Research on the Comparison and Collaboration Paths between AI Translation and Human Translation: A Case Study of the Reports on the Work of the Government of China (2024-2026)

The Report on the Work of the Government of China serves as a core text for the international community to understand China’s policies and governance. However, some Western media reports have seriously misrepresented the true logic of China’s governance model, highlighting the urgency of strengthening the foreign translation of political texts to reduce international cognitive biases. Although AI translation technology has developed rapidly, its practical effectiveness in translating political texts remains to be empirically tested. Grounded in Vermeer’s Skopos Theory, this study compares AI translations with official human translations of Chinese idioms extracted from the Reports on the Work of the Government of China (2024-2026), focusing on strategies such as literal translation, amplification, and category shift. The comparative analysis reveals that AI can quickly produce lexically corresponding versions, but it falls short in adjusting strategies according to communicative purposes, compensating for cultural defaults, and conveying the intensity of policy implementation. Human translators, by contrast, are better at balancing the overall communication goals of external publicity and harmonizing accuracy with readability. In response, this study proposes a collaborative framework in which AI generates initial drafts while human translators lead the review and refinement process, and suggests building a strategy-instance database to enhance AI’s contextual decision-making. This study provides empirical evidence for the division of labor and cooperation between human and machine translation.