Artificial-Intelligence-Assisted Workflow Optimization for Digital Media Art Creation

Digital media artists often face fragmented workflows, slow translation of ideas into concrete assets, limited personalised support and high technical barriers. This paper proposes an artificial-intelligence-assisted workflow designed to streamline the full creative pipeline for digital media art. We decompose creation into four stages—creative stimulation, asset generation, iterative co-creation and output optimisation—and embed generative models, computer vision and natural language processing at each stage. A central AI services layer combines a prompt engine, diffusion-based visual generators and retrieval modules, while a lightweight user-preference model adapts recommendations to individual styles. The system integrates into standard illustration, motion graphics and interaction design tools through dockable panels rather than replacing existing software. We evaluate the framework with 48 students and early-career practitioners completing illustration, motion and interactive tasks under two conditions: a traditional digital workflow and the proposed AI-assisted workflow. Metrics include ideation time, number of useful assets produced per hour, number of iterations to reach an acceptable concept, objective expert ratings and participant-reported outcomes. Results show that the AI-assisted workflow reduces early ideation time by an average of 43.2%, increases the rate of usable asset generation by 51.7% and lowers the number of required iterations by 29.4%, without a statistically significant drop in perceived authorship or originality. Participants report that AI support reduces technical friction and expands the range of explored visual directions while keeping final decisions in human hands. The study suggests that carefully designed AI assistance can lower barriers for digital media art creation and improve efficiency, while preserving room for individual expression and critical judgement.