When Systems Recognise Abuse but Still Fail to See How Control Operates
THE DIRECTIVE™
Coercive Control, Digital Safeguarding, Financial Remedies and the Emerging Institutional Capability Gap
By Samantha Avril-Andreassen, LLB (Hons), LLM, LPC, FRSA
Founder, SAFECHAIN™
There is a significant shift taking place across safeguarding.
For years, much of the debate centred on recognition.
Do professionals recognise coercive control?
Do institutions understand economic abuse?
Do organisations recognise technology-facilitated abuse?
Do courts understand that domestic abuse can continue after separation?
Those questions remain important.
But they are no longer enough.
The emerging challenge is more difficult:
Can our systems actually detect how ordinary, legitimate systems are being converted into instruments of control?
Because coercive control does not always require an obviously abusive mechanism.
Sometimes the mechanism is perfectly legitimate.
A bank account.
A smartphone.
A location-sharing application.
A smart doorbell.
A joint mortgage.
A family court application.
Financial disclosure.
A company.
A payment platform.
A shared calendar.
A cloud account.
A legal process.
A procedural right.
A perfectly ordinary transaction.
The safeguarding question is therefore changing.
It is no longer simply:
“Do we recognise abuse?”
It is becoming:
“Can we recognise when legitimate infrastructure is being weaponised to produce it?”
That distinction represents one of the most important safeguarding challenges of our time.
The Convergence Is the Story
Developments across 17–23 August 2026 point towards something larger than a collection of separate safeguarding issues.
Digital safeguarding.
Coercive control.
Economic abuse.
Financial-remedy proceedings.
Artificial intelligence.
Evidence integrity.
Institutional capability.
These may appear to belong to different professional disciplines.
They do not.
They are increasingly converging around the same fundamental problem:
Modern coercive control operates across systems.
A survivor does not experience “digital abuse” on Monday, “economic abuse” on Tuesday and “legal proceedings” on Wednesday as unrelated events.
They experience a pattern of power.
Technology may provide surveillance.
Money may provide dependency.
Information may provide leverage.
Legal proceedings may provide continued access.
Institutional fragmentation may prevent anyone from seeing the whole.
That is precisely why safeguarding systems built around isolated incidents can struggle with coercive control.
The survivor experiences a pattern.
The institution often receives fragments.
The Instrument May Be Legitimate. The Use May Not Be.
This is where safeguarding needs to become considerably more sophisticated.
Take location technology.
Location sharing can protect children.
Help families coordinate.
Find lost devices.
Improve personal safety.
But the same technology can potentially be used to monitor another person's movements.
Take financial technology.
Joint banking can make family finances easier.
Payment applications make transactions faster.
Digital accounts improve accessibility.
But financial infrastructure can also become a mechanism through which spending is monitored, resources restricted or dependency reinforced.
Take smart-home technology.
Cameras, locks, thermostats, speakers and security systems are designed to make homes more convenient and secure.
But access and control over those systems can create very different experiences depending upon who possesses the credentials.
Take legal proceedings.
Courts exist to resolve disputes.
Disclosure exists to establish truth.
Applications exist to enable access to justice.
Procedural rights exist for good reason.
Yet legitimate processes can still be used strategically.
That gives us an important principle:
Safeguarding risk does not depend solely upon whether the instrument is legitimate. It depends upon how power is exercised through it.
The Smart Home Can Become a Controlled Environment
We increasingly describe homes as “smart”.
But safeguarding must ask:
Smart for whom?
Who controls the account?
Who controls the password?
Who receives the notifications?
Who sees the camera?
Who controls the heating?
Who controls the locks?
Who can view location history?
Who knows when someone enters or leaves?
Who can revoke access?
Who possesses administrative privileges?
The technological question may be:
Does the system work?
The safeguarding question is:
Who has power over whom through the system?
That is a fundamentally different design question.
A technology can function exactly as intended technically while producing a profoundly unsafe experience relationally.
Financial Technology Can Turn Everyday Transactions Into Data About Behaviour
The same principle applies to money.
Modern financial systems generate extraordinary quantities of information.
Where.
When.
How much.
To whom.
How frequently.
What subscription.
What journey.
What purchase.
What location.
