Fraud is now identified as a major money-laundering risk in 90% of assessed jurisdictions, and in the UK alone it accounts for more than 40% of all recorded crime. What stands out in the FATF report is not only the scale, but the structure of modern fraud. These are no longer isolated scams. Digitalization has lowered the barriers to entry where criminal groups can now run industrialized operations.

Messaging platforms, social networks, instant payments and crypto rails have created a global operating environment where fraud can be organized, outsourced and scaled.

The structure is strikingly modular. One group specializes in recruiting victims through social engineering or investment scams. Another maintains phishing infrastructure, while others focus on laundering funds through mule networks or moving proceeds through payment systems and digital assets. The participants may never meet, but the system functions as a coordinated whole.

Financial institutions face asymmetry here. Fraud networks operate across organizations and jurisdictions, while detection remains largely confined within institutional boundaries: banks see transaction flows, while law enforcement and FIUs see fragments after the fact.

The FATF report returns repeatedly to this point: relevant intelligence often exists, but it sits in separate organizations and in datasets that cannot easily be combined. Legal obligations, privacy rules, and commercial sensitivities make the sharing of raw customer data difficult, and in many cases undesirable.

For years, the industry has treated this as a governance problem. In reality, it is also a technical one.

Encrypted computing offers a different approach. In particular Multi-Party Computation allows organizations to analyze data collectively without revealing the underlying records. Each participant keeps control of its own information, while the analysis runs across encrypted inputs.

The output is not a shared database but shared insight: the identification of overlapping mule networks, repeated scam accounts or cross-institution transaction patterns.

The FATF report stops short of prescribing technologies, but its conclusions point to a clear direction. Fraud detection must become more networked, more collaborative, and more intelligence-led. For banks, that means challenging the assumption that better analytics inside a single institution will be sufficient.

To stay ahead of incoming regulation, banks must look at their collaboration strategies to comply with PSR Art 83 and AMLR Art.75 without creating new risks around privacy, competition or data sovereignty. This exists in solutions that allow banks to connect signals while keeping data under their own control. Multi-Party Computation is the only technology that moves with the speed, precision and security that these regulations demand to create anti-financial crime networks. 

If cyber-enabled fraud has become an ecosystem, the response needs to look more like one as well.