Beyond the Buzzword: The Dos and Don’ts of AI in AML
- Sep 18, 2025
- 3 min read
Updated: Aug 4
Last updated: July 27, 2026 | By: Ryan McInerny and Haig Douzdjian
Key Summary
Responsible AI in AML should assist experienced professionals, preserve explainable reasoning, protect sensitive data, and fit the institution's controls. RiskScout treats AI as optional support, not an autonomous replacement for AML judgment. Its five AI Scouts assist with investigations, behavioral analysis, regulatory filings, onboarding, and due diligence.
AI is everywhere - but in BSA/AML, not all AI is created equal. From flashy promises that don’t deliver to real-world tools that save hours of busywork, knowing the difference is critical for BSA Officers and back office teams.
You’ll walk away knowing the right questions to ask vendors, how to spot red flags, and where AI can genuinely support your program without creating new risks. Plus, we’ll share stories of how community banks and credit unions are already putting AI to work to save time and stay exam-ready.
Whether you’re curious, cautious, or already leveraging an AI-enabled tool, this session will give you the clarity you need to make smart decisions for your institution.
Watch this session to learn:
Data & Privacy Concerns: Understand where sensitive information could be at risk and how to safeguard it.
What to Ask Vendors: A guide for questions to uncover how AI models handle your data, privacy, and controls.
What are the Types of AI: Learn the difference between predictive and generational AI.
How RiskScout's Five AI Scouts Assist AML Teams
RiskScout uses five optional Scouts to reduce repetitive work while keeping professionals responsible for decisions:
Investigative Scout: Evaluates alert configurations and parameters to support investigation work.
Behavior Scout: Analyzes transaction patterns and activity signals to support customer/member risk profiles.
Filing Scout: Helps prepare SAR and CTR filings through data population, templates, and deadline tracking.
Onboarding Scout: Collects risk data and documents through guided onboarding workflows.
Diligence Scout: Supports enhanced due diligence collection and examiner-ready reporting.
The useful question is not whether a platform uses hard rules or behavioral AI. Institutions need the right mix of controls for their risk, with a human in the loop reviewing the results and making material decisions.
FinCEN supports responsible financial services innovation that can help institutions strengthen AML programs and improve BSA recordkeeping and reporting.
Frequently Asked Questions
Should AI make SAR filing decisions?
AI can assist with research, organization, and drafting, but the institution remains responsible for the filing decision. Material judgments should remain subject to qualified human review and approved procedures.
What is explainable AI in AML?
Explainable AI gives the institution enough information to understand, review, and document how an output was produced. The team should be able to connect the output to the underlying information and preserve a defensible record. If you can’t explain how you reached a decision to your examiner, then you need more insight into how your system works.
Is rules-based monitoring incompatible with AI?
No. Hard rules and behavioral analysis can both be valuable. RiskScout's position is that institutions need the right mix of controls, with people responsible for reviewing results and making decisions.
What should institutions ask AI vendors?
Institutions should ask what data the tool accesses, how outputs are explained, which decisions require human approval, how changes are tested, how input data is stored and secured, and whether features can be used selectively. An AI feature should strengthen a controlled process, not create a new black box.
Meet the Presenters:
Ryan McInerny, Director of Product Strategy
Ryan McInerny is the Director of Product Strategy at RiskScout, where he is passionate about creating solutions to enhance the day-to-day lives of financial institution employees. He brings a wealth of regulatory and technical experience to his role as a former National Bank Examiner with the Office of The Comptroller of the Currency (OCC). At the OCC, Ryan was a member of the OCC’s northeast district committees for compliance and commercial credit, where he helped to identify changing industry trends and risk conditions. Ryan brings a strong business sense with his banking expertise, having received his Masters Degree in Business Analytics from UVA's Darden School of Business.
Haig Douzdjian, AI Business Founder and Product Leader
Haig is a former AI founder and product leader, previously Head of Product at a Series B fintech AI startup. He drove product at LlamaIndex, a leading platform for AI agents, and now focuses on product and AI at The Browser Company.
Related: BSA/AML Platform
