Nasdaq testing artificial intelligence systems to track rogue traders
Source: Bill
South Korean professional Go player Lee Se-Dol (R) puts his first stone against Google's artificial intelligence program, AlphaGo. Nasdaq is trying to use artificial intelligence screening against rogue traders.
Nasdaq Inc. is trying to identify would-be white-collar criminals by using artificial intelligence systems originally built to track terrorists and sex traffