logo
ResearchBunny Logo
Experimental narratives: A comparison of human crowdsourced storytelling and AI storytelling

Interdisciplinary Studies

Experimental narratives: A comparison of human crowdsourced storytelling and AI storytelling

N. Beguš

This research by Nina Beguš explores a groundbreaking framework that merges behavioral and computational experiments using fictional prompts to unveil cultural artifacts and social biases in storytelling. Delving into 250 human stories and 80 AI-generated narratives, it uncovers how AI, particularly GPT-4, portrays progressive themes in gender roles and sexuality, while highlighting the imaginative superiority of human storytelling. Discover how fiction can bridge understanding between human and AI social dynamics.

00:00
00:00
Playback language: English
Abstract
This paper introduces a framework combining behavioral and computational experiments using fictional prompts to investigate cultural artifacts and social biases in storytelling by humans and generative AI. Analyzing 250 human stories and 80 AI-generated stories (GPT-3.5 and GPT-4), the study reveals that AI narratives, particularly those from GPT-4, show more progressive gender roles and sexuality than human-authored texts. While AI can offer innovative plot twists, human stories demonstrate more imaginative scenarios and rhetoric. The framework suggests fiction as a tool for understanding human and AI collective imaginary and social dimensions.
Publisher
HUMANITIES AND SOCIAL SCIENCES COMMUNICATIONS
Published On
Oct 28, 2024
Authors
Nina Beguš
Tags
storytelling
cultural artifacts
social biases
AI narratives
gender roles
human imagination
Listen, Learn & Level Up
Over 10,000 hours of research content in 25+ fields, available in 12+ languages.
No more digging through PDFs, just hit play and absorb the world's latest research in your language, on your time.
listen to research audio papers with researchbunny