- Ted Sausen, Director and anti-money laundering subject matter expert, NICE Actimize
- Evan Weitz, Managing director and regional head of controls, Standard Chartered Bank
- Jayati Chaudhury, Financial crime anti-money laundering transaction monitoring, Global investment banking lead, Barclays
- Moderator: Duncan Wood, Editor-in-chief, Risk.net
As money laundering fines have grown, so have regulatory expectations and banks’ determination to meet them. But this isn’t easy – criminal typologies are also changing, and banks can find it hard to keep up.
This combination of factors is placing major strain on the financial investigation units responsible for identifying and reporting suspicious activity – false-positive rates can be as high as 95%.
It’s no surprise banks are drafting in robots to help – machine learning tools, for example, seem ideally suited to weeding out suspicious activity. It can, however, be hard to explain how some forms of machine learning reach their conclusions, presenting the industry with a different challenge.
Key topics explored include:
- What regulators expect from crime detection and analytics capabilities
- Where machine learning can help – and where should it be avoided
- Model risk challenges for machine learning
- Addressing concerns about explainability.