<![CDATA[The integration of generative artificial intelligence (AI) into English for Specific Purposes (ESP) instruction opens new possibilities for adapting authentic materials to meet the needs of novice learners. This study examines the effectiveness of different prompt engineering strategies in modifying the readability level of authentic English for Transportation texts while preserving essential domain-specific technical vocabulary. Five authentic texts drawn from the transportation and logistics domain, including port news articles, logistics manuals, maritime safety regulations, cargo handling procedures, and shipping route reports, served as source materials. Three types of prompts were applied to a generative AI tool (ChatGPT): Prompt A (a general simplification command), Prompt B (a CEFR A2-targeted rewriting command), and Prompt C (a structured prompt specifying CEFR A2 level and explicit instruction to retain all technical transportation terms). Readability was measured using the Flesch Reading Ease (FRE) formula, and technical term retention was evaluated through systematic lexical analysis. Results demonstrated that the original texts averaged an FRE score of 24.88 (Very Difficult), while Prompt A, Prompt B, and Prompt C produced average FRE scores of 50.18, 65.32, and 63.74, respectively. Notably, Prompt C achieved a 100% technical term retention rate, compared to 20% for Prompt A and 56% for Prompt B. The findings confirm that detailed, structured prompts specifying the target language proficiency level and vocabulary preservation requirements yield the most pedagogically appropriate ESP materials for vocational students at the beginner level. This study offers practical implications for ESP instructors seeking to leverage generative AI in developing teaching modules and course books for transportation programs.]]>
Copyrights © 2026
