Online Dating Scams: How Romance Fraud Turns Emotional Trust into Financial Loss

Online dating scams are often described as situations in which a criminal creates a false profile, develops a relationship and eventually asks the victim for money. That description captures the visible outcome but understates the sophistication of the operating model.

Modern romance fraud can combine stolen identities, artificial intelligence, social-media reconnaissance, synthetic images, scripted conversations, fraudulent investment platforms, cryptocurrency payments and money-mule networks. The relationship may be cultivated for weeks or months before any financial request is made, allowing the offender to understand the victim’s circumstances, test their responses and create emotional dependency.

The victim may then be persuaded not only to transfer money, but also to disclose personal information, borrow funds, liquidate investments, receive third-party payments or participate in activity they do not recognise as criminal.

For financial crime teams, romance fraud is therefore not solely a consumer-awareness issue. It sits at the intersection of authorised push payment fraud, investment fraud, identity abuse, customer vulnerability, money laundering, human trafficking and cyber-enabled organised crime.

Listen the podcast

Watch the video

Why online dating scams matter now

Romance fraud continues to grow despite sustained public awareness campaigns. UK Finance recorded £39.2 million in romance-scam losses during 2025, the highest annual total in its data series. The number of reported cases increased by 22% to 6,335, marking the fifth consecutive annual rise.

The payment pattern is particularly important. Victims made an average of 9.7 payments per case—more than for any other authorised push payment scam category. This reflects the gradual nature of the manipulation. Criminals commonly begin with smaller requests, establish a history of successful payments and increase the amounts as the relationship deepens.

The point of origin is also changing. Dating applications remain important, but criminals increasingly approach potential victims through social media, gaming platforms, professional networks and private messaging services. Federal Trade Commission data indicate that nearly 60% of people who reported losing money to a romance scam during 2025 said the contact began through social media.

The underlying threat is broader than individual fraudsters operating fabricated profiles. INTERPOL has documented the global expansion of industrial-scale scam centres, some of which rely on people trafficked through false employment opportunities and forced to conduct romance, investment and other social-engineering scams. The criminal operation may therefore exploit two groups simultaneously: the person being financially manipulated and the person compelled to conduct the conversations.

How criminal actors manufacture a relationship

The attack begins with victim selection. Criminals review dating profiles, social-media posts, relationship status, employment, interests and major life events. Bereavement, divorce, retirement, relocation and social isolation can provide openings for a carefully tailored approach.

The false identity is designed to match the victim’s expectations. The supposed partner may present themselves as a military officer, engineer, medical professional, entrepreneur, aid worker or contractor based abroad. These occupations provide a credible explanation for distance, irregular communication and repeated difficulty meeting in person.

Photographs may be stolen from genuine users, altered through image-editing tools or created synthetically. Criminals may maintain several profiles across different platforms to reinforce the identity. A search of the supposed name may produce professional pages, social-media accounts and photographs that appear mutually consistent because the entire digital footprint was constructed for the fraud.

The conversation then moves away from the original platform. The offender may claim that private messaging is more convenient or secure. This reduces the dating service’s visibility and allows the criminal to preserve access if the original profile is reported or removed.

Communication becomes frequent and emotionally intense. The offender mirrors the victim’s values, shares fabricated personal experiences and introduces long-term commitments unusually quickly. Discussions of marriage, travel or a shared home can create the perception that the relationship has advanced beyond an ordinary online connection.

This stage is not incidental. It is the principal control mechanism. The criminal needs the victim to interpret future inconsistencies as temporary problems within a genuine relationship rather than as evidence of fraud.

From emotional trust to financial escalation

The first request is often designed to appear limited, understandable and temporary. The supposed partner may need help with a medical bill, travel document, customs fee, business emergency or flight ticket.

The amount may be low enough to avoid bank intervention and emotionally framed so that refusal appears uncaring. Once the victim sends money, the criminal has established both financial compliance and psychological commitment.

Subsequent requests become larger or more complex. A new obstacle prevents the promised meeting, a previous payment fails to resolve the emergency, or an unexpected fee is required to release money supposedly belonging to the offender.

