Llms help data visualization designers create charts in new ways
Vibe Analysis: Exploring LLM Adoption by Data Visualization Practitioners
Human-Computer InteractionArtificial Intelligence
Summary
Data visualization designers are starting to use large language models (LLMs) to help with making charts and understanding data faster and more creatively. These models can assist in both technical tasks and creative parts of creating visuals, but they sometimes make mistakes because they weren’t specifically designed for visualization work. The researchers found that designers have developed a new workflow called vibe analysis to work alongside these AI tools. Some challenges include verifying charts and handling visualization-specific knowledge gaps. This study suggests future tools could combine AI with data visualization guidance to improve the process.
What this means in practice
- •For data visualization designers: Use AI-assisted techniques to speed up both creative and technical steps in making visualizations through vibe analysis workflows.
- •For business intelligence teams: Incorporate LLMs with visualization tools to reduce manual data prep and create better-checked charts faster.
Authors
Shani C Spivak, Aditi Krishna, Mahsan Nourani, Melanie Tory
Abstract
Large language models (LLMs) are enticing in their promise to support data visualization (Vis) through faster and simpler workflows for data prep, analysis, and visualization creation. Yet LLMs are notoriously error-prone and not built for data visualization tasks. Few studies have explored LLM adoption among Vis practitioners. To fill this gap, we conducted semi-structured interviews with members of the Data Visualization Society, a global community of data visualization designers. Our findings show that Vis designers actively use LLMs for both creative and technical aspects of the visualization process. A new visualization workflow is emerging, a process we call vibe analysis, analogous to vibe coding. Some key challenges raised by participants parallel those of vibe coding, while others are Vis-specific, like gaps in Vis knowledge and chart verification. This work opens up opportunities for research combining LLM-mediated work with Vis tools that incorporate data visualization guidance, constraints, and best practices.