credit · EU AI Act

TransUnion

Ravna tracks 4 AI features shipped by TransUnion. Classifications are independent interpretations of public documentation, dated and sourced.

AI featureRiskDefault stateConfidenceAssessed
TruIQ Analytics Studio — ML Model Development
Cloud data-science environment (on Amazon SageMaker Studio) where lenders build and train machine-learning models on TransUnion credit data to develop credit-risk scoring for underwriting.
redOn when feature licensedlikely
TruVision Alternative Bank Risk Score
Credit-risk score that evaluates a consumer's banking/checking activity, including thin-file and subprime borrowers, to predict likelihood of default for lending decisions.
redOn when feature licensedlikely
TruValidate Device Risk — ML Fraud Decisioning
Adaptive ML models that learn from device signals and global fraud-consortium data to produce real-time risk scores and fraud decisions at credit-application touchpoints.
redOn when feature licensedlikely
AI Analytics Orchestrator Agent (TruIQ / OneTru)
Agentic AI (on OneTru with Google Gemini) that turns natural-language questions into production credit-analytics workflows and enables autonomous decision-making, adaptive learning and predictive scoring.
redOn when feature licensedlikely
See the full heatmapCheck your Article 50 exposure

Classifications are Ravna’s independent interpretation of public vendor documentation — an opinion, not a statement of fact, not legal advice, and not a determination that TransUnion or any deployer is non-compliant. TransUnion and related marks are the property of their respective owners, used for identification and commentary only; Ravna is not affiliated with or endorsed by TransUnion. Represent TransUnion? Request a correction — we respond within 10 days. See the full disclaimer.