AI-Powered Data Storytelling: Transforming Raw Data into Actionable Insights

• Lilibeth DG. Antonio, DIT

  College of Information and Communications Technology, Bulacan State University, Philippines

  Author

The aim of this study was to determine the effectiveness of AI-driven data storytelling systems in converting raw data into actionable insights through the quantification of user ratings most specially in terms of clarity, accuracy, trustworthiness, and usefulness. The need for systems that can convert information into readable stories has increased as information complexity in different industries has continued to rise. In this research, the effectiveness of AI-stories in naturalistic data interaction situations was compared with human-authored stories generated through natural language generation (NLG) among a sample of 120 participants from different industries, such as government, healthcare, education, IT, and business. Participants were randomized into a control group, and human-written stories, or an experimental group, and AI-generated stories from a GPT-4–driven NLG system and Power BI dashboards. The findings revealed that the experimental group rated AI-generated stories significantly higher in intelligibility (M = 4.38, p < .001), user trust (M = 4.13, p < .01), and usefulness (M = 4.35, p < .001). While the control group attained a marginally higher accuracy rating, and the difference predicted was non-statistically significant. Thematic analysis revealed three broad experiential themes: improved accessibility and user engagement, guarded optimism over narrative trust, and perceived usefulness for decision-making environments. These findings support the potential of artificial intelligence as an assistive data narrative companion that can augment user understanding and engagement when used with existing data systems. The study implies that the optimal solution to address risks and improve trust is to integrate human-in-the-loop design, transparency features, and domain-specific calibration. This study adds a validated framework for the evaluation of AI-generated stories and provides real-world implications on the effective and ethical dissemination of information.

• PDF

Copyright (c) 2025 Package Printing

This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.

All articles published are made available under the Creative Commons Attribution–NonCommercial 4.0 International (CC BY-NC 4.0) licence.