For most people, this is convenience.
Inside coercive control, transaction data can potentially become information about another person's life.
Financial abuse therefore cannot be understood only as:
“Did somebody have access to money?”
We increasingly need to ask:
Who could see the transactions?
Who controlled the accounts?
Who received alerts?
Who had authority?
Could access be withdrawn?
Was expenditure questioned?
Did one person have independent financial privacy?
Were debts created?
Was income restricted?
Was employment interfered with?
The account balance tells us something.
The architecture surrounding the account may tell us considerably more.
Financial Remedy Is Now Part of This Conversation
This week's professional discussion around coercive control in financial-remedy proceedings is particularly significant.
On 20 August 2026, Resolution published professional analysis on recognising coercive control in financial-remedy cases.
The significance is not simply that economic abuse is being discussed within financial proceedings.
It is the movement towards recognising that coercive control may affect:
financial resources;
access to information;
evidence;
disclosure;
costs;
participation;
and ultimately the practical circumstances in which financial decisions are made.
That changes the analytical question.
Not simply:
“Was money withheld?”
But:
“How has coercion shaped the person's financial position and their ability to navigate the proceedings intended to resolve it?”
That is a major shift.
The Balance Sheet Does Not Tell You How the Balance Sheet Was Created
Financial-remedy proceedings necessarily deal with numbers.
Assets.
Liabilities.
Income.
Property.
Businesses.
Pensions.
Needs.
But numbers have histories.
A debt does not explain how the debt arose.
A depleted account does not explain why it became depleted.
A low income does not explain whether earning capacity was restricted.
A business valuation does not reveal whether both parties had equal access to the information necessary to test it.
An absence of savings does not explain whether somebody had meaningful control over household resources.
And financial disclosure does not become accurate merely because it has been placed before a court.
This is why evidence integrity becomes inseparable from safeguarding.
The balance sheet does not tell you how the balance sheet was created.
Control Can Migrate From the Home Into the Proceedings
This is another area in which institutional understanding needs to evolve.
We often imagine separation as the point at which domestic control ends.
But separation may simply remove one mechanism.
The relationship ends.
The financial connection remains.
Property remains.
Debt remains.
Businesses remain.
Children may remain a connection.
Disclosure becomes necessary.
Litigation begins.
Control can therefore potentially migrate.
From controlling money—
to controlling financial information.
From domestic uncertainty—
to procedural uncertainty.
From withholding information within the relationship—
to disputes about disclosure.
From economic dependency—
to unequal capacity to fund litigation.
From exhausting someone emotionally—
to exhausting them procedurally and financially.
This does not mean every difficult financial-remedy case involves coercive control.
It means that where coercive control is credibly raised, the proceedings themselves cannot automatically be treated as a completely separate environment.
Economic Control Can Become Procedural Disadvantage
The pathway can be stark:
Economic Control → Reduced Resources → Information Asymmetry → Reduced Exit Capacity™ → Impaired Participation → Procedural Disadvantage → Financial Outcome
And potentially:
Financial Outcome → Continuing Dependency → Post-Separation Control
This matters because two people can possess identical procedural rights while having profoundly different practical abilities to exercise them.
One may have resources.
Documents.
Professional advice.
Financial knowledge.
Institutional confidence.
The other may be attempting to reconstruct years of financial history while simultaneously managing the effects of prolonged abuse.
Formal equality cannot, by itself, resolve that difference.
Technology, Money and Legal Process Share One Critical Feature
At first glance, a banking application, smart-home device and financial-remedy proceeding have very little in common.
But from a safeguarding perspective, they share something fundamental.
They all create interfaces of power.
Someone can access.
Someone can authorise.
Someone can deny.
Someone can monitor.
Someone can disclose.
Someone can challenge.
Someone can delay.
Someone possesses information.
Someone may depend upon another person's action.
That is why safeguarding cannot remain focused solely on the instrument.
We must analyse the relationship of power created through the instrument.
The Institutional Fragmentation Problem
Here lies perhaps the greatest weakness.
Different institutions see different fragments.
The bank sees unusual transactions.
The technology company sees account access.
The police see an incident.
The court sees litigation.
The GP sees distress.
The housing provider sees instability.
