Incipit adds explicit planning layer to help writers guide AI prose
Incipit: Axiom-Grounded Scaffolding for Human-AI Literary Creation
Computers and Society
Summary
Writing long stories with AI can be hard because you don't see the ideas that guide the story. The authors created Incipit, a tool that adds a planning step before the AI writes paragraphs. This step uses a big list of story ideas called literary axioms to help shape the story’s main points and characters openly. Writers can look at and change the plan before the AI writes scenes, making creativity more connected and clear. The tool is a prototype without tested proof that it makes stories better yet.
What this means in practice
- •For creative writing teams: Create detailed story plans using literary knowledge before AI generates narrative scenes to improve human-AI writing collaboration.
- •For game narrative designers: Use axiom-grounded blueprints to structure interactive storytelling elements before generating dialogue or story text with AI.
A position paper. It proposes an approach and reports no results.
Authors
Qiang Liu, Chunyi Zhao
Abstract
Large language models can produce fluent prose from short prompts, but direct prompt-to-text interaction gives writers limited access to the assumptions that shape a long narrative. We present Incipit, an implemented research prototype that introduces an explicit planning layer between a writer's intent and generated prose. This layer is grounded in literary axioms, defined as curated and reusable propositions about human experience and narrative craft. The prototype connects a knowledge base of 1455 axioms and 472 typed relationships with a five-round direction dialogue, a retrieval-and-selection pipeline, and a three-level blueprint covering creative premises, story beats and character arcs, and chapter outlines. Writers can inspect and edit these structures before using them as context for scene generation. Additional modules support real-event abstraction and five-dimensional diagnostic feedback. We describe the system design rationale, data flow, implementation boundaries, and a worked design example. As no controlled user study or independently rated output study has yet been completed, we do not claim that the system improves literary quality. Instead, we outline a future preregistered evaluation designed to distinguish the contribution of axiom grounding from that of hierarchical planning. The paper contributes a concrete architecture for using literary knowledge as an inspectable coordination object in human-AI writing.