A Comparative Study on the Effectiveness of ChatGPT and DeepSeek in the English Translation of Characteristic Terms from Classical Chinese Medical Texts—A Case Study of Huangdi Neijing: Jue Lun Chapter 45

Abstract

This study employs Chapter 45 (Jue Lun) of Huangdi Neijing (Yellow Emperor’s Inner Classic) as the source text to compare the performance of ChatGPT and DeepSeek in the English translation of characteristic terms from traditional Chinese medicine (TCM) classics. Results indicate that while ChatGPT produces fluent translations with strong cross-contextual adaptability, it exhibits terminological inconsistencies and a tendency to dilute deep cultural connotations. In contrast, DeepSeek, leveraging a domain-specific knowledge base, demonstrates greater stability in recognizing the underlying medical rationale of core terminologies, thereby generating translations with higher academic rigor. However, neither model fully reconciles the competing demands of terminological precision, textual readability, and philosophical depth. Consequently, this study proposes a “Generation-Proofreading-Optimization” human-AI collaborative translation model. This approach integrates the technological strengths of both systems while emphasizing the translator’s pivotal role in strategy formulation and cultural gatekeeping, offering a viable pathway for the intelligent translation and international dissemination of TCM classics.