The stories are adaptable. A criminal may claim to have received an inheritance, completed a profitable contract or acquired valuable goods that cannot be accessed until taxes or legal charges are paid. The victim is encouraged to view each payment as the final step before the relationship can become real.

This creates a sunk-cost dynamic. The more the victim has transferred, the more difficult it becomes to accept that the relationship and the expected repayment were fabricated. Admitting the fraud may mean confronting not only financial loss but also the collapse of an emotionally significant relationship.

Criminals understand this pressure and may actively isolate victims from family, friends and bank staff. They may claim that outsiders are jealous, prejudiced or attempting to interfere. The victim can be instructed to conceal the relationship or provide a false explanation when questioned about a payment.

Romance baiting and fraudulent investments

One of the most damaging developments is the convergence of romance and investment fraud. The criminal does not initially ask for emergency assistance. Instead, they present themselves as financially successful and gradually introduce the victim to cryptocurrency, foreign exchange or another investment opportunity.

The term “pig butchering” has frequently been used for this model, but INTERPOL and other organisations increasingly favour “romance baiting” because the older language can stigmatise victims and reduce willingness to report.

The offender may casually discuss their own investment performance before offering to help the victim. Early payments are directed towards a professional-looking website or mobile application. Fabricated dashboards display profits, and the victim may be allowed to withdraw a small amount to demonstrate that the platform is genuine.

The apparent success encourages larger deposits. The criminal may recommend borrowing, transferring pension savings or liquidating legitimate investments to take advantage of a time-limited opportunity.

When the victim attempts to withdraw substantial funds, the platform demands additional payments for tax, insurance, verification or account release. Each charge is presented as necessary to unlock a much larger balance. In reality, the displayed profits never existed.

This hybrid model is difficult for banks to identify because the customer may describe the transaction as an investment rather than support for a partner. The beneficiary may also be a company, payment processor, cryptoasset service or mule account with no obvious connection to the online relationship.

Money mules, account access and criminal facilitation

Romance scams do not always end with the victim’s own money. Once trust is established, the offender may ask the victim to receive funds on their behalf.

The explanation may involve difficulty accessing a bank account, international business activity, an inheritance or payments from clients. The victim is instructed to accept incoming transfers and forward them to other accounts, purchase cryptocurrency or withdraw cash.

Those funds may belong to victims of other frauds. The dating-scam victim has now become an unwitting money mule, creating distance between the predicate offence and the criminal organisers.

The FBI specifically warns that an online companion requesting access to a bank account or asking someone to receive and forward money is likely using the account to facilitate other theft and fraud. Continuing after warnings from the bank can also change how the individual’s conduct is assessed, particularly where they begin ignoring obvious indicators or retaining a commission.

Criminals may also persuade victims to open companies, establish payment accounts, purchase mobile phones or receive parcels. The relationship becomes an infrastructure-recruitment mechanism through which genuine identities and accounts are placed under criminal influence.

Financial institutions must therefore consider both victimisation and facilitation. A customer displaying mule-account characteristics may still be under emotional manipulation, but safeguarding does not remove the need to restrict criminal fund movement and preserve evidence.

Sextortion and intimate-image abuse

A relationship can also be weaponised through intimate content. The offender persuades the victim to share photographs, videos or participate in a private video call, then threatens to send the material to family, employers or social contacts unless payment is made.

Synthetic or manipulated images can intensify the threat. Criminals may create sexualised material using photographs taken from the victim’s profile even where no genuine intimate content was shared.

The payment request may be repeated because compliance demonstrates that the victim can be pressured. The criminal may also sell the victim’s details to another group or later approach them through a supposed recovery service.

INTERPOL’s latest financial-fraud assessment notes that sextortion is increasingly being integrated into romance and investment scams, often supported by scripted communication and AI-generated content. This convergence shows why romance fraud cannot be addressed through payment controls alone.

Why romance fraud is difficult to detect

The defining activity takes place outside the financial institution. A bank sees transfers, cash withdrawals, card payments or cryptoasset purchases. It does not automatically see the relationship, messages, photographs, threats or fabricated investment platform that explain the customer’s behaviour.

The payments may also appear technically legitimate. The customer uses their normal device, completes authentication and confirms the beneficiary personally. Successful authentication establishes who performed the action, but not whether their understanding was created through deception.

