What this page is for
This is a plain-language explanation of how AI tools are used across ftrcrp.org. It exists so a reader can know, at a glance, what role AI played in any piece they are reading. The legal scaffolding is in the editorial standards page, this one is for people who want a straight answer.
How AI is used
The site publishes across several sections, each handled differently.
Signals. Short, time-sensitive notes. Fully human-authored. No AI is used for drafting, generation, or research. The disclosure block at the foot of each Signal says so explicitly.
Investigations. Longer reporting work. AI tools are used for research and for parts of the generation. A human is in the loop at every stage, drafts are reviewed, edited, fact-checked, and signed off by a named natural person before publication. The per-article block at the foot of each piece states this and lists the tool categories used.
Analysis, opinion, guides. Same posture as investigations. AI-assisted research and generation, human in the loop, named editorial responsibility. The per-article block at the foot states this.
Security Digest. A weekly automated pipeline. Sources are collected by automated scrapers, summarised by local LLM, and drafted into the issue end-to-end. Human review is light-touch and limited to publication gating. Every digest issue carries a disclosure block at the foot stating this explicitly.
Legacy posts. Anything published before 19 May 2026, when the formal disclosure regime came into effect. These carry a general retro disclaimer noting that AI tools were used to polish and generate some text in the article. They are not retroactively re-labelled per the new mode taxonomy, that would be guesswork.
Editorial responsibility
All content on ftrcrp.org is editorially the responsibility of Thomas A. Kleppestø as a named natural person. This is the Article 50(4) editorial-responsibility carve-out from the EU AI Act, invoked because the published work is journalistic and academic in character, with human editorial responsibility for content, accuracy, and corrections.
A real person reads, edits, and signs off on what is published. Automated content does not bypass this gate, the gate is light-touch on digest issues by design but it is not absent.
Hva betyr dette på norsk
ftrcrp.org bruker AI-verktøy i flere deler av innholdsproduksjonen. Hver artikkel har en åpenhetsblokk nederst som forteller hvordan AI er brukt i akkurat den teksten, fra fullt menneskeskrevet til helautomatisert. Redaktøransvar ligger hos Thomas A. Kleppestø som navngitt fysisk person, etter unntaket i EU AI Act Artikkel 50(4) for journalistisk og akademisk arbeid med menneskelig redaktøransvar. Personopplysningsloven §3 dekker det samme prinsippet på den norske siden.
Hvis du har spørsmål om en bestemt artikkel, bruk kontaktsiden.
Article 4, AI literacy
ftrcrp.org is a deployer of AI systems within the meaning of the EU AI Act. Article 4 requires deployers to ensure a sufficient level of AI literacy among staff and contractors operating those systems. The author of the site is the operator. AI literacy is established through ongoing security and AI training at Noroff University College, formal coursework on AI risk, and direct operational experience building AI pipelines for the work that ends up on the site.
This page is also part of meeting Article 4 toward the reader, by being honest about what AI is, what it does on this site, and where the human in the loop lives.
Questions
If you want to know more about how a specific article was produced, or you spot something that looks wrong, the contact page is the way through.