Editorial Policy
This page explains how NetRiskScan's Knowledge Hub content gets made: how topics are chosen, what role AI plays, who reviews what, and how we handle mistakes.
How topics are chosen
Guides are built around the questions our own tools raise — what a DNS leak is, why a VPN might show a datacenter flag, what the NetRiskScan Index can and can't tell you — so that a reader who lands on a result page has somewhere to go for the explanation behind it.
The role of AI
AI may assist with drafting, research summarization, and producing initial structure for an article. It is a drafting aid, not an unsupervised publisher: content is expected to be reviewed and edited by a person before it counts as reviewed and is made publicly indexable.
A concrete case: our own review backlog
In August 2026 we identified a batch of guide articles that had been generated procedurally from a fixed template rather than written and reviewed individually, and that carried publish dates that had been artificially spread across a fabricated multi-week history rather than the one real date they were actually published. Both issues are exactly the kind of pattern that makes content look automated and low-value, independent of whether any individual sentence in it is accurate. We corrected the dates to the one real date on record and removed the entire batch from search indexing pending individual review — each article is being read, and either substantively rewritten, merged with an overlapping one, or retired, before it's allowed back into search results. We're describing this here rather than quietly fixing it, because an editorial policy that doesn't survive contact with its own mistakes isn't worth much.
What "reviewed" means
A reviewed article has been read start to finish by a person who checked it against our own tools' actual behavior (not just general knowledge about the topic), confirmed it doesn't overstate what a signal or score can prove, and confirmed it isn't substantially duplicating another article already on the site.
Corrections and updates
Articles carry published and updated dates that reflect real events, not a manufactured publishing cadence. We revise content when browser behavior, protocols, provider practices, or NetRiskScan's own tools materially change. To report an error, see Contact.
Sourcing and attribution
Where a guide describes a third-party service or technology, we write our own explanation rather than copying vendor marketing text; where we rely on a specific external claim, we aim to say so rather than presenting it as our own finding.
Independence
NetRiskScan does not accept undisclosed paid placement inside editorial guides. Our relationship with the third-party IP intelligence providers we query is that of a paying customer of their data — not a sponsorship, and it does not influence how we describe their limitations.