Romance fraud frequently develops gradually. A low-value transfer to a new recipient may not be unusual. Risk emerges through the sequence: repeated payments, increasing values, international beneficiaries, borrowing followed by immediate transfer, new cryptocurrency activity or movement of savings that is inconsistent with previous behaviour.

Victims may give misleading explanations. The FCA found this in 42% of the romance-fraud cases it reviewed. Some customers claimed to be purchasing property, supporting family members or paying for legitimate services because the fraudster had coached them to defeat bank controls.

Inconsistency should therefore prompt careful investigation rather than automatic accusations. Genuine victims may appear defensive, irrational or evasive because they remain emotionally committed to the relationship.

What a resilient control stack looks like

The first layer is platform-level identity and behaviour monitoring. Dating and social-media services should assess profile creation, device linkage, image reuse, rapid messaging patterns, requests to move off-platform and networks of accounts sharing infrastructure.

Verification badges should not be treated as guarantees. Criminals may compromise verified accounts, recruit genuine individuals or pass an identity check before another person controls the conversation.

The second layer is financial behavioural detection. Institutions should monitor repeated transfers to new beneficiaries, sudden international payments, unusual use of money-transfer services, first-time cryptocurrency purchases and borrowing followed by rapid dispersal.

The third layer is relationship-aware intervention. Generic warnings are unlikely to persuade someone who believes they are supporting a trusted partner. Staff should ask how the customer met the recipient, whether they have met safely in person, why the funds are required and whether repayment or investment returns can be independently verified.

The fourth layer is receiving-account intelligence. A beneficiary receiving funds from several unrelated customers, followed by rapid pass-through activity, may be part of a wider scam network. Banks should examine linked devices, contact details, companies, downstream recipients and previous fraud reports.

The fifth layer is cross-institution information sharing. Victims may move payments to another bank after one institution intervenes or open a new account at the fraudster’s direction. Known-victim markers and scam-beneficiary intelligence can help firms identify activity that would otherwise appear to be a new customer relationship.

The sixth layer is safeguarding. Staff require training to recognise emotional dependency, coercion and isolation. Effective engagement should be persistent but non-accusatory, allowing the customer to reconsider the relationship without feeling humiliated or blamed.

Finally, confirmed cases should generate preventive intelligence. Profiles, phone numbers, email addresses, domains, cryptoasset wallets, beneficiary accounts and payment narratives should be connected and distributed through lawful channels before they are reused against another victim.

What an effective victim response looks like

Once a scam is suspected, contact with the offender should stop and evidence should be preserved. Messages, profiles, transaction details, payment instructions, websites and wallet addresses may assist financial institutions, platforms and law enforcement.

The victim’s bank or payment provider should be contacted immediately. Rapid notification may allow transfers to be recalled, receiving accounts to be restricted or remaining funds to be protected.

The wider identity environment should also be reviewed. The criminal may possess identity documents, passwords, financial information, intimate images or access to email and social-media accounts. Credentials should be changed from a trusted device, active sessions revoked and recovery details checked.

Support should continue after the payment case closes. The FCA found that 15% of the cases it reviewed involved customers who had previously experienced fraud with the same firm. Recovery scammers commonly approach known victims with promises to recover lost money, creating a material risk of repeat victimisation.

What this means for financial crime leaders

Online dating scams should be governed as complex, long-duration financial-crime events rather than isolated payment disputes.

The operating model connects identity deception, social engineering, authorised payments, cryptoassets, mule accounts, sextortion and organised laundering. Some of the people conducting the conversations may themselves be victims of trafficking and forced criminality, adding further complexity to disruption and investigation.

Financial institutions need to connect customer behaviour, beneficiary intelligence, vulnerability and cross-channel evidence. Dating and social-media platforms must detect coordinated identities and suspicious interaction patterns before the relationship moves beyond their visibility.

The strongest interventions will not begin by telling the customer that the relationship is obviously false. They will help the individual test the story, identify contradictions and regain independent control over financial decisions.

Romance fraud succeeds because emotional trust is converted into financial authority. Institutions that recognise the gradual escalation, intervene with evidence and disrupt the accounts receiving the proceeds will be better positioned to protect both the victim’s finances and the wider payment system.