. This research highlights the continued necessity of skilled human translators in the legal field and suggests a hybrid approach that leverages AI tools to augment human expertise in translation processes. Cite as: Altakhaineh et al., JLL 14 (2025), 63-80, DOI: 10.14762/jll.2025.063