The employer sees absence.
The safeguarding professional sees risk.
Each may be acting correctly within its own remit.
And yet nobody necessarily sees:
the pattern connecting them.
This is the Institutional Fragmentation Problem.
Coercive control crosses organisational boundaries more easily than safeguarding intelligence does.
That creates an enormous advantage for control.
Because a pattern divided across six institutions may look like six unrelated administrative events.
To the person experiencing it, it is one continuous reality.
The institution sees separate incidents. The survivor experiences one life.
Artificial Intelligence Raises the Stakes Again
AI introduces another layer.
Artificial intelligence can potentially improve safeguarding.
Pattern recognition.
Risk identification.
Document analysis.
Fraud detection.
Case triage.
Translation.
Accessibility.
Evidence organisation.
But AI can also increase the scale at which manipulation, impersonation, monitoring and information abuse may occur.
Deepfakes.
Synthetic communications.
Automated harassment.
Voice cloning.
Fabricated material.
Data aggregation.
Highly personalised manipulation.
The challenge therefore cannot simply be:
“How do we use AI?”
It must also be:
“How do we preserve human rights, evidence integrity, consent, privacy and safeguarding as AI becomes embedded within everyday systems?”
Technological capability without safeguarding governance is not neutral progress.
Evidence Integrity Is Becoming a Safeguarding Issue
This convergence makes evidence increasingly important.
Digital systems produce evidence.
Messages.
Metadata.
Bank records.
Location histories.
Device logs.
Images.
Audio.
Video.
Account access histories.
Court documents.
Corporate records.
But the existence of digital material does not automatically establish its meaning.
Who created it?
Who had access?
Can authenticity be established?
Is context missing?
Has material been altered?
Is the chronology reliable?
Could an account have been accessed by another person?
Was consent present?
Does the evidence show an isolated act—or part of a pattern?
As synthetic media becomes easier to create, those questions become even more important.
This is why Digital Evidence Integrity™ must sit at the centre of modern safeguarding.
Without trustworthy evidence, systems cannot reliably distinguish allegation, manipulation, context and fact.
The Architecture of Entrapment™ Is Becoming Cross-System
SAFECHAIN™'s Architecture of Entrapment™ becomes particularly important when these developments are viewed together.
Within an intimate relationship, the pathway may develop through:
Grooming → Commitment → Dependency → Reduced Exit Capacity™ → Domestic Manufactured Choice™ → Apparent Compliance → Entrapment
But modern systems can extend that architecture.
Digital Access → Surveillance
Financial Integration → Economic Dependency
Information Control → Evidential Disadvantage
Separation → Procedural Dependency
Litigation → Financial Depletion
Institutional Fragmentation → Failure to Recognise the Pattern
The control is no longer located in one place.
It is distributed.
That requires a distributed safeguarding response capable of reconnecting the evidence.
The Capability Gap
This brings us to the central issue.
We have spent years improving awareness.
But awareness and capability are not the same thing.
An organisation can have:
a domestic-abuse policy;
a safeguarding policy;
staff training;
a vulnerable-customer policy;
an equality statement;
a trauma-informed commitment;
and still fail someone.
Why?
Because the real test is operational.
Can staff identify patterns?
Can information travel safely?
Can concerns be escalated?
Can professionals challenge assumptions?
Can evidence be preserved?
Can vulnerability be recognised where the person appears capable?
Can systems distinguish genuine consent from coerced compliance?
Can technology identify misuse without creating additional surveillance?
Can organisations work across institutional boundaries?
Can the survivor participate meaningfully?
Can failures be audited?
Can somebody be held accountable when safeguarding repeatedly breaks down?
That is institutional capability.
Safeguarding Maturity Must Be Measurable
This is precisely why SAFECHAIN™ developed the Digital Safeguarding Maturity Model™.
Safeguarding cannot remain a statement of intention.
Organisations need to know:
Where are we now?
What can our systems actually detect?
Where are our blind spots?
What happens when technology is misused?
What happens when information crosses departments?
Who owns the risk?
How quickly can concerns escalate?
Are survivors involved in system design?
Can we evidence that safeguards work?
What happens when they fail?
A mature safeguarding organisation does not simply possess policies.
