The Pandemic Has Changed How Criminals Hide Their Cash — and AI Tools Are Trying to Sniff It Out


The pandemic has forced crim­i­nal gangs to come up with new ways to move money around. In turn, this has upped the stakes for anti-money laun­der­ing (AML) teams tasked with detect­ing sus­pi­cious finan­cial trans­ac­tions and fol­low­ing them back to their source.

Key to their strate­gies are new AI tools. While some larger, older finan­cial insti­tu­tions have been slower to adapt their rule-based legacy sys­tems, small­er, newer firms are using machine learn­ing to look out for anom­alous activ­i­ty, what­ev­er it might be.

It is hard to assess the exact scale of the prob­lem. But accord­ing to the United Nations Office on Drugs and Crime, between 2% and 5% of global GDP — between $800 bil­lion and $2 tril­lion at cur­rent fig­ures — is laundered every year. Most goes unde­tect­ed. Estimates sug­gest that only around 1% of prof­its earned by crim­i­nals is seized.

And that was before covid-19 hit. Fraud is up, with fears around covid-19 cre­at­ing a lucra­tive market for coun­ter­feit pro­tec­tive gear or med­ica­tion. More people spend­ing time online also cre­ates a bigger pool for phish­ing attacks and other scams. And, of course, drugs are still being bought and sold.

Lockdown made it harder to hide the pro­ceeds — at least to begin with. The prob­lem for crim­i­nals is that many of the best busi­ness­es for laun­der­ing money were also those hit hard­est by the pan­dem­ic. Small shops, restau­rants, bars, and clubs are favored because they are cash-heavy, which makes it easier to mix up ill-gotten gains with legal income.

With bank branch­es closed, it has been harder to make large cash deposits. Wire trans­fer ser­vices like Western Union — which usu­al­ly allow anyone to walk in off the street and send money over­seas — shut their premis­es, too.

But crim­i­nals are noth­ing if not oppor­tunis­tic. As the normal chan­nels for money laun­der­ing closed, new ones opened up. Vast sums of money have start­ed flow­ing into small busi­ness­es again thanks to gov­ern­ment bailouts. This cre­ates a flurry of finan­cial activ­i­ty that pro­vides cover for money laun­der­ing.

Breaking the rules

The upshot is that there are more demands being placed on AML tech. Older sys­tems rely on hand-craft­ed rules, such as that trans­ac­tions over a cer­tain amount should raise an alert. But these rules lead to many false flags and real crim­i­nal trans­ac­tions get lost in the noise. More recent­ly, machine-learn­ing based approach­es try to iden­ti­fy pat­terns of normal activ­i­ty and raise flags only when out­liers are detect­ed. These are then assessed by humans, who reject or approve the alert.

This feed­back can be used to tweak the AI model so that it adjusts itself over time. Some firms, includ­ing Featurespace, a firm based in the US and UK that uses machine learn­ing to detect sus­pi­cious finan­cial activ­i­ty, and Napier, anoth­er firm that builds machine learn­ing tools for AML, are devel­op­ing hybrid approach­es in which cor­rect alerts gen­er­at­ed by an AI can be turned into new rules that shape the over­all model.  

The rapid shifts in behav­ior in recent months have made the advan­tages of more adapt­able sys­tems clear. Financial reg­u­la­tors around the world have released new guid­ance on what sort of activ­i­ty AML teams should look out for but for many it was too late, says Araliya Sammé, head of finan­cial crime at Featurespace. “When some­thing like covid hap­pens, where every­body’s pay­ment pat­terns change sud­den­ly, you don’t have time to put new rules in place.”

You need tech that can catch it as it is hap­pen­ing, she says: “Otherwise by the time you’ve detect­ed some­thing and alert­ed the people who need to know, the money is gone.” 

For Dave Burns, chief rev­enue offi­cer for Napier, covid-19 caused long-sim­mer­ing prob­lems to boil over. “This pan­dem­ic was the tip­ping point in many ways,” he says. “It’s a bit of a wake-up call that we really need to think dif­fer­ent­ly.” And, he adds, “some of the larger play­ers in the indus­try have been caught flat-footed.”

But that doesn’t simply mean adopt­ing the latest tech. “You can’t just do AI for AI’s sake because that will spew out garbage,” says Burns. What’s needed, he says, is a bespoke approach for each bank or pay­ment provider.

AML tech­nol­o­gy still has a long way to go. The pan­dem­ic has revealed cracks in exist­ing sys­tems that have people wor­ried, says Burns. And that means that things could change faster than they were going to. “We’re seeing a greater degree of urgency,” he says. “What is tra­di­tion­al­ly very long, bureau­crat­ic deci­sion-making is being accel­er­at­ed dra­mat­i­cal­ly.”

MIT Technology Review source|articles

Recommended Posts

Start typing and press Enter to search