These définitions rely on public agencies and established research labs (EEAS, Hybrid CoE, EU DisinfoLab, Viginum, Council of Europe, Meta Threat Reports). We do not invent définitions. When a term is contested or evolving, we say so.
A pattern of mostly non-illegal behaviour that threatens or has the potential to negatively impact values, procédures and political processes. Conducted in an intentional and coordinated manner by state or non-state actors, including their proxies, inside and outside their own territory.
The practice of masking the sponsors of a message or organisation to make it appear as though it originates from grassroots participants. Commonly uses fake accounts, paid users or fabricated civic groups.
Simultaneous or tightly-timed sharing of the same content across accounts, outlets or networks in a way that is unlikely to be spontaneous. Not automatically malicious - coordination must be observed before intent is assumed.
Synthetic audio, image or vidéo generated or manipulated using artificial intelligence in a way that produces realistic but fabricated content attributed to real people.
The process of moving a narrative from a low-credibility origin (anonymous channel, state media, fake think tank) through successive intermediaries until it reaches mainstream outlets, stripped of its original source.
Information explicitly produced, funded or directed by a state actor to promote a political objective. Being state-funded does not automatically make a piece of content false; the distinction matters for attribution and transparency.
The original document, recording or first-hand account of an event. Disinformation analysis prioritises primary sources (official filings, raw footage, direct statements) over secondary reporting that can reintroduce bias or errors.