Editorial Standards
These standards are operational commitments, not aspirations. They are public because accountability requires it — and because our readers are expert enough to evaluate whether we meet them.
01Coverage Scope
The Cyber Observer covers three interconnected domains: threat intelligence, digital policy and regulation, and geopolitics as it relates to cyber power. A story qualifies only if it touches at least two of the three domains.
Intelligence briefs cover developments significant enough to warrant professional attention in the next 48 hours. Analysis covers structural dynamics, actor profiles, and forward-looking assessments with a longer shelf life.
Not in scope: consumer security tips, vendor product announcements, conference coverage as events, anything requiring us to act as a marketing channel.
02Source Standards
Primary sources — government advisories, regulatory texts, court filings, official statements — are cited directly with links. We do not paraphrase in ways that change meaning.
Vendor intelligence is treated as a source, not gospel. We note commercial context and sensor coverage limitations. Vendor reports are not the sole basis for attribution claims.
Open source intelligence conclusions are labelled as such. We note when readers with access to classified data may hold a materially different picture.
03Confidence Labels
We use structured confidence labels derived from intelligence community practice. These appear in the body of analysis wherever relevant. Collapsing the distinction between analysis and established fact is an editorial failure.
04Independence & Conflicts
No advertising. No vendor sponsorship. No commercial partnerships that constrain editorial judgment. Revenue comes exclusively from reader subscriptions.
Any material conflict of interest — financial, professional, or personal — is disclosed in the body of the article. If a conflict cannot be adequately disclosed, the piece is not published under that author.
05Corrections
Errors of fact are corrected promptly, visibly, and without minimisation. Corrections appear at the top of the relevant article with the date, nature of the error, and the corrected information.
Errors of judgment — analytical conclusions that subsequent evidence contradicts — are addressed in follow-up analysis, not treated as corrections. We distinguish between being wrong about a fact and being wrong about an inference.