AI and Fair Lending: UDAAP Risk in Algorithmic Decisioning
CFPB's UDAAP-as-discrimination gambit was vacated, but adverse action notice requirements still bite. Here's what AI lenders actually owe consumers in 2026.
Practical insights from the field — for financial services risk and compliance professionals.
CFPB's UDAAP-as-discrimination gambit was vacated, but adverse action notice requirements still bite. Here's what AI lenders actually owe consumers in 2026.
The method you choose for BIA data collection determines whether your RTOs reflect operational reality or wishful thinking. A practitioner's guide to surveys, interviews, and workshops — when each method works, where each fails, and how to combine them.
The DOJ's NFED consolidates healthcare fraud, tax, and market fraud units under one command. Here's what changes for compliance officers and risk managers.
A 47-page BIA full of RTOs and dependency tables won't get board buy-in for BCP investment. Here's how to translate BIA findings into an executive summary that drives decisions and satisfies FFIEC board reporting requirements.
When a vendor deploys AI in the service they provide you, your institution's model risk responsibility doesn't disappear. Here's the due diligence framework, questionnaire areas, and contract provisions you need before deploying a vendor's AI.
The CFPB fired its director, dropped 40+ enforcement actions, and withdrew nearly 70 guidance documents. Here's what actually changed — and what compliance obligations remain regardless.
These are the AI governance deficiencies regulators are actually finding in exams — incomplete model inventories, missing validation records, unmanaged vendor AI — and what to do about each one.
CFPB examiners don't just check whether you respond to complaints — they evaluate your entire complaint management infrastructure. Here's exactly what the exam manual requires and where programs typically fall short.
A step-by-step scoring methodology for identifying and tiering critical business functions in your BIA — with impact dimensions, scoring criteria, and real financial services examples.
Iran demanding crypto tolls at the Strait of Hormuz exposes massive sanctions risk for banks, shipping companies, and crypto exchanges. Here's what compliance teams need to do now.
How to derive RTO and RPO from real BIA data, set defensible numbers using the MTD hierarchy, and pass FFIEC examiner scrutiny on recovery objective methodology.
How to actually implement SR 11-7 for LLMs: model inventory, governance ownership, documentation standards, and validation scope for in-house and vendor AI.
When examiners evaluate your AI governance program, they're checking specific items. Here's the complete checklist — mapped to SR 11-7, OCC guidance, and the GAO's 2025 findings.
UDAAP exam findings still happen even when the CFPB is pulling back. Here's how to conduct a real UDAAP product risk assessment across all three prongs.
The CPPA issued over $2.3 million in fines across multiple enforcement actions in 2025. Here's exactly what they found, what the common violation patterns are, and what compliance teams need to fix before they're next.
A Las Vegas business owner just got 54 months in prison for filing 1,227 fraudulent ERC returns. If your company claimed the credit, here's what to do now.
FinCEN's 2026 proposed rule fundamentally reforms BSA AML/CFT program requirements—shifting from paperwork to effectiveness. Here's what changes and what to do now.
A practitioner's guide to running a business impact analysis that satisfies FFIEC examiners and ISO 22301 requirements—from scoping and data collection through RTO/RPO setting, dependency mapping, and board reporting.
SR 11-7 wasn't written for language models. Here's what model risk management teams actually need to test for LLMs and generative AI—hallucination evaluation, red-teaming, adversarial testing, and continuous drift monitoring.
92% of compliance professionals say their role has become harder, and 77% still track regulatory changes manually. Here's how to build a program that closes the gap between when rules take effect and when your policies reflect them.
AI systems fail differently than traditional IT. Here's how to build AI failure scenarios into your BCP, set recovery objectives for models, and satisfy emerging regulatory requirements on AI resilience.
Build an annual BCP testing calendar that satisfies FFIEC, ISO 22301, and NCUA examiners. Covers exercise types, scheduling by function criticality, tracking, and how to handle mid-year changes.
DOJ's $160M resolution with AssuredPartners and APSF for ACA enrollment fraud shows exactly what happens when broker compliance programs ignore third-party marketer risk.
DORA went live January 17, 2025. If your ICT vendor contracts haven't been updated, your Register of Information isn't filed, and you haven't mapped your concentration risk — here's exactly what needs to happen.
The OCC and FDIC finalized a rule prohibiting examiners from using 'reputation risk' in supervision, effective June 9, 2026. What bank compliance teams need to update now.
Flagstar said it had 'no evidence of unauthorized access.' The SEC disagreed. A practical breakdown of Form 8-K Item 1.05, how to build a defensible materiality determination process, and what early enforcement actions reveal about where companies are getting it wrong.
SEC filed just 456 enforcement actions in FY2025—fewest in two decades. Here's what compliance officers must know about the SEC's priorities shift.
Supply chain disruptions expose BCP gaps faster than almost any other event. COVID, the Suez blockage, and the chip shortage showed exactly where plans failed. Here's how to fix yours.
A practical after-action report template for BCP tabletop exercises. Covers the hot wash, finding format, corrective action tracking, and what FFIEC and ISO 22301 require.
A practical guide to the Business Continuity Maturity Model (BCMM). Learn the 5 maturity levels, how to self-assess your program, and how to prioritize improvements.
COVID forced every BCP program to confront a reality most hadn't planned for: running critical operations with 80% of your team working from kitchen tables. Five years later, hybrid work is the permanent baseline — and most BCPs still haven't caught up.
Most organizations run cyber incident response and BCP as separate programs — and that gap showed up badly in Change Healthcare, MOVEit, and dozens of other major ransomware events. Here's how to build a unified framework that actually works.
Most business continuity plans fail not because they're wrong, but because they've never been tested. Here are 10 tabletop exercise scenarios — with facilitator guidance — to stress-test your plan before a real incident does.
OCC and NCUA examiners don't just check if you have a BCP — they test whether it actually works. Here's what national banks and credit unions need to meet regulatory expectations and avoid exam findings.
Small businesses don't need a 200-page BCP. Here's a minimum viable framework covering OSHA, SBA, and HIPAA requirements — built for teams of 1-50.
SaaS companies have two BC obligations: protecting their own operations and surviving as the vendor when their customers' examiners come calling. Here's how to build a program that covers both.
SVB hadn't tested its discount window access since 2022. Learn how to design CFP tabletop exercises, stress scenarios, and simulation drills that satisfy OCC, FFIEC, and FINRA examiners.
OCC and FDIC finalized a joint rule on April 7, 2026 banning reputation risk as a basis for supervisory action. Here's what changes for compliance programs.
SVB's own stress tests predicted its failure eight months early. Management changed the assumptions instead of fixing the balance sheet. Here's what the 2023 bank failures reveal about CFP design, testing, and governance — and what regulators now expect from every institution.
Healthcare BCP isn't just about uptime — it's about patient safety. Here's what HIPAA, CMS, and The Joint Commission actually require, and how to build a continuity plan that survives an OCR audit.
Most contingency funding plans fail in execution, not design. The reason is almost always governance — unclear ownership, no board-level accountability, and triggers that nobody has authority to pull. Here's how to build a CFP governance structure regulators can actually examine.
Build a practical EWI framework for liquidity stress monitoring. Covers the indicators regulators expect, how to set escalation thresholds, and the governance structure to act on signals before they become crises.
FinCEN hit Canaccord Genuity with the largest-ever BSA penalty against a broker-dealer — $80M, coordinated with SEC and FINRA for $120M total. Here's what failed and what to fix now.
Not all contingent funding sources are created equal. Here's how to rank your backup liquidity options by reliability, cost, and access speed — before you actually need them.
ISO 22301:2019 mandates specific documented information across Clauses 4-10. Here's the complete list of required policies, procedures, and records — and what auditors actually check.
ISO 22301 gap analysis maps where your BCMS falls short clause by clause. Use this template and scoring guide to assess maturity and prioritize before your certification audit.
