From text-to-image to text-to-video: the reconfiguration of prompting practice in human-AI co-creation in 2022–2025

Abstract

Introduction The movement of generative artificial intelligence from still-image to moving-image creation has reshaped the symbolic mechanism of human-AI co-creation. Methods This comparative observational study asks how prompt expression changes when the output shifts from static images to moving images, comparing two large-scale public prompt corpora under one neural topic-modeling protocol that combines lexical analysis of style, quality, technical, motion, camera, and format vocabulary with manual family coding and validity checks, and adds an exploratory user-focused topic structure. Results Video prompts weaken the style, quality, and technical decoration characteristic of image prompts and instead organize meaning around action, scene, and screen format. Because the corpora differ in platform ecology, model generation, period, and corpus construction as well as in modality, these differences are described rather than causally attributed, and the format result is the most interface-sensitive of them. Artist-style imitation loses its earlier prominence, a decline that survives the coding reliability checks, while everyday action and scene description forms the largest family in the adjudicated solution, a position that is sensitive to those checks. Discussion Prompting practice moved from selecting aesthetic styles toward specifying how a scene moves and how it is framed, and the prompting grammar observed here appears bound to the modality and the period. The terms that organized image prompts did not carry over intact to video prompts. These patterns suggest that the symbolic interface of human-AI co-creation is not a fixed grammar but an adaptive arrangement that responds to the salient capabilities of a generative system. The study contributes a comparative account of how the division of creative labor is reconfigured across media modalities, with implications for co-creation theory, generative interface design, and media literacy education.

The authors' abstract, as published at the source. Frontiers in Human Dynamics, 2026 · DOI ↗