Paper 2606.11176
Data Journalist Agent: Transforming Data into Verifiable Multimodal Stories
- Published
- Jun 2026
- Research lab
- Independent
- Citations
- 0
- GitHub
- 149 stars
01 In brief
Summary
The paper introduces Data Journalist Agent (Data2Story), a multi-agent framework that transforms raw data into verifiable, multimodal articles.
It orchestrates seven roles (Detective, Analyst, Editor, Designer, Programmer, Auditor, Inspector) into a virtual newsroom.
Key innovations are evidence-grounded claims via an Inspector that links each claim to code or references, and multimodal generative storytelling (e.g., interactive maps, audio).
Evaluated on 18 articles against human-written pieces, Data2Story showed competitive performance, with strengths in transparency and auditability (93% of claims traceable vs.
25% for human articles).
Human raters (n=53) preferred Data2Story in 74% of cases, but human articles retained an edge in editorial angle, creative design, and informative presentation.
The authors position Data2Story as a collaborator, not a replacement, for journalists, enabling more evidence-based and verifiable reporting.
Limitations include full automation without human feedback loops and a gap in narrative depth compared to human writers.
The work demonstrates potential for augmenting newsroom workflows and surfacing niche datasets, but acknowledges human strengths in reporting and design craft remain unmatched.
Future work includes incorporating human feedback and direct comparisons of narrative depth.
The paper also discusses evaluation protocols including human-agent angle coverage, rubric evaluation, computer-use agents as judges, and verifiability checks.
Overall, Data2Story advances automated data journalism by prioritizing evidence traceability and multimodal engagement, though it does not yet match human creativity and editorial judgment.
The authors emphasize the importance of verifiability in AI-generated journalism, addressing hallucination risks.
The system's ability to discover original findings on underexplored data (e.g., FIFA 2026 schedule, arXiv submissions) highlights…
02 From the paper
Abstract
Data tells stories that shape society; the data journalist's job is to turn raw information into stories non-experts can trust. A high-quality news feature takes a newsroom team weeks: hunting for context, running statistics, choosing an angle, and designing visuals. Recent agents handle individual steps well: data-science agents close the analysis loop, while design agents synthesize beautiful websites. But can an agent serve as a data journalist end to end? We introduce Data Journalist Agent (Data2Story), a multi-agent framework that orchestrates specialized roles into a single virtual newsroom. Data2Story contributes two innovations. (i) Claims are evidence-grounded: an Inspector links every number, angle, and asset back to data, code, or an external reference. (ii) Articles are multimodally generative: rather than defaulting to plain text and static charts, Data2Story reasons about what readers will want to see, then deploys multimodal tools, such as interactive maps for geography and audio for music. We evaluate Data2Story on 18 articles, each paired with the originally published expert piece, along four axes: (a) human-agent angle coverage; (b) rubric evaluation with 53 participants across five dimensions; (c) computer-use agents as judges, a cost-saving proxy for how readers navigate interactive articles; and (d) verifiability, where a coding verifier re-executes statements against the data and checks claims against references. Data2Story produces competitive, evidence-traceable multimedia stories, with particular strength in transparency and auditability. Human articles retain an edge in editorial angle, creative design, and presentation. We position Data2Story as a collaborator for journalists, enabling more evidence-based, transparent, and verifiable reporting. Code and demos are available at https://data2story.github.io.