The OCC, FDIC, and Fed repeatedly flag the same CFP deficiencies across examination cycles. Here's exactly what they find, why SVB is the case study, and what remediation actually looks like.
Learn how to create a robust contingency funding plan (CFP) for your financial institution with our step-by-step framework, covering regulatory requirements and best practices for liquidity risk management.
ISO 22301 Clause 9.2 requires documented internal audits at planned intervals. Use this clause-by-clause checklist to find gaps before your external auditor does.
Build a defensible CFP liquidity stress test: three required scenarios, assumption documentation, survival horizon metrics, and lessons from SVB's $18B 30-day deficit.
The OCC rescinded 12 CFR 30 Appendix E, eliminating mandatory recovery planning for $100B+ banks effective May 1, 2026. Here's what that means for your program.
Three major global regulatory frameworks — BCBS 2021, UK PS6/21, and EU DORA — have redefined business continuity into something practitioners barely recognize. Here's what changed and what it means for your program.
Most BIAs skip IT dependency mapping entirely — or treat it as an afterthought. Here's how to build the technology layer that makes your BIA actually useful for recovery planning.
A practical BIA scoring methodology for financial services. Score impact across 4 dimensions, assign criticality tiers, and set defensible RTO targets.
A contingency funding plan that sits in a drawer fails the moment you need it. Here are the components OCC, Fed, and FDIC examiners actually check — and how to build a CFP that survives both a liquidity event and a regulatory exam.
CFP and BCP sound similar but serve completely different functions. Here's how to tell them apart, who owns each, and when both trigger at the same time.
FINRA's proposed Rule 4610 would impose liquidity risk management requirements on about 125 broker-dealers. Here's what the rule covers, the controversial 'rebuttable presumption' conditions, and what firms should be doing now.
Your BIA isn't a one-time project. Learn FFIEC and ISO 22301 requirements for BIA review frequency, which triggers mandate an update, and how to build a defensible maintenance schedule.
Learn how to create a robust contingency funding plan (CFP) with our step-by-step framework, covering regulatory guidance from FFIEC, OCC, FDIC, NCUA, and FRB.
Navigate AI risks and regulatory demands with a robust AI Impact Assessment (AIIA) guide and template. Essential for financial services.
Understand the differences and synergies between ISO 22301 (Business Continuity) and ISO 27001 (Information Security) for robust financial services resilience.
Vincent Camarda of A.G. Morgan Financial Advisors pleaded guilty to $160M investment fraud. Here's what went wrong and the compliance red flags every firm should watch for.
How state and federal regulators define consequential AI decisions — and what compliance teams must do before June 2026 to avoid enforcement.
How consumer data rights like deletion, opt-out, and access apply when businesses use AI for automated decisions — mapped across CCPA, Colorado, Virginia, and 17 other state laws.
A practitioner's guide to ai model validation techniques that satisfy OCC SR 11-7, FFIEC, and CFPB requirements for ML and LLM models in financial services.
Business impact analysis vs risk assessment — learn the key differences, when to use each, and how to integrate both into your BCM program.
Contingency funding plan requirements vary by regulator, but most banks and larger credit unions need a CFP now. Here’s what OCC, Fed, FDIC, NCUA, and FINRA expect.
How to build an AI training data governance program that covers data quality, consent, provenance tracking, and regulatory compliance for financial services.
Advanced Urology and Dr. Jitesh Patel will pay $14M to settle DOJ allegations of fraudulent billing and unnecessary procedures. Key compliance takeaways inside.
Using HireVue or AI screening in Illinois? HB 3773 just expanded your obligations. A federal court ruled BIPA and AIVICA apply simultaneously — here's what that means for your consent flow, notice requirements, and litigation exposure.
NYC Local Law 144 requires annual bias audits for AI hiring tools. Learn AEDT requirements, penalties, audit process, and what the Comptroller's enforcement review means for 2026.
Practical guide to detecting, protecting, and managing PII in LLM systems — covering GLBA, CCPA, de-identification, and vendor contract requirements.
