The state of AI-assisted data visualization research
An explanatory field guide to goals, use contexts, techniques, convergence, disagreement, failure evidence, benchmarks, and the 36-month capability timeline.
OpenScope
One maintained body of research, with different ways in.
The visual experiences support comparison and exploration. The written reports preserve the complete arguments, methods, caveats, and source links. This edition is dated because model capability, tools, and practice are moving quickly.
Visual experiences
Explore the evidence and its relationships.
The practitioner and reader experience
What people are trying to accomplish, which tools they reach for, what feels newly possible, and where correction, trust, maintenance, and reader experience still break down.
OpenHow close are we to the ideal?
Context-specific scorecards for generation and understanding, reconstructed backward through time and connected to explicit forecasts.
OpenMap of the research corpus
A non-linear view of the questions, reports, evidence layers, gaps, and relationships across the complete working corpus.
OpenWho needs this research?
A reader-first map of audiences, decisions, useful formats, discovery routes, and evidence needed to distinguish possible output from accomplished work and actual use.
OpenWritten reports
Follow a question through the full analysis.
Core synthesis
Executive summary
The central findings, limits, and practical implications in one short reading path.
OpenResearch review
The full literature synthesis and technique history behind the visual field guide.
OpenPractitioner and reader experience
Tool landscape, jobs, creator accounts, reader reactions, and the gap between possibility, accomplishment, and experience.
OpenIdeal state and capability scorecards
Definitions of excellent generation and understanding, current scores by context, historical reconstructions, and forecast states.
OpenCapability and evidence deep dives
What the benchmarks actually measure
A crosswalk from benchmark scores to the capability demonstrated—and the claims each benchmark cannot support.
OpenData-visualization skills for coding agents
A comparative reading of skill packages, their techniques, adoption signals, testing evidence, and likely durability as base models improve.
OpenSpecialized vision models
Where chart-specific perception, OCR, structured reconstruction, and visual critics outperform general models—and where they do not.
OpenHuman skills and banked gains
What novices, intermediate practitioners, and experts gain today, what remains unproven, and which human capabilities become more important.
OpenLearning when AI can make the chart
What becomes easier to learn, what can be lost when learners skip construction, and how curricula should rebalance.
OpenWhat would unlock the next capabilities?
Measurable gates, evidence-based timeframes, and the distinction between gains from stronger models and gains from better technique.
OpenPublishing and use
Audience, decisions, and rollout
Who may use the work, what decisions it should help, and how to gather evidence of actual use without confusing attention with impact.
OpenEditorial architecture
How the comprehensive report and a non-linear series can coexist as two views of one maintained research corpus.
Open