The Anatomy of Romance and Investment Fraud: Evolving Typologies Demand New Defences

Romance scams and investment fraud have converged into a single, devastating crime category that now accounts for the largest share of reported fraud losses in multiple jurisdictions. In the United States, the FBI's Internet Crime Complaint Center reported $4.6 billion in investment fraud losses in 2025 — a figure driven overwhelmingly by schemes that begin with social engineering through dating platforms, social media, and messaging applications. The United Kingdom, Australia, and Singapore report similar patterns.

The 'pig butchering' typology — in which victims are cultivated through fabricated romantic or social relationships before being directed to fraudulent investment platforms — has evolved significantly. Criminals now deploy AI-generated personas, deepfake video calls, and sophisticated fake trading platforms that convincingly simulate real market movements. The technical barrier to entry has fallen even as the psychological manipulation has become more refined.

For financial institutions, these fraud typologies present a dual challenge: protecting customers from becoming victims through outbound payment controls and customer education, while simultaneously detecting the money laundering flows generated by fraud proceeds as they are layered through the financial system. Both dimensions require urgent attention.

Regulatory, Enforcement, and Market Context

The UK's Payment Systems Regulator introduced mandatory reimbursement for authorised push payment fraud in October 2024, creating a powerful financial incentive for payment service providers to invest in fraud prevention and detection. Early data from 2025 shows that while reimbursement rates have improved significantly, the volume of romance and investment fraud continues to grow — suggesting that prevention, rather than compensation, must be the primary focus.

In Australia, the National Anti-Scam Centre has expanded its operations and now coordinates intelligence sharing between banks, telecoms, and digital platforms. AUSTRAC has issued specific typology guidance on the financial indicators of romance and investment fraud, including patterns of escalating payments to overseas accounts, use of cryptocurrency exchanges for cash-out, and the involvement of domestic money mule accounts.

FATF's updated risk indicators for fraud-related money laundering, published in early 2026, recognise romance and investment fraud as distinct predicate offences requiring specific detection strategies. The guidance calls on financial institutions to develop fraud-specific monitoring scenarios that go beyond traditional AML typologies.

What the Data Is Showing

Data from the Australian Competition and Consumer Commission shows that Australians lost over $2.7 billion to scams in 2025, with investment fraud accounting for 47% of total losses and romance scams for 12%. Critically, the average loss per romance scam victim exceeds $100,000, reflecting the extended cultivation period and deep psychological manipulation involved. Victims aged 55 and over are disproportionately affected.

Research by Europol's European Cybercrime Centre indicates that romance and investment fraud networks are increasingly linked to organised crime groups operating from Southeast Asian scam compounds. The financial flows follow a consistent pattern: victim payments to domestic mule accounts, rapid dispersal across multiple accounts, conversion to cryptocurrency, and final cash-out through exchanges in jurisdictions with weak AML controls.

Implications for Financial Institutions

Banks and payment service providers must deploy targeted intervention mechanisms for payments that exhibit romance and investment fraud characteristics. This includes real-time payment screening that identifies high-risk payment patterns — such as first-time international transfers to known fraud corridors, escalating payment amounts, and transfers to cryptocurrency exchanges — combined with customer friction measures such as warnings, delays, and human intervention for flagged transactions.

The integration of fraud and AML detection (FRAML) is particularly relevant here. Romance and investment fraud generate money laundering flows that traditional AML monitoring may not detect because the predicate offence involves authorised payments from the victim's own account. FRAML approaches that combine fraud indicators with money movement analysis can identify both the victim-side and the laundering-side of these schemes.

Customer education remains a critical but insufficient defence. Institutions should complement awareness campaigns with embedded warnings in digital banking platforms, triggered by behavioural signals that suggest a customer may be under the influence of a scam.

Conclusion

Romance and investment fraud represent a clear and growing threat that demands coordinated action across the financial system. The human cost is enormous, the financial losses are staggering, and the criminal infrastructure behind these schemes is becoming more sophisticated. Financial institutions that fail to build targeted detection and intervention capabilities are not only failing their customers — they are facilitating the money laundering that sustains these criminal enterprises.

Suggested Next Steps

• Implement real-time payment screening rules specifically designed for romance and investment fraud typologies, including escalating payment patterns and first-time transfers to high-risk corridors.

• Deploy customer-facing intervention mechanisms — including in-app warnings, payment delays, and call-back procedures — for transactions exhibiting fraud risk indicators.

• Develop FRAML detection scenarios that identify both victim-side authorised payments and the downstream money laundering through mule accounts and crypto exchanges.

• Partner with law enforcement and industry intelligence-sharing platforms to report and receive real-time intelligence on active fraud campaigns and mule account networks.

Sources: FBI IC3 Annual Report 2025, UK PSR APP Fraud Reimbursement Data, ACCC Scamwatch Annual Report 2025, AUSTRAC Romance and Investment Fraud Typology Guidance, FATF Fraud-Related ML Risk Indicators 2026, Europol IOCTA 2025.

TrustSphere helps financial institutions design and deploy intelligent fraud and financial crime detection solutions. Visit www.trustsphere.ai