Nasdaq Turns to AI to Detect Market Manipulation

Nasdaq undertakes 14 million trades a day on its exchange making it nearly impossible to detect market abuses taking place. However, with the help of artificial intelligence (AI), Nasdaq is cracking down on the bad behavior.

There are several things that Valerie Bannert-Thurner, senior vice president and head of risk and surveillance at Nasdaq, considers when it comes to bad behavior on the exchange.

Speaking to the American Banker, she said:

If people are excessively profitable given how they trade and in comparison to everybody else trading the same instruments with similar styles, then we ask, is this luck or something else?

Yet, with an excessive amount of trades taking place each day at Nasdaq, keeping track of who’s doing what can seem a daunting task.

In a bid to lighten the workload, the exchange is turning its attention to AI to help, and is one of the first exchanges to use AI as well.

Maintaining the Exchange’s Integrity

As Nasdaq is a well-known exchange that operates 25 exchanges in Canada, Europe, and the U.S. it needs to maintain its trustfulness or else companies will cease from using it.

According to Bannert-Thurner, Nasdaq exists because people believe in its abilities to detect market manipulation and abuse. So-much-so, that 45 outside exchanges and 13 regulators use its SMART trade surveillance platform.

Typically, Nasdaq has large teams of people using the SMART platform to detect bad behavior on the exchange. They do this while also connecting what traders are saying in emails and chats with what their actual trades are doing for them to catch them in the act.

However, it was in 2014, that the Libor scandal uncovered the manipulation of Forex markets, with U.S. regulators investigating many incriminating chats to fix rates.

To prevent situations like this taking place again, Nasdaq is employing AI software from Digital Reasoning.

AI to Detect Market Abuses

Using language processing and machine intelligence, the software can detect the language traders are using and identify any key triggers, which will alert those seeking out market abuse to look at what trades were made.

To make sure the software can detect what is going on, Nasdaq is feeding it alerts and chats from previous cases of manipulation.

Bannert-Thurner said:

We think machine learning will make a significant difference in how we detect behaviors, how we help people review and investigate.

The Digital Reasoning software will screen the threat indicators first before humans see them. It is hoped that the software will reduce the number of alerts down to only the relevant ones.

Featured image from Shutterstock.