Trust at AuditAI
Open metrics on the AI we send into your audit. Public eval scorecard, transparency report, and the methodology behind how framework IDs are cross-checked.
Service issue detectedLatest eval pass-rate
-
no runs yet
Engagements delivered
5+
across 1 industries
Frameworks supported
7
NIST AI RMF, NIST AI 600-1 (GenAI Profile), ISO 42001, …
Eval scorecard
Every AI route is scored nightly. Workflow classification runs against curated ground-truth cases; the other 12 routes run structural shape evals. Pass rate, average score, and per-case breakdown.
Open scorecardMethodology
Which Claude models we use, how citations are validated, and how human auditors review every AI output.
Read methodology