0%

Book Description

As new digital channels emerge for monetary transactions, financial crime continues to soar. The good news is that recently developed AI-based crime-fighting systems are already having a positive impact. In this report, Atif Kureishy (Teradata Consulting) and Chad Meley (Teradata) examine online criminal activity and describe the benefits and challenges of deploying AI models for fighting digital crime.

Roughly two-thirds of all businesses around the globe experienced financial criminal activity in 2017—up 58% from the year before. Legacy practices and traditional rules engines simply can’t keep up. This report delves into research on the current state of AI adoption worldwide and discusses the advantages of AI models as well as the difficulties of putting them into practice.

With this report, you’ll explore:

  • Different types of financial crime, including sophisticated fraud schemes, cybercrime, and money laundering
  • The fallout that successful criminal schemes have on financial services firms
  • Challenges to staying ahead of financial crime, such as regulatory complexity, real-time transactions, and pressure to innovate
  • The state of today’s anticrime measures in financial institutions and the benefits of AI-based models
  • Challenges that crop up when deploying AI models for fighting financial crime

Table of Contents

  1. 1. Fighting Financial Crimes with Artificial Intelligence
    1. Executive Summary
    2. Financial Crime Continues to Increase
    3. Different Types of Financial Crime
      1. Fraud
      2. Money Laundering
      3. Cybercrime
    4. Fallout to Financial Services Firms of Successful Crime
    5. Challenges to Keeping on Top of Financial Crime
      1. Criminals Are Innovating Relentlessly
    6. Current State of Anticrime Measures in Financial Institutions
      1. Today’s Transaction Monitoring Systems
      2. The Burden of False-Positive Alerts in a TMS
    7. The Emergence of AI-Based Crime-Fighting Systems
      1. Benefits of AI-Based Models to Fight Financial Crime
    8. Challenges of Deploying AI Models when Fighting Financial Crime
      1. AI Models Can’t Just Be Handed Over to IT
      2. Model “Explainability” Can Be Difficult
      3. Fragmentation of Teams—and Therefore Siloed Data
      4. Real-Time AI Deployment: Current Realities and Constraints
      5. Managing AI Models in Production
      6. Managing Alerts from AI-Based and AI-Enhanced Monitoring Systems
      7. Other Operational Challenges
      8. Accelerating AI Model Deployment with AI-based AnalyticsOps
      9. Don’t Replace Your TMSs—Supercharge Them
    9. Real-World Case Study: Danske Bank
      1. Overcoming Challenges
      2. From Machine Learning to Deep Learning
      3. A Platform for the Future
    10. Conclusion