What is this course about?
AML programs consume enormous analyst time on alerts that are overwhelmingly false positives. This course covers where AI reduces that burden — screening, triage, narrative drafting — while keeping the decision trail an examiner will scrutinize, and is explicit that closing an alert remains a documented human judgment.
Who is this course for?
- BSA officers and AML compliance staff
- Financial crimes investigators
- Transaction monitoring analysts
- Internal audit covering BSA/AML
What do I need before starting?
- BSA/AML program experience
- Familiarity with your transaction monitoring system
What will I be able to do afterwards?
- Identify where AI reduces cost in a BSA/AML program
- Improve name and sanctions screening precision without losing recall
- Triage alerts while preserving an examinable decision trail
- Draft SAR narratives with mandatory human review and sign-off
- Explain to an examiner why an alert was closed
What does each module cover?
Where does the cost actually sit in a BSA program?
40 minMapping program cost to find where AI changes the economics rather than the appearance.
Objectives
- Map program cost across activities
- Identify where false positives dominate
- Set realistic reduction targets
Topics
Activity. Map your program's cost and quantify the false positive burden.
How do you improve screening precision?
55 minName and sanctions screening where recall cannot fall and precision must rise.
Objectives
- Improve match precision without reducing recall
- Handle transliteration and name variants
- Validate changes against known true positives
Topics
Activity. Tune screening and validate that recall against known true positives did not fall.
How do you triage alerts defensibly?
55 minReducing analyst load while every closure remains a documented human decision.
Objectives
- Prioritize alerts by likely disposition
- Keep closure decisions human and documented
- Prevent triage becoming automated closure
Topics
Activity. Triage a real alert set and verify every closure carries a human decision record.
How do you tune transaction monitoring?
50 minThreshold and scenario tuning supported by AI analysis, validated properly.
Objectives
- Analyze scenario performance
- Support tuning decisions with evidence
- Validate tuning against known typologies
Topics
Activity. Analyze one scenario's performance and propose a validated tuning change.
How do you draft a SAR narrative?
50 minNarrative drafting with mandatory human review, sign-off, and accountability.
Objectives
- Draft narratives covering the required elements
- Enforce human review and sign-off
- Maintain filing quality and timeliness
Topics
Activity. Draft SAR narratives and route each through the review and sign-off gate.
How do you research beneficial ownership?
45 minOwnership research across unstructured sources with documented provenance.
Objectives
- Research beneficial ownership across sources
- Document provenance for every finding
- Handle deliberately obscured structures
Topics
Activity. Research a complex ownership structure with full source documentation.
How do you explain a closure to an examiner?
45 minExplainability where the examiner asks why this alert was closed and that one escalated.
Objectives
- Produce explanations for individual decisions
- Maintain consistency across analysts
- Prepare for examination questioning
Topics
Activity. Explain ten closures to a colleague playing an examiner.
Building the triage workflow
50 minThe lab module: triage with a complete decision audit trail.
Objectives
- Build the triage workflow
- Verify the audit trail completeness
- Measure analyst time saved
Topics
Activity. Build the workflow and verify every decision is reconstructable from the trail.
What is the capstone project?
AML triage workflow with a complete decision audit trail
Build an AML workflow with tuned screening validated for recall, defensible alert triage with human closure, monitoring tuning evidence, SAR narrative drafting with sign-off, and a decision audit trail that lets an examiner reconstruct any closure.
Deliverable: A working triage workflow with examination-ready audit trail and measured time savings.
How are learners assessed?
- Screening recall against known true positives must not fall
- Every alert closure must carry a reconstructable human decision record
- Examiner role-play conducted on ten real closures
What ships with the course?
Facilitator guide
Session-by-session running order, discussion prompts, and the questions that reliably derail a room.
Learner workbook
Exercises, checklists, and the templates each module's activity produces.
Hands-on lab environment
A sandboxed ibl.ai deployment so exercises run against real agents, not screenshots.
Assessment bank
Scenario questions and rubric criteria mapped to each stated learning outcome.
Source bibliography
Every primary regulation and standard cited on this page, linked and dated.
Which AI agents does this course use?
The hands-on modules run against agents already deployable on the ibl.ai platform for financial services.
Where does the course material come from?
Every module is grounded in primary sources — the regulation, standard, or research itself, not a summary of it. Each was resolved at authoring time.
- FinCEN
U.S. Treasury
SAR requirements, beneficial ownership rules, and AML guidance.
- FFIEC
Federal Financial Institutions Examination Council
BSA/AML examination expectations governing the audit trail design.
- Artificial Intelligence
FINRA
Regulatory guidance on AI use in financial compliance.
- AI Risk Management Framework
NIST
Validation and monitoring structure for the screening changes.
Delivery notes
Binding guidance for anyone preparing and delivering this course.
- Module 3's boundary is the regulatory crux. AI may prioritize alerts; it may not close them. Make this structural in the workflow, because an automated closure discovered at examination is a consent-order-level finding.
- Module 2's recall validation is non-negotiable. Precision improvements that quietly reduce recall are the single most dangerous change an AML program can make.
- Use synthetic alert data with known dispositions. Real alert data cannot leave the compliance function and known ground truth is required to measure anything.
- Module 7's examiner role-play should be run by someone who has been examined. Examination questioning has a specific character that a generic role-play misses.
- Have BSA counsel review the course. AML requirements are prescriptive and an error here creates regulatory exposure.
Why run AI training on a platform you own?
You own the course, not a licence to it
Course content, learner data, and the platform run inside your perimeter — you own all the code and the data.
Model-agnostic delivery
Run the course's AI components on any LLM — Claude, GPT, Llama, Gemini, Command — and switch anytime.
No per-seat training licences
Usage-based or self-hosted, so cost tracks actual use rather than headcount.
Deploy anywhere
Cloud, private VPC, on-premise, or fully air-gapped — including for cohorts that cannot use public AI tools.
Frequently asked questions
What does the KYC and AML with AI: Screening, Alerts, and SAR Support course cover?
AML programs consume enormous analyst time on alerts that are overwhelmingly false positives. This course covers where AI reduces that burden — screening, triage, narrative drafting — while keeping the decision trail an examiner will scrutinize, and is explicit that closing an alert remains a documented human judgment. It runs 6 hours across 8 modules across 8 modules, at advanced level, and closes with a capstone: AML triage workflow with a complete decision audit trail.
Who should take KYC and AML with AI: Screening, Alerts, and SAR Support?
It is written for BSA officers and AML compliance staff, Financial crimes investigators, Transaction monitoring analysts, Internal audit covering BSA/AML. Prerequisites: BSA/AML program experience; Familiarity with your transaction monitoring system.
Can we run this course on our own infrastructure?
Yes. ibl.ai is model-agnostic and deploy-anywhere — cloud, private VPC, on-premise, or fully air-gapped — and you own all the code and the data. Cohort data, submissions, and any material learners upload stay inside your perimeter, which matters for financial services teams that cannot send work to a public AI tool.
How do we get access to KYC and AML with AI: Screening, Alerts, and SAR Support?
Request access and we will set it up for your cohort — hosted by ibl.ai, or running against your own deployment. Tell us the group size and timing you need, and whether it should run inside your own perimeter.
How much does AI training for financial services cost on ibl.ai?
There is no per-seat pricing — you pay for usage or self-host and pay only for the infrastructure, so a 5,000-person rollout does not cost 5,000 licences. 1.6M+ users across 400+ organizations run the platform this way, including NVIDIA, MIT, and Syracuse University.