What will an examiner ask?

Where is this system in your inventory, what tier did you assign, and who signed it off? What counts as a model here, and what does your validation cover for the parts that are not? Show me the data lineage behind the retrieval set and the training or tuning data. What can the system do without a person, and where is that written down? How do you know it is still working: which evals run, how often, and what happened the last time one failed? What did you do about the vendor: due diligence, contract, exit plan, concentration? Walk me through one wrong output from production and what the customer, if any, saw. Who can switch it off, and has that been tested?

AML / KYC: AI Design Brief for Banks

Q: Which steps belong to a person, the model and a system?

1. Classify the alert: the model. Routing to a typology-specific path. 2. Gather evidence: a system. Tool calls to the case system, transactions and KYC store, in parallel. 3. Draft the investigation narrative: the model. Every statement cited to an evidence item. 4. Check the draft against the evidence: the model. An evaluator pass with explicit criteria; failures go back or to a person. 5. Dispose and file: a person. The investigator decides; the SAR is theirs.

AML is where supervisors have been most encouraging and where the volume argument is strongest: thousands of alerts, most of them false positives, each needing a file. The winning pattern is routing plus evaluator: classify the alert, gather the evidence from the case system and the transaction store in parallel, draft the narrative, then check it against the evidence before an investigator sees it. The investigator still disposes of the alert and signs the report; the model removes hours of assembly.

adjust the answers for your bank →every document for this use case →

The call

WorkflowRoutingEvaluator and optimizerTier 2: act with approvalautonomy level 2

AML / KYC: The model prepares and may act within limits; a person approves anything that reaches a customer or cannot be undone. Several model calls orchestrated by your code, in a sequence or a graph you designed. Material but recoverable. The model may prepare and, within limits, act, with a person approving anything that leaves the bank or touches a customer.

PatternWorkflow. Your code lays out the steps, the cost of error is real, and you need to observe every step.RoutingClassify the input first, then send it to a specialised path: a knowledge answer, a transaction on a live system, or a person.Evaluator and optimizerOne model produces, another checks against explicit criteria, and the loop repeats until the check passes or a person is called.KnowledgeRetrieval over the governed document set, with a citation on every answer and a refusal when nothing relevant is found. Tool calls to the system of record for anything live. Never retrieval for a balance, a case status or a limit. Documents to reason with, systems to check against: the workflow decides which question goes where before the model answers.Delivery routeThrough a cloud platform: two third parties in the chain and a concentration question; often the faster route through security review.

Decomposition

Which steps belong to a person, the model and a system?

#	Step	Owner	Note
1	Classify the alert	The model	Routing to a typology-specific path.
2	Gather evidence	A system	Tool calls to the case system, transactions and KYC store, in parallel.
3	Draft the investigation narrative	The model	Every statement cited to an evidence item.
4	Check the draft against the evidence	The model	An evaluator pass with explicit criteria; failures go back or to a person.
5	Dispose and file	A person	The investigator decides; the SAR is theirs.

Architecture

Which control layers carry the weight?

Each layer is described, with its controls and documents, on the control plane page.

Governance and accountability: core for this design.
Identity and entitlements: core for this design.
Action gateway: core for this design.
Data and knowledge: core for this design.
Models and vendors: core for this design.

Documents

Which rules and guidance does this design answer to?

Document	Authority	Why it applies here	Status
2018 Joint Statement on BSA/AML Innovation	FinCEN	Agencies encourage innovative approaches, including AI, in BSA/AML programmes.	In force
2026 AML/CFT Program Proposed Rule	FinCEN	'Effective use of artificial intelligence' counted in an institution's favour, if finalised.	Proposed
SR 26-2	Federal Reserve	Replaces the 2021 BSA/AML model-risk statement; monitoring models are models.	In force
FIN-2024-Alert004 (Deepfake Media)	FinCEN	Deepfake documents at onboarding are the threat KYC models now face.	In force
SR 23-4	Federal Reserve	US third-party risk management, including the model provider.	In force
SB 26-189	Colorado AI Act	Colorado: notice, explanation and human review for consequential automated decisions from Jan 1, 2027.	Final
BCBS Third-Party Risk Principles (Dec 2025)	Basel Committee	Nth-party supply chains and concentration on cloud providers.	In force
BCBS ICT Risk Management Report (June 2026)	Basel Committee	How supervisors look at ICT and cloud dependencies.	Final
OCC Bulletin 2026-13	OCC	For a national bank, the OCC's copy of the 2026 model-risk guidance.	In force
NIST AI RMF 1.0	NIST	The voluntary Govern, Map, Measure, Manage frame for everything model-risk guidance leaves out.	In force

Evals

How will you know it works, before and after launch?

A golden dataset of at least 150 real cases with expected outputs, including adversarial inputs: wrong documents, unusual formats, prompts that try to change the task.
A judge model scoring against a written rubric (accuracy, completeness, tone, citation present), calibrated against a human-scored sample every month. Threshold set from the human sample, not guessed.
Citation checks: every factual claim resolves to a passage in the governed set; unsupported claims below 2% of answers.
Trace evals per step, not only end to end: which step fails, how often, at what cost, so a prompt or model change can be judged step by step.
The same suite reruns on every prompt change, model version change and retrieval change; a regression blocks the release. That is what ongoing monitoring and outcomes analysis mean in model-risk terms.

Human gates

Where must a person be in the loop?

Per-action approval by a competent reviewer for anything customer-facing or irreversible; sampled review for the rest.
Validation proportionate to materiality, with monitoring for drift on inputs and outputs.
An escalation route to a person that the customer can reach in one step.
Monthly review of the evals and the exception log by the accountable owner.

Pitfalls

Letting the model close alerts on its own; the 2026 proposal rewards effective AI use, not unsupervised disposal.
Retrieval over stale customer data instead of a live tool call.
No record of what the model read, which makes the file indefensible.

Examiner questions

What will a validator or an examiner ask?

Where is this system in your inventory, what tier did you assign, and who signed it off?
What counts as a model here, and what does your validation cover for the parts that are not?
Show me the data lineage behind the retrieval set and the training or tuning data.
What can the system do without a person, and where is that written down?
How do you know it is still working: which evals run, how often, and what happened the last time one failed?
What did you do about the vendor: due diligence, contract, exit plan, concentration?
Walk me through one wrong output from production and what the customer, if any, saw.
Who can switch it off, and has that been tested?

For the board

What it does: triages alerts, assembles cases and drafts investigations for AML and KYC teams; the model prepares and may act within limits; a person approves anything that reaches a customer or cannot be undone.

Pattern: workflow (routing, evaluator and optimizer); tier 2: act with approval.

Rules it answers to: 4 documents across US, each linked in the brief.

How we know it works: a golden dataset, automated checks on every release, and human review at the level the tier demands.

What could go wrong and who answers: the accountable owner, the kill switch, the escalation route.

Peer evidence

Which of the 100 largest US banks have put this use case on the record?

Other briefs

AML / KYC: a design brief for banks.