How AI Refines Authorship: Evidence from Text and Content Analysis
(A2026-133915)

AUTHORS

Zizhou Peng, University of Birmingham; Yuan Yuan, Warwick Business School; Seungcheol Lee, Texas Tech University; Yedo Ji, Texas Tech University

ABSTRACT

Generative AI transforms creative work, yet how humans and AI co-author narratives remains unclear. This study examines how interaction depth shapes authorship and perceived creativity. We collected 268 stories written under human-only and AI-assisted conditions, evaluated via rater assessments and linguistic analysis. Human stories were judged more useful than AI-assisted ones. However, deeper human–AI interaction improved creative evaluations compared to minimal prompting. Linguistic analyses revealed that AI-assisted stories emphasized scenic description and positivity, while human stories showed greater subjectivity, conflict, and emotional nuance. These findings demonstrate how iterative interaction activates human–AI complementarity, advancing understanding of distributed creativity and informing the design of co-creative systems.