When Algorithms Read Folklore: AI, Simulated Reception, and Cross-Cultural Studies of Literary Interpretation

Abstract

As artificial intelligence increasingly enters literary classrooms and digital humanities research, the question is no longer whether AI can produce interpretations, but whether it can meaningfully “read” culturally embedded texts. This study addresses a critical gap in AI-literature scholarship: limited attention has been given to AI-generated interpretation as a form of simulated reception, particularly in cross-cultural folklore contexts where meaning depends on emotion, moral judgment, and cultural situatedness. Grounded in reader-response and reception theory, this qualitative comparative study examines human and AI responses to four folktales: Bawang Merah Bawang Putih, Timun Mas, Cinderella, and Jack and the Beanstalk. Data were drawn from 30 undergraduate students’ reader responses and AI-generated interpretations using the same interpretive prompts. The analysis focused on three dimensions: intellectual, emotional, and cultural interpretation. The findings show that human readers produced varied, affective, morally evaluative, and culturally grounded responses, while AI generated structurally coherent but consistently affirmative interpretations. AI responses displayed algorithmic positivity, emotional flattening, and cultural abstraction, especially when confronting moral ambiguity, gendered agency, suffering, punishment, and cross-cultural value conflict. The originality of this study lies in conceptualizing AI as a simulated reader rather than an interpretive subject and in showing how folklore exposes the cultural limits of algorithmic interpretation. The study contributes to literary education, reception theory, folklore studies, and digital humanities by positioning AI not as a replacement for human interpretation, but as a comparative artifact for developing critical AI literacy and culturally responsive literary pedagogy.