BANKING
Customer onboarding, identity verification, enhanced due diligence
← FUNCTION-BASED WORKFLOWS
AI IN KYC · WORKFLOW EXPLORATION
How should AI actually change the way KYC work gets done?
AI is already being discussed across banking and financial services. Your KYC team may already be asking how it can use it.
The difficult part isn't finding an AI tool—or calling the answer KYC automation.
It's knowing what should change in the work.
KYC & CUSTOMER DUE DILIGENCE
A workflow carried differently across regulated organisations.
FINANCIAL SERVICES
Customer review, regulatory compliance, case handling
FINTECH & LENDING
Digital onboarding, verification, risk review
INSURANCE
Customer verification, policyholder due diligence, exceptions
REGULATED ORGANISATIONS
Evidence-based customer and case assessment
THE CURRENT REALITY
The request usually sounds simple.
“Can we use AI to make KYC faster?”
An employee wants to work faster. They ask for access to AI.
The organisation responds with governance. A policy is created. Access is granted. People are told what they can and cannot do. Then the organisation waits for adoption.
Sometimes it happens. Sometimes the policy becomes another layer of work. The people already comfortable with AI continue using it. Everyone else becomes more cautious.
Meanwhile, the subscription is being paid for, the promised productivity improvement hasn't appeared, and management starts asking a different question:
Why are we paying for this if people aren't using it?
The organisation solved the question of whether AI could be used.
It never solved the question of how the work should change.
That is where the question changes.
There are two ways to approach this.
CHOOSE A PATH TO EXPLORE
Select a path to see how the starting question changes.
That is where the question changes.
Instead of asking how someone can use AI, we ask what the person is actually doing.
LET'S LOOK AT THE WORK ITSELF.
Where does the difficulty sit in your workflow?
Select the stage that feels most difficult. The page will help you examine what may be sitting underneath it.
What does KYC actually carry?
SELECT A STAGE TO EXAMINE IT
SELECT A STAGE
Information → Interpretation → Verification → Exception → Decision
Choose the point where the work becomes difficult for your organisation.
The sequence is recognisable. Its operating context is not always the same.
IS THIS YOUR WORLD?
The workflow changes shape by industry.
WORKFLOW → INDUSTRY CONTEXT
The underlying workflow is similar. The pressures around it are not.
The sequence is simple. The work isn't.
THE CHARACTER OF THE WORK
The interesting part isn't the sequence.
It is where the character of the work changes.
Collecting a document is different from interpreting it.
Checking a field is different from deciding whether the evidence is sufficient.
Following a rule is different from deciding what to do when the rule doesn't quite fit.
And an exception is not simply another task.
An exception is where the organisation's judgement becomes visible.
That is often where the greatest opportunity—and the greatest risk—sits.
ONE USEFUL INSIGHT
Rules organise the routine. Judgement carries the exception.
WHERE THE WORK CHANGES CHARACTER
A rule can determine whether information is present or a condition is met.
The work changes when evidence is incomplete, conflicting or does not fit the expected path.
- What the evidence establishes
- Where the normal rule stops being sufficient
- Who owns the consequential judgement
Rule → Judgement.
ILLUSTRATIVE WORKFLOW EXAMPLE
A matched rule can still require judgement.
EXAMPLE · NOT A DIAGNOSIS
- A customer record triggers a routine ownership check.
- The submitted evidence meets the standard rule, but the ownership chain includes an unusual control arrangement.
- AI retrieves the relevant records and explains the apparent match without treating it as a clearance decision.
- A qualified reviewer resolves the exception and records the evidence, judgement and outcome.
Rule → Judgement.
So the examination follows the places where the work changes character.
How we would examine it
DIMENSIONS · NOT A CHECKLIST
These aren't a checklist to complete. They are dimensions through which we examine how the workflow actually operates.
These questions reveal the direction of the examination. The scoring, weighting, task libraries and intervention-selection rules remain part of the private engagement.
This is also where AI requires more care.
AI PARTICIPATION · CONTROL
And this is where AI gets interesting.
AI can help interpret information.
It can compare evidence.
It can surface inconsistencies.
It can summarise.
It can identify patterns.
It can also be wrong.
AI can hallucinate facts, misinterpret context, overlook an exception, or present an uncertain conclusion with more confidence than it deserves.
A mistake in summarising a document is not the same as a mistake that changes a customer's eligibility, triggers a regulatory action, or influences a meaningful transaction. The more consequential the decision, the stronger the control around AI should be.
This is why we don't treat AI participation as an on/off decision. We examine what the AI is doing, what evidence it is allowed to use, what it is allowed to conclude, when a human must intervene, and what needs to be recorded.
Evidence constraints
What information can the system rely upon?
Validation
How is an AI-generated interpretation checked?
Decision boundaries
What can AI recommend, and what can it actually decide?
Human control
Where is human judgement mandatory?
Exception handling
What happens when confidence is low or evidence conflicts?
Auditability
Can the organisation understand what happened afterwards?
We don't remove human judgement because AI has arrived. We make human judgement more valuable by determining where it belongs.
Governance shouldn't sit beside the workflow.
It should be designed into it.
So what would actually change if we redesigned the work?
What could change?
THE WORK · NOT ONLY THE TECHNOLOGY
An intervention might ultimately change much more than the technology involved.
INFORMATION · VERIFICATION
Compress the work
Remove unnecessary or repetitive tasks.
INTERPRETATION · DECISION
Improve the work
Give people better information at the point where judgement is required.
ALL FIVE STAGES
Reshape the work
Change which tasks belong together and where AI participates.
EXCEPTION · DECISION
Clarify the work
Make responsibilities and decision boundaries explicit.
VERIFICATION · EXCEPTION
Make governance usable
Connect policy to the actual work people perform.
WORKFLOW OUTCOME
Measure the work
Establish whether the intervention improves the workflow rather than simply increasing AI usage.
The objective isn't simply to make KYC faster. It is to make the KYC workflow more capable.
ONE QUESTION WORTH ASKING
Where does your KYC workflow stop being a rule and start becoming a judgement?
You may already know the answer.
You may not.
Either is a useful place to begin.
FIND THIS WORK IN CONTEXT
One workflow, connected to the organisation around it.
Return to the wider context without creating a separate version of this workflow.
Relevant functions
Risk, Compliance & Assurance →YOUR WORKFLOW
How does your KYC workflow operate today?
The page above examines KYC as a workflow. Now apply the same lens to your own organisation. Answer eight questions and get a structured read on its operating condition.
Take the KYC workflow assessment →You've looked at the workflow.
Now tell us what matters most.
YOUR WORKFLOW · YOUR PRIORITY
What do you want to change?
You know your workflow better than we do. Start by telling us where it hurts.
Tell us what matters most. We'll use it to understand where the workflow may be carrying unnecessary work, risk or judgement.
ALREADY KNOW YOU WANT TO TALK?
Bring the workflow as it operates today.
You do not need to prepare a presentation or redesign the process first. Bring the workflow as people experience it now—including the points where evidence, judgement, authority and AI participation become difficult.