The emergence of generative artificial intelligence (AI) has introduced new possibilities for human–AI collaboration in storytelling, spanning domains such as literature, education, media, games, and multimodal systems. However, current research remains fragmented, with few attempts to synthesize co-creative practices across these fields. This study presents a scoping review of 44 peer-reviewed publications (2020–2025), using Arksey and O’Malley’s framework to examine how AI-assisted storytelling is conceptualized, implemented, and evaluated. Through domain-based analysis, we identify distinct trajectories: literary systems focus on authorship and narrative coherence; educational applications emphasize engagement, creativity, and language development; game-based systems highlight adaptive narratives and player agency; media and film studies examine authorship and workflow transformation; and multimodal platforms enable accessible visual storytelling. Across these domains, AI is increasingly positioned as a creative collaborator, supporting ideation, personalization, and co-authorship. Yet challenges remain, including narrative inconsistency, limited user control, and questions surrounding authorship and agency. This review highlights the need for inclusive design principles, clearer conceptual frameworks for co-creativity, and interdisciplinary approaches to address ethical, cultural, and practical concerns. By mapping current trends and tensions, the paper offers a foundation for future research and responsible development of AI-assisted storytelling systems.
