Why every threat score must be defensible
A number between 0 and 100 is easy to produce and easy to distrust. The difference between an opinion and an assessment is whether you can show your work.
Threat scoring is only useful if it survives scrutiny. A score that an analyst cannot explain — and that oversight cannot audit — is worse than no score at all, because it lends false confidence to a black box.
CLERINT Fusion computes a 0–100 threat score from seven weighted, transparent factors: criminal history, network position, activity recency, communication patterns, financial indicators, geographic risk and intelligence flags. The factors, the weights and the evidence trail behind each are visible to the analyst and to oversight.
That transparency is what turns a number into an assessment. When a person of interest crosses a threshold, the analyst can see exactly which factors drove it, and trace each back to the source record. The score becomes a starting point for judgement — not a substitute for it.
It also changes the conversation with oversight. A defensible score is one a supervisor, a lawyer or a court can interrogate: which factors, what weight, whose evidence. Opacity forecloses that conversation; transparency invites it. In work where the stakes include someone's liberty, that difference is not academic.
Time saved is time spent on judgement — the work only people can do. Automation should clear the noise so the analyst can reason about what matters, with the evidence in front of them, and stand behind the answer.