<![CDATA[The emergence of generative artificial intelligence (AI) fundamentally challenges traditional literary paradigms that burden the modern reader with authenticating the text's source. This study empirically investigates the reception of AI-generated flash fiction (Text A) and human-written flash fiction (Text B) by 16 4th Semester English Literature students at Universitas Ngudi Waluyo. The research utilizes a mixed-methods approach grounded in Hans Robert Jauss’s reception theory, particularly the concept of the horizon of expectations. The core research methodology employs a "mirrored prompt" approach to ensure high internal validity, giving the human and AI authors the same core narrative and emotional task. The questionnaire analyzed the students' literary experience, Technological Horizons, and Interpretative Horizons. The results show a consistent and significant preference for Text B (Human-written) across all measured dimensions of the Interpretative Horizon, particularly in terms of emotion and stylistics, compared to Text A. Eleven out of 16 students (68.75%) accurately identified Text A as AI-generated and Text B as human-written. Qualitative data reveal that students critique text A for its lack of affective resonance, while text B has a "natural and flowing style”. This finding empirically validates that the reader’s interpretative horizon, particularly the expectation for deep emotion and unique style, is the primary factor in determining the perceived authenticity of a text, thus updating Jauss's theory to include the challenge of algorithmic works. The accuracy rate (68.75%) is significantly higher than previously reported research, suggesting that academic literary competence may increase the ability to discern AI-generated fiction.]]>
