Overview

Public discourse is shaped less by “facts” than by the stories used to organize them. Across long-form media, speakers frame problems, assign blame, define victims and heroes, and advance solutions through recurring narrative patterns. These storylines shape how audiences interpret complex social issues, yet they are difficult to study systematically at scale.

Storyline develops an auditable framework for tracing those patterns across podcasts, conversations, and news. The project uses AI-assisted analysis to identify storylines, compare how different sources frame the same issue, and surface where narratives converge, diverge, or directly conflict. Its outputs are designed to remain traceable to source evidence, with verification workflows that help researchers inspect, challenge, and refine machine-generated interpretations.

By making narrative structure legible and comparable, Storyline supports new forms of computational social science and human-centered sensemaking. The project aims to help researchers understand how public meaning is constructed, how narrative conflict travels across media ecosystems, and how AI can support rigorous discourse analysis without obscuring provenance or human judgment.