AI Fair Lending & Consumer Risk for Financial Institutions
A Model Doesn’t Get a Pass on Fair Lending
Confidence for examiners. Clarity for the board. Assurance for every applicant.
What We Review
When an AI model influences a credit decision, it carries the same fair lending obligations as a human underwriter. When an AI-powered pricing engine determines the rate a customer receives, ECOA and UDAAP requirements apply, regardless of whether a person was involved in that specific decision.
The CFPB has made this explicit: algorithmic tools do not insulate institutions from fair lending liability. In some cases, they increase it, because biased inputs, flawed training data, or discriminatory correlations in a model can produce disparate outcomes at scale, across thousands of decisions, before anyone notices. AuditOne provides independent review of AI-driven credit and consumer-facing tools, evaluating compliance with fair lending requirements, adverse action notice obligations, and UDAAP standards.
AI-Assisted Credit Decisioning
- Fair lending compliance for AI-assisted underwriting and credit scoring
- Disparate impact analysis for protected class outcomes
- Model inputs and proxy variables: identification of prohibited-basis correlations
- Adverse action notice accuracy and specificity for AI-assisted decisions
- Documentation of the AI’s role in credit decisions for regulatory examination purposes
AI-Driven Pricing and Rate Setting
- ECOA compliance for AI-assisted rate, fee, or product pricing
- Disparate pricing analysis for protected classes
- Documentation and governance of pricing model inputs and outcomes
- Examination readiness for pricing-related fair lending reviews
Customer-Facing AI Tools
- UDAAP risk from AI-powered chatbots, virtual assistants, and customer communication tools
- Accuracy, fairness, and consistency of AI-generated customer communications
- AI tools used in collections or loss mitigation: consumer protection compliance
- Disclosure obligations for AI use in customer interactions
AI Model Governance for Consumer Risk
- Governance and oversight structure for consumer-facing AI models
- Testing and validation of AI models for fair lending outcomes
- Documentation supporting examiner review of AI-assisted consumer decisions
- Board and management reporting on consumer risk from AI programs
