StoryScope: Investigating Idiosyncrasies in AI Fiction
A new study titled 'StoryScope' investigates the differences between AI-generated and human-written fiction. The research focuses on discourse-level narrative choices rather than stylistic elements to distinguish between the two. Findings suggest that narrative construction differences can effectively separate human-authored works from AI-generated stories.
- ▪The study analyzes 10,272 writing prompts from human authors and five language models, resulting in 61,608 stories.
- ▪Narrative features alone achieved a 93.2% macro-F1 score for distinguishing human from AI writing.
- ▪AI-generated stories tend to over-explain themes and have simpler plots compared to the more complex narratives of human authors.
Opening excerpt (first ~120 words) tap to expand
Computer Science > Computation and Language arXiv:2604.03136 (cs) [Submitted on 3 Apr 2026 (v1), last revised 13 Apr 2026 (this version, v4)] Title:StoryScope: Investigating idiosyncrasies in AI fiction Authors:Jenna Russell, Rishanth Rajendhran, Chau Minh Pham, Mohit Iyyer, John Wieting View a PDF of the paper titled StoryScope: Investigating idiosyncrasies in AI fiction, by Jenna Russell and 4 other authors View PDF Abstract:As AI-generated fiction becomes increasingly prevalent, questions of authorship and originality are becoming central to how written work is evaluated.
…
Excerpt limited to ~120 words for fair-use compliance. The full article is at arXiv.org.
Discussion
0 commentsMore from arXiv.org
