Can AI tell good stories? Narrative transportation and persuasion with ChatGPT

Abstract Storytelling is a human universal. The ubiquity of stories and the rapid development in Artificial Intelligence (AI) pose important questions: can AI like ChatGPT tell engaging and persuasive stories? If so, what makes a narrative engaging and persuasive? Three pre-registered experiments comparing human-generated narratives from existing research and the ChatGPT-generated versions using descriptions and materials from these studies show that labeling AI as a narrative source led to lower transportation, higher counterarguing, and lower story-consistent beliefs. However, AI-generated narratives led to lower (Study 1 and 3) or similar levels (Study 2) of counterarguing than the human-generated version. Readers showed lower (Study 2) or similar levels of transportation (Study 1 and 3) when reading the AI- than the human-generated stories. We suggest the AI model’s linguistic competence and logical coherence contribute to its stories’ verisimilitude. However, AI’s lack of lived experience and creativity may limit its storytelling ability.

Authors

Institutions

Publication Details

Journal
Journal of Communication
Published
2024-09-13
DOI
https://doi.org/10.1093/joc/jqae029
Citations
31
Primary Topic
Media Influence and Health
Type
article
Field-Weighted Citation Impact
121.16
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
article

Can AI tell good stories? Narrative transportation and persuasion with ChatGPT

Sixiao Liu, Haoran Chu
31 citations
Journal of Communication
Media Influence and Health
121.16
article

Can AI tell good stories? Narrative transportation and persuasion with ChatGPT

Sixiao Liu, Haoran Chu
article en
31 citations

Abstract

Abstract Storytelling is a human universal. The ubiquity of stories and the rapid development in Artificial Intelligence (AI) pose important questions: can AI like ChatGPT tell engaging and persuasive stories? If so, what makes a narrative engaging and persuasive? Three pre-registered experiments comparing human-generated narratives from existing research and the ChatGPT-generated versions using descriptions and materials from these studies show that labeling AI as a narrative source led to lower transportation, higher counterarguing, and lower story-consistent beliefs. However, AI-generated narratives led to lower (Study 1 and 3) or similar levels (Study 2) of counterarguing than the human-generated version. Readers showed lower (Study 2) or similar levels of transportation (Study 1 and 3) when reading the AI- than the human-generated stories. We suggest the AI model’s linguistic competence and logical coherence contribute to its stories’ verisimilitude. However, AI’s lack of lived experience and creativity may limit its storytelling ability.

Journal of CommunicationVol. 74(5)
University of Central Florida (US), University of Florida (US)
Quality Education
Openalex Percentile: Top 0%
Media Influence and Health
121.16
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.