Introduction
We all know that criminals try to stay one step ahead of the law. That’s one of the biggest challenges in fighting financial crime. Yesterday’s detection techniques are woefully inadequate today. Rules-based (or signature-based) cyber detection techniques have been a source of concern among crime prevention practitioners for the past several years due to the inability of those techniques to keep pace with emerging new attack vectors. Most in the financial industry have embraced various artificial intelligence (AI) and machinelearning (ML) techniques, essentially to allow the model to learn the rules required to detect attacks. The hypothesis has been that, with a big enough sample of historical attacks, one model can be trained to recognize a new attack and block it as it occurs.
