GTA funded - Controlled Character-Driven Story Generation with AI University of Leicester in United Kingdom

GTA funded PhD studentship in Computing

Highlights

Integration of AI and traditional narratology

Analysis of the trade-off between structure and agency

Study of the interactions of generative agents in a simulation environment

Project

This research investigates the intersection of traditional narratological theories and advanced artificial intelligence (AI) methodologies, with a particular focus on the automated generation of structured, character-driven narratives. The project explores the use of AI-powered generative agents, built on large language models (LLMs), as narrative characters. Each agent is initialized to embody the complexity and depth characteristic of well-developed fictional characters.

A key aspect of this study is the balance between control through established narratological frameworks (e.g., Freytag’s Pyramid, the Hero’s Journey, or Propp’s morphology of folktales) and the agency of generative character agents. By structuring agent interactions and narrative development within predefined narrative models, the research aims to ensure that the resulting stories adhere to recognized storytelling principles while maintaining coherence and engagement.

The generative agents operate autonomously and interact with each other in a simulated environment within the constraints of their predefined arcs, contributing to a balance between predictability and variation in the narrative. This aspect allows for an exploration of the trade-offs between structured narratology and the emergent agency of AI-driven characters.

The findings of this research could be relevant across multiple domains. In the entertainment industry, potential applications include video games, where character-driven AI could enhance immersion, as well as film and television, where AI-generated scripts and interactive narratives may provide new creative opportunities. In a commercial context, AI-driven storytelling could be applied to advertising and marketing strategies or integrated into chatbot technologies to improve user engagement.

Enquiries to project supervisor Dr Marco Volpe [email protected]

General enquiries [email protected]

Please carefully read the information on our web page before applying

How to Apply https://le.ac.uk/study/research-degrees/funded-opportunities/computer-science-gta

There are 3 GTA studentships available. You can only apply for one project