Explainability
How Cyberdroid AI shows the reasoning and evidence behind its output.
What each finding shows
Cyberdroid AI Detection is available now to select organizations, and Cyberdroid AI Investigation is in early access. Both attach the evidence to each output and note any telemetry gap.
How it works
-
Telemetry health
Source availability and quality help the reviewer judge what the evidence supports. A missing feed remains a visible gap.
-
Threat intel source
Threat intel shows its source, its age and how far it is trusted.
-
Verdict review
A separate review step tests the evidence behind each verdict.
Limits
Scores
A score helps prioritize review. The assessment still needs supporting events and a reasoned interpretation.
Incomplete cases
When the available evidence does not support a conclusion, the case remains incomplete and identifies the unanswered questions.
Language
Preserve source identifiers and technical meaning in the record, so reviewers can trace an assessment back to the evidence.
Standards
Naming a clause is not a claim of certification.
NIST
AI RMF MAP 2.2 asks for documented knowledge limits.
EU
Article 14 of the EU AI Act asks that people can interpret output and stay aware of automation bias.
