Search

Generative AI and the Cross-Cultural Narration of Chinese Intangible Cultural Heritage: A Case Study of Shadow Puppetry

Author(s)

Zixuan Xu

Affiliation(s)

Department of Artificial Intelligence, Macau University of Science and Technology, Macau, China

Abstract

Generative artificial intelligence is now frequently used to draft museum labels, tourism introductions, bilingual public texts, and short educational materials. In the field of intangible cultural heritage, however, a fluent paragraph can still miss the practical knowledge carried by performers, craftspeople, and local communities. Taking Chinese shadow puppetry as a case, this study examines how ChatGPT and ERNIE Bot narrate one living tradition for English-speaking readers. The analysis uses six heritage dimensions and four evaluation indicators: factual accuracy, cultural specificity, narrative accessibility, and commercialization risk. The comparison shows that both tools can produce readable first drafts and summarize basic facts, but their strongest passages often remain close to common public descriptions. More delicate elements, such as regional singing styles, hand manipulation behind the screen, apprenticeship, ritual occasions, and the economic pressure of tourism, are less consistently represented. The paper therefore argues for an AI-assisted, human-reviewed model of heritage communication, in which machine-generated drafts are treated as raw material rather than cultural authority.

Keywords

Generative AI; Intangible Cultural Heritage; Shadow Puppetry; Cultural Communication; AI Narrative

References

[1] Bender, E. M., Gebru, T., McMillan-Major, A., & Shmitchell, S. (2021). On the dangers of stochastic parrots: Can language models be too big? In Proceedings of the 2021 ACM Conference on Fairness, Accountability, and Transparency (pp. 610-623). ACM. [2] UNESCO. (2003). Convention for the safeguarding of the intangible cultural heritage. UNESCO. [3] UNESCO. (2011). Chinese shadow puppetry. Representative List of the Intangible Cultural Heritage of Humanity. [4] Smith, L. (2006). Uses of heritage. Routledge. [5] Tilden, F. (1977). Interpreting our heritage (3rd ed.). University of North Carolina Press. [6] Cameron, F., & Kenderdine, S. (Eds.). (2007). Theorizing digital cultural heritage: A critical discourse. MIT Press. [7] Kirshenblatt-Gimblett, B. (2004). Intangible heritage as metacultural production. Museum International, 56(1-2), 52-65. [8] Nora, P. (1989). Between memory and history: Les lieux de memoire. Representations, 26, 7-24. [9] OpenAI. (2023). GPT-4 technical report. arXiv. https://arxiv.org/abs/2303.08774 [10] Wang, Q., Ding, K., Gao, H., Wang, H., & Xu, R. (2025). Error comparison optimization for large language models on aspect-based sentiment analysis. Unpublished manuscript.