It can demonstrate that its systems behave safely under pressure.
Safeguarding by Design Must Replace Safeguarding by Reaction
The larger lesson from this week's developments is that intervention after harm is no longer enough.
Products need safeguarding during design.
Financial services need safeguarding during design.
Court processes need safeguarding during design.
AI systems need safeguarding during design.
Housing systems need safeguarding during design.
Evidence systems need safeguarding during design.
This is Survivor Privacy by Design™.
Trust by Design™.
Consent Integrity™.
Trauma-Informed Digital Design™.
Participation by Design™.
And ultimately:
Safeguarding by Design.
Do not wait until a system has been weaponised against somebody to ask whether misuse was foreseeable.
Ask before deployment.
The Question Every Institution Should Be Asking
The question is no longer simply:
“Does our system work?”
It is:
“What happens when somebody deliberately uses our system to control another person?”
What happens when someone weaponises the joint account?
What happens when someone weaponises location sharing?
What happens when someone weaponises the smart home?
What happens when someone weaponises financial dependency?
What happens when someone weaponises procedural delay?
What happens when someone weaponises disclosure?
What happens when someone weaponises artificial intelligence?
What happens when somebody uses an institution exactly as designed—but for a purpose the designers never anticipated?
That is where modern safeguarding begins.
From Recognition to Institutional Capability
The next stage of safeguarding reform cannot simply be another awareness campaign.
Recognition matters.
Training matters.
Policies matter.
But they must translate into capability.
The movement now needs to be:
Recognition → Detection → Evidence → Interpretation → Intervention → Accountability → Prevention
And across that entire chain:
Participation. Privacy. Consent. Integrity.
Because recognising coercive control after years of harm is important.
Designing systems capable of interrupting it earlier is better.
Conclusion — The System Is Now Part of the Safeguarding Environment
This may be the most important conclusion from the developments of 17–23 August 2026.
Safeguarding no longer exists only in the relationship between a professional and a vulnerable person.
It exists in:
the software;
the account permissions;
the banking architecture;
the evidence systems;
the financial-remedy process;
the disclosure regime;
the smart home;
the algorithm;
the organisational handoff;
the court order;
and the institutional design itself.
That changes everything.
Because the question is no longer simply whether an individual professional recognised abuse.
We must ask whether the system made recognition possible.
Whether evidence could be connected.
Whether participation was meaningful.
Whether privacy was protected.
Whether control could migrate unnoticed from one environment into another.
Whether safeguards survived institutional boundaries.
And whether the organisation had the capability to respond when an apparently legitimate process became an instrument of coercion.
This is where safeguarding is moving.
From incidents to patterns.
From individual products to ecosystems.
From awareness to capability.
From retrospective intervention to preventative design.
From asking:
“Was this abusive?”
towards asking:
“How was power exercised through the systems surrounding this person?”
That is a much harder question.
But it is the question modern safeguarding must learn to answer.
Because coercive control evolves.
Technology evolves.
Financial systems evolve.
Legal processes evolve.
The methods of exploitation evolve.
Safeguarding must evolve with them.
THE DIRECTIVE™ | SAFECHAIN™
“The instrument may be legitimate. The use may not be.”
“The institution sees separate incidents. The survivor experiences one life.”
“Information asymmetry can become power asymmetry.”
“A mature safeguarding organisation does not simply possess policies. It can demonstrate that its systems behave safely under pressure.”
“The question is no longer simply whether our systems work. It is what happens when somebody deliberately uses them to control another person.”
“Safeguarding must evolve as quickly as the systems through which control can now be exercised.”
© 2026 Samantha Avril-Andreassen. All Rights Reserved.
Published by SAFECHAINN Ltd as part of The Directive™ public-interest safeguarding and systems-reform series.
SAFECHAIN™ concepts and frameworks referenced in this article include Reduced Exit Capacity™, Commitment Trap™, Domestic Manufactured Choice™, The Architecture of Entrapment™, Grooming-to-Control Continuum™, Digital Evidence Integrity™, Digital Safeguarding Maturity Model™, Survivor Privacy by Design™, Trust by Design™, Consent Integrity™, Trauma-Informed Digital Design™, and Participation by Design™.