Recently HSBC announced that it is implementing AI technology to automate AML investigations – (Source: Reuters 01-June-2017)
Similarly, many more banks have started using AI technology in various fields. Since most of us are aware about Machine Learning (ML), Artificial Intelligence (AI) and Robotic Process Automation (RPA), we will not delve into the same.
In general, earlier technology products only offered solutions based on the fixed set of rules, e.g. If X event happens then do this, else do that. However, in AI, products develop repository rules of their own based on the huge volume of data, also known as pattern recognition. So, decision making on basis of pattern recognition has lots of scope in banking and financial applications. Some of the areas in banking where AI is being used extensively are:
• Anti-Money Laundering (AML) Pattern Detection
• Chat Bots
• Fraud Detection
• Stock Trading
We will only cover AML Pattern Detection in this series.
AML pattern Detection: Earlier detection of Money Laundering involved monitoring of bank accounts by its staff.
Let’s understand evolution of AML with the help of an example:
Pre computerization era: Earlier cashier used to report if unusual amounts of transactions start taking place in customer’s account. Risk and compliance department used to investigate and monitor accounts and put forward the recommendations like account closure or close monitoring or reporting to concerned legal authorities.
Computerization Era: Then arrived the software products which used to trigger alerts on the basis of certain events such as high cash deposit, rapid movement of money between multiple accounts, and so on. Now other than Bank staff computer too was alerting Risk and compliance team.
AI Era: In above case computers generated only alerts based on events and hence lots of false alerts were being generated. E.g.: There were many business organizations like restaurants whose nature of business involved heavy cash handling. In such cases system ended up generating alerts for such accounts although humans (bankers) could easily judge that there is no anomaly in such cash deposits given the nature of business of restaurants where transactions are mostly in cash.
What was missing was the ability of computers to take judgement calls based on the patterns and the past history. Now this piece is falling in place with the help of AI, and banks are expecting not only to reduce false alerts, but, also catch those cases where transactions are below thresholds but don’t fit into the pattern. A typical system flow is as under:
Thank you for reading this blog. Please standby for our next Blog on AI in Banking – Chat Bots.
- AI in Banking(2) – Chatbots
Hello readers, as promised, we are here with our 2nd blog in the series of blogs on “AI in Banking and Financial Services”, focusing on Chatbots. As discussed in the…
- AI in Banking – Fraud Detection
Hello readers, as promised, we are here with our third blog in the series “AI in Banking and Financial Services”. While our previous two blogs focused on Anti-Money Laundering (AML)…
- An Insight into the World of Datorama
Gartner marketing analytics research highlights that in 2017-18 about 9.2 percent of marketing budgets were consumed by marketing analytics alone. However, a majority of marketing teams continue to wrangle data,…
- An Insight into the Magical World of IoT – Part 1
IoT has become a major phenomenon globally because of its impact on our lives, particularly as a lifesaving system. However, before we dive into the technology, we first have to…
- All You Need to Know About Blockchain
Blockchain is quite the buzzword these days, but how many of us really know about it? Don’t fret if you don’t, this blogpost will prove to be a great source…
- KAFKA-Druid Integration with Ingestion DIP Real Time Data
The following blog explains how we can leverage the power of Druid to ingest the DIP data into Druid (a high performance, column oriented, distributed data store), via Kafka Tranquility…