"This could save us months of work" -- Use Cases of AI and Automation Support in Investigative Journalism

TL;DR

The paper investigates how automation and AI, including Programming-by-Demonstration and LLMs, can support investigative journalism amid growing data complexity. Through eight qualitative interviews with German investigative journalists and a speculative JournalXRecorder prototype, it derives a taxonomy of practical use cases across data collection and reporting, and provides design guidelines for end-user–focused automation tools. Key contributions include a detailed breakdown of collecting activities (monitoring, filtering, documenting, storing, augmenting) and reporting activities (analyzing, labeling, writing), plus considerations for multi-modal agents, cloud storage, and transparent documentation. The findings highlight both the potential for speeding up investigations and the need to manage risks such as hallucinations, bias, and data literacy, underscoring the importance of human-in-the-loop, co-creative tool development for trustworthy AI-assisted journalism.

Abstract

As the capabilities of Large Language Models (LLMs) expand, more researchers are studying their adoption in newsrooms. However, much of the research focus remains broad and does not address the specific technical needs of investigative journalists. Therefore, this paper presents several applied use cases where automation and AI intersect with investigative journalism. We conducted a within-subjects user study with eight investigative journalists. In interviews, we elicited practical use cases using a speculative design approach by having journalists react to a prototype of a system that combines LLMs and Programming-by-Demonstration (PbD) to simplify data collection on numerous websites. Based on user reports, we classified the journalistic processes into data collecting and reporting. Participants indicated they utilize automation to handle repetitive tasks like content monitoring, web scraping, summarization, and preliminary data exploration. Following these insights, we provide guidelines on how investigative journalism can benefit from AI and automation.

Figures (1)

• Figure 1: JournalXRecorder, prototype user interface. A) Main page where user can see their workflow, track events, and ask LLMs to automate. B) Running automation and generating scripts. C) Chat with and analyze the collected data.