AI tool helping news gathering
Using AI to enable an industrial-scale newsgathering assistant might be crucial for small and medium-sized newsroom because they are usually very limited in the kinds and quantities of news they can gather. This was the case with the mid-sized newsroom, Magyar Jeti’s 444.hu in Hungary: with only 70 employes, the outlet would not normally be able to assign correspondents to cover the range of stories developing across the world, or trending across social media. But AI changes that equation.
Using funding from Thomson Foundation, Magyar Jeti was able to build an AI-powered semi-automated news monitoring system that enabled a dramatic increase in the number of stories that journalists could access. The early results have been significant, enabling the processing of around 200,000 content items per month and the constant monitoring of 115 news sources across the country and the world. This new capability, when fully adopted, will be like adding dozens of new reporters to 444.hu’s newsroom. Before the project, journalists did what most mid-sized newsrooms do: monitor key websites, social feeds and wires by hand, one tab and one notification at a time. But the outlet wanted to broaden its international coverage and react more quickly to stories with relevance for local audiences. The answer was designing an AI-supported monitoring and story-discovery system that was able to dramatically expand the newsroom’s field of vision while fitting real editorial workflows. Read more about the details of this journey in this article.
