Money Laundering

End the Plague of False Positives and False Negatives

AML fines are currently over $14B per year and on the rise. To avoid getting fined, compliance teams are monitoring any sign of potential AML activity. However, traditional approaches to AML transaction monitoring are rigid, prone to false alarms and missing many of the most sophisticated incidents of money laundering. A recent PWC report estimates that 90-95% of alerts are false positives. Using an unsupervised machine learning approach, DataVisor provides the most advanced AML transaction monitoring solution that can drastically reduce false positive and false negative problems in your organization.

Request Trial

Common Attack Techniques

Creating mule accounts

Money launderers utilize synthetic or stolen identities to create fake accounts to move funds

Structuring

Money launderers avoid CTR reporting through depositing funds in small denominations

Layering

Money launderers transfer funds through multiple mule accounts to mask money flows

Fund transfer to/from high-risk locations

Money launderers send funds to financial secrecy havens to protect illegal funds

Why Unsupervised Machine Learning for Transaction Monitoring

Traditional transaction monitoring solutions rely on rules or supervised machine learning models that require constant tuning as bad actors discover new ways to evade detection and banks are continuously adding new financial products that need to be protected. Powered by our big data architecture, our unsupervised machine learning model analyzes hundreds of data fields across hundreds of millions of accounts and events. This approach dramatically reduces false positives while also uncovering attack techniques that went previously undetected by traditional AML solutions.

Lower false positives

Reduce costly investigations driven by false alerts, allowing you to focus on investigating alerts that lead to substantive SAR filings

Automated rules tuning

Spend less time managing rules by automatically generating and sunsetting rules derived from unsupervised machine learning detection results

Lower false negatives

Easily adapt to evolving money laundering techniques and changing product types

Key Features

Account linkage visualization

Show linkage among suspicious accounts with unusual activities and track money flow through accounts with campaign visualization

Automated Rules Engine

Auto-maintain human readable rules (powered by UML) to eliminate the need for manual tuning and easy reporting to auditors. Allow users to manually create rules, test effectiveness, then deploy

Complement existing TMS

Run in parallel with existing TMS and case management tools, eliminating need to retrain or modify existing workflows

Account linkage view to discover hidden links between money laundering accounts

What’s Happening in AML

BLOG POST

Guest Post: End the False Positive Alerts Plague in Anti-Money Laundering (AML) Systems

Traditional transaction monitoring systems (TMS) suffer a plague of false positives. In this guest post from Keith Furst, founder of Data Derivatives, Keith discusses the problem with false positives and how banks should embrace unsupervised machine learning to raise the stakes in the fight against money laundering.

Read Blog

Getting Started

Want to get started and find out how DataVisor can help find malicious accounts hiding inside your online service? Request a security assessment today!

Request Trial