Every rating shows its reasons, so you can disagree with an input.
A rating is likelihood times impact, built from recorded facts about the asset, and each fact carries its proof and what would remove it.
What would lower it
For each fact present, what the rating becomes without it. In the example, putting a person in the loop, or pinning the version and adding endpoint protection, lowers the rating.
Gaps are listed, not guessed
An unrecorded fact counts as nothing, so missing data never inflates a rating. Each gap shows what the rating would be in the worst case.
A visible second opinion
Where the facts support a different rating than the detection engine's label, the page says so in a sentence. In the example, the engine said critical and the facts support high.
Your decision is recorded
Adopting the calculated rating records an override with an owner and a reason. The engine's label is never silently replaced, and the override expires.
The same panel everywhere
The rating appears on the asset page and in every component, tool-server and finding panel. The risk register uses the same model for a risk that names one asset.
Step by step.
Facts are recorded
A set of facts about the asset is recorded, some about likelihood and some about impact.
Each fact adds points
Points turn into a likelihood and an impact.
The rating shows its reasons
You see the proof for every fact and what would remove it.
Common questions.
Can we change the weights?
What if the facts disagree with the engine?
Where does the rating appear?
How are ratings banded?
- Nothing to configure. Engine weights can be tuned and simulated under Policy and scoring.
The scoring model, its factors and its version are described in the technical pack, shared with your team under NDA.
Request the demo and technical packSee the AI on your fleet in about a week.
Run the sensor across your estate and receive an AI risk report and an AI Bill of Materials. The assessment is free and carries no commitment.