As the saying goes, “Money makes the world go round.” While that’s true in a legal sense, it is also true for criminals and terrorists who launder money to finance their operations. Although this might not seem like an everyday problem, data from the United Nations suggests otherwise. It says that between two and five percent of the global gross domestic product (GDP) is laundered every year. That adds up to around $800 million to $2 trillion.
In the hands of wrongdoers, that sort of funding can do a lot of harm. That’s why Intel is teaming up with a company called Consilient to do something about it. Together, the duo is using artificial intelligence (AI) and federated learning to stop money laundering with a data-driven approach.
Trying to put together a system that monitors financial data around the world without the help of artificial intelligence would be nearly impossible. That’s part of the reason why such a system hasn’t existed until now.
Juan Zarate, the first-ever assistant secretary of the U.S. Treasury for terrorist financing and financial crimes, says, “When banks try to detect illicit and fraudulent activity, the system is highly inefficient and ineffective, with over 95 percent of transaction monitoring rendering false positives and institutions unable to see risk beyond their own walls.”
The extremely high error rate allows money laundering schemes to slip by undetected unless they are fairly blatant. Zarate continues, “With Consilient’s federated machine learning technology, backed by Intel SGX, we are redesigning the way financial institutions and authorities discover and prevent financial crime risk dynamically and securely.”
It’s clear that addressing financial crimes requires a joint effort. Individual institutions simply can’t see everything. Consilient’s shared approach makes it possible to see more of what’s going on in the world’s finances.
At the same time, sensitive data and customer information isn’t revealed. Intel notes that this can help “accurately and efficiently detect illicit activity, with lower false positive rates, helping to combat financial crime, thwart higher-value money laundering, and enable legitimate individuals and businesses to manage risk more effectively.”
The entire system is made possible thanks to federated learning. For those who aren’t familiar, this is a form of shared machine learning that allows companies to train AI systems faster without sacrificing privacy. It utilizes a network of computers and decentralizes data while training the algorithms on multiple datasets.
The system is possible thanks to Intel’s Software Guard Extensions (SGX). It says that SGX “uses a hardware-based trusted execution environment (TEE) to help isolate and protect specific application code and data in memory.”
Consilient’s system is designed for governments and financial institutions alike since both play a role in stopping financial crime. It will be interesting to see how it is adopted and the sort of impact it can have in the real world.
If the Consilient system is effective at stopping money laundering and terrorist financing schemes, it could make the world a much safer place.