Visual storytellers keep control while using large language models
Harnessing LLMs Without Surrendering Control: Delegation Boundaries in Visual Data Storytelling Authoring
Human-Computer Interaction
Summary
People who create stories using charts and visuals often work alongside large language models (LLMs) but don’t let these AI systems take full control. Instead, the storytellers let the AI handle routine writing tasks while they make the main decisions about what the story means and how it’s told. The researchers found that AI helps most after humans provide clear guidance and review its work carefully. This shifts the human effort from making content to checking it.
What this means in practice
- •For data visualization teams: Build authoring tools that let humans set boundaries while LLMs assist with drafting visual narratives.
- •For marketing content creators: Use LLMs to draft routine parts of visual story reports while maintaining manual control over key messages.
Authors
Zhuojun Jiang, Yuki Ueno, Chris Bryan
Abstract
Despite the emergence of large language models (LLMs) for visual data storytelling workflows, there are open questions about how authors decide what activities or tasks to entrust to them and what should be "protected" or maintained under human control. To investigate this, we interviewed a cohort of 12 expert visual data storytellers. Our analysis shows that participants rarely treated LLMs as autonomous storytellers. Instead, they tend to selectively delegate execution-oriented tasks to LLMs while retaining control over activities that shape narrative intent and story meaning. Our findings show that LLM assistance is most productive after human seeding and constraint-setting, and that it shifts labor from production to verification. We discuss design implications for boundary-aware authoring tools, data-grounded generation, low-fidelity ideation, and reporting practices for LLM-based visualization research. Supplemental materials for this paper are available at https://osf.io/hcnp6.