Artificial Intelligence and the Translation of Culture-Specific Terms on Arabic Monolingual Signs

What is it about?

Artificial intelligence (AI) tools are increasingly used in various fields, including the translation of signs within linguistic landscapes. This field-based study investigates the performance of AI tools, namely, Google Gemini and DeepSeek, in translating culture-specific terms (CSTs). To this end, Arabic monolingual signs at tourist sites in Aseer region in Saudi Arabia were documented and a comparative content analysis approach was employed to evaluate the accuracy, cultural sensitivity, and contextual relevance of the AI-generated translations. These translations were then compared to human translations using the same criteria designed for independent human evaluation. The findings indicated that Gemini consistently outperformed DeepSeek in translating CSTs, demonstrating superior accuracy, greater sensitivity, a stronger ability to consider the context of the terms, and better orthography. While both tools employed transliteration and addition strategies, DeepSeek’s frequent use of substitution and struggles with orthography often resulted in inaccurate translations. The study’s unique insights highlighted Gemini’s effective use of translation strategies and its consistent, accurate performance with Optical Character Recognition (OCR), a feature that DeepSeek struggled with. The findings also highlight the crucial role that human translators play in proofreading and refining AI-generated translations loaded with CSTs, ensuring accurate and high-quality translations that lead to effective communication.

Featured Image

Why is it important?

AI and translation