Vincent Camarda of A.G. Morgan Financial Advisors pleads guilty to defrauding 400+ clients of $160M. What compliance professionals need to know about this investment adviser fraud case.
45 states have introduced 1,561 AI bills in 2026 — already surpassing 2024's full-year total. Colorado, Texas, and California are the three to watch. Every enacted state AI law, organized by what your compliance team actually needs to do.
A contingency funding plan (CFP) maps how your institution survives a liquidity crisis. Learn what a CFP is, who needs one, key components, and regulatory requirements.
Your AI model is making bad decisions in production. Do you have a documented shutdown plan? Most banks don't. Here's the kill switch framework examiners expect — decision criteria, authority matrix, fallback ops, and what the EU AI Act requires.
Practical guide to AI model monitoring and drift detection — types of drift, statistical tests, alert thresholds, and regulatory expectations for production ML systems.
How to build AI operational resilience for financial services — dependency mapping, vendor concentration risk, BCP planning, and tabletop exercises for AI failures.
How financial institutions can detect and defend against deepfake fraud — from voice cloning scams to KYC bypass attacks. Practical controls, FinCEN red flags, and detection tech.
The SEC settled with Chicago-based P/E Capital and CEO Eliseo Prisno for charging 200+ clients $2.4M in undisclosed fees — including hijacking client login credentials to approve charges.
NYC LL 144 already requires annual bias audits — and a 2025 Comptroller report found most companies aren't complying. Colorado SB 205 hits Feb 2026. Here's the full audit lifecycle: scoping, testing methods, remediation, and documentation.
A lender just paid $2.5M because their AI model discriminated against minority borrowers. The CFPB is watching yours next. Three testing layers you need running now — statistical analysis, LDA searches, and adverse action compliance.
OCC Bulletin 2011-12 now applies to AI and ML models. Here's what national bank examiners expect, common MRA findings, and how to build a defensible program.
OWASP ranks prompt injection #1 on their LLM Top 10. Gartner predicts 50%+ of successful AI attacks will use it through 2029. If your firm deploys any LLM, here are the four controls you need before the first incident — not after.
The SEC filed fraud charges against Jon Fullenkamp and Scott Sand for misappropriating millions through sham agreements and fraudulent preferred share issuances at two penny stock companies.
AI agents can now initiate payments autonomously. Your existing fraud controls were built for humans. Here's the threat model and control framework fintechs need now.
Step-by-step ai impact assessment guide template covering NIST AI RMF, EU AI Act, CFPB explainability, and SR 11-7. Risk tiers, timelines, owner assignments.
Build an AI model risk tiering methodology that accounts for autonomy, explainability, and data sensitivity. Includes a decision-tree framework and tier-specific oversight requirements.
A complete business impact analysis questionnaire template with 50 questions across 10 categories. Based on FFIEC, NIST SP 800-34, and ISO 22301 guidance.
ISO 22301 certification costs $15K-$60K+ depending on org size. Get realistic timelines, a month-by-month implementation roadmap, and tips to avoid common pitfalls.
ISO 22301 vs ISO 27001 compared side-by-side: scope, controls, certification process, and whether you need one, both, or neither.
The SEC secured final judgments against Titanium Capital LLC and founder Henry Abdo for a Ponzi scheme that defrauded 162 investors of $5.3 million. Here's what happened and what compliance teams should learn.
How to meet AI explainability requirements from the OCC, Fed, CFPB, and EU AI Act — with practical techniques for every model type.
AI impact assessments are now required under Colorado SB 205 and the EU AI Act. Learn who needs one, what to include, and how to build the process.
Map SR 11-7 and OCC 2011-12 documentation requirements to AI and ML models. Section-by-section template for model cards, training data provenance, and examiner-ready documentation.
Traditional model validation breaks down with AI. Learn the testing techniques — from adversarial red-teaming to drift detection — that actually work for ML and LLM models in financial services.
The SEC secured a final judgment against Kenneth Welsh, a former Wells Fargo advisor who misappropriated $2.86M+ from clients over five years through 137 fraudulent transactions.
Practical governance framework for agentic AI systems. Covers new risk categories, permission models, audit trails, and the human-on-the-loop debate for financial services.
SEC charged Tulsa college student Krish Kumar with misappropriating nearly $7M from two investment funds. Here's what compliance officers at investment advisers need to know.
Learn the essential AI bias testing methodologies for fair lending compliance—disparate impact analysis, counterfactual fairness, calibration testing, and more—before your next exam.
Learn how to prevent AI data leakage from LLMs in financial services. Covers the 5 leakage vectors, OWASP LLM top risks, NIST controls, and a 90-day implementation roadmap.
Your employees are already using ChatGPT — do you have a policy? Build an AI acceptable use policy with data classification rules, prohibited uses, and tool approval workflows.
The SEC's final consent judgment against Commonwealth Financial Network for undisclosed revenue-sharing conflicts offers a critical compliance lesson: fiduciary duty means disclosing who pays you, fully.
How to build an AI incident response plan that covers model failures, hallucinations, bias events, and drift — with severity tiers, escalation paths, and containment controls.
The first question your examiner will ask isn't about bias or governance — it's 'show me your model inventory.' Most banks can't. Here's the SR 11-7 fields examiners expect, how to find shadow AI, and the vendor tracking gap that gets flagged every time.
How to detect, measure, and mitigate LLM hallucination risk in financial services — with real controls, metrics, and a regulatory-ready framework.
The SEC charged four individuals with fabricating documents to defraud investors in a $284 million municipal bond offering for a Mesa, Arizona sports complex. Here's what went wrong and what compliance teams can learn.
How financial institutions should build business continuity programs that satisfy OCC, FDIC, and Fed operational resilience expectations — with real enforcement examples and implementation guidance.
Colorado SB 205 takes effect June 30, 2026. Learn who's covered, what counts as a high-risk AI system, required impact assessments, consumer notices, and your compliance checklist.
Your auditor wants ISO 22301 alignment? Here's exactly what each clause requires, how it maps to FFIEC, and whether certification is actually worth the cost.
The NIST AI RMF is in active revision as of 2026. Here's what's changing, what's staying stable, and what your AI risk program should do right now.
The SEC obtained a final judgment against Stuart Frost for extracting $14M in undisclosed incubator fees from VC fund investors. Key lessons for investment adviser compliance programs in 2026.
Shadow AI is spreading through financial services whether you know it or not. Here's how to detect it, assess the risk, and build a governance framework that actually works.
SR 11-7 was written for spreadsheet models, not LLMs. Here's how each pillar of the framework must adapt for AI/ML — and where traditional MRM breaks down completely.
Build a crisis communication plan that covers customers, regulators, employees, and partners — with pre-drafted templates, escalation timelines, and real-world lessons.
SEC closes $50M Ozy Media fraud case — revenue inflated 100%, a YouTube exec impersonated on an investor call. What compliance teams must learn from this textbook failure.
Step-by-step tabletop exercise template with facilitator guide, scenario injects, and 3 ready-to-use scenarios for business continuity testing.
Learn how to assess vendor business continuity plans, monitor third-party resilience, and meet FFIEC requirements for vendor BCP oversight.
Free business continuity plan template with the 8 sections every BCP needs. Step-by-step guide for financial services teams building or rebuilding their BCP.
Learn 5 types of business continuity testing, from checklist reviews to full-scale exercises, with practical guidance for financial institutions.
Business continuity vs disaster recovery explained — what each covers, where they overlap, and why treating DR as your whole continuity program is a regulatory red flag.
TSB Bank spent £330M learning what a proper BIA would have caught. Free template included. Step-by-step: identify critical processes, map dependencies, set RTO/RPO targets, and build the documentation your examiner expects.
Step-by-step disaster recovery plan template with recovery tiers, DR strategies, and testing schedules. Build a DRP aligned to your BIA and RTO/RPO targets.
A practical breakdown of the FFIEC BCM booklet requirements — governance, BIA, risk assessment, testing, and third-party resilience — with what examiners expect and common MRA triggers.
RTO vs RPO explained with practical guidance on setting recovery objectives, tiering critical functions, and avoiding the mistakes that turn outages into disasters.
The SEC's Legacy Cares case is a textbook municipal bond fraud—fabricated contracts, forged signatures, and a near-total investor wipeout. Here's what compliance practitioners need to know.
AI ethics and AI governance are not the same thing. Learn how ethics, governance, and model governance layer together — and why you need all three.
Build an effective AI governance committee with the right roles, a defensible charter, and a meeting cadence that actually works. Practical guide for financial services.
AI risk ownership is broken at most firms. Learn how to apply three lines of defense, assign accountability, and stop the 'everyone owns it' trap.
Scale AI governance across a large organization without killing innovation. Federated vs. centralized models, shadow AI controls, model inventories, and board reporting.
The SEC's Legacy Cares case shows how fabricated revenue documents collapsed a $284M municipal bond deal. Here's what compliance and risk teams need to know.
Build an AI compliance framework that survives regulatory exams. Model inventories, risk assessments, testing evidence, and documentation that proves you're compliant.
Build an AI governance policy that actually works. Covers scope, risk classification, approval workflows, monitoring, and exceptions — with section-by-section guidance.
Navigate the 2026 AI regulatory landscape — EU AI Act deadlines, state laws in Colorado, Illinois, and Texas, SEC enforcement priorities, and what compliance teams should do now.
Map the NIST AI Risk Management Framework against EU AI Act requirements. Build one AI governance program that satisfies both — with a practical crosswalk for financial services teams.
The SEC's enforcement action against Legacy Cares executives exposes how fabricated documents slipped past underwriters. Here's what compliance teams need to know.
Tactical AI governance best practices from financial services, healthcare, and insurance. Model inventories, tiered oversight, cross-functional committees, and documentation that survives exams.
Build an AI risk management framework that identifies, assesses, and mitigates real AI risks. Includes risk taxonomy, tiering model, and 90-day roadmap.
Map the NIST AI RMF's voluntary framework against the EU AI Act's mandatory requirements. Build one AI risk program that satisfies both.
Build a responsible AI framework that turns fairness, transparency, and accountability principles into operational controls. Includes bias testing, impact assessments, and 120-day roadmap.
The SEC obtained a final judgment against Bin Hao and Qidian LLC for a Ponzi scheme that targeted Chinese-American investors. Here's what compliance teams need to know about affinity fraud detection and controls.
Build an AI governance framework that actually works. 8 core components, maturity model, and 90-day implementation roadmap for risk practitioners.
Build a business continuity plan that survives regulatory exams. Step-by-step guide covering FFIEC requirements, BIA, RTO/RPO, testing, and common MRA findings.
A practical guide to implementing the NIST AI RMF across Govern, Map, Measure, and Manage — with actionable steps for financial services teams.
13 state AGs sued OneMain Financial for loan packing and junk fees on March 16, 2026. Here's what the case means for add-on product controls, fee disclosure, and state AG enforcement trends.
A practical guide to building an operational risk management framework — RCSA, KRIs, loss event tracking, and the ORM lifecycle for mid-size banks and fintechs.
The SEC obtained a final consent judgment ordering $106.5M in disgorgement against Ofer Abarbanel for orchestrating a mutual fund fraud scheme. Here's what happened and why it matters for fund compliance.
A new House bill would overhaul GLBA Title V and preempt state privacy laws for financial institutions. What practitioners need to know and do now.
Build a defensible incident response plan template for your fintech. Covers NIST phases, regulatory notification requirements, and what regulators actually check.
The FS AI RMF gives financial institutions 230 AI control objectives. A practical guide to prioritizing what matters and building your implementation roadmap.
New survey data shows most financial institutions can't identify vendor AI use. A practical vendor AI risk assessment guide with due diligence questions and implementation roadmap.
A comprehensive guide to identifying, assessing, and mitigating AI risks in regulated financial institutions—from model governance to third-party AI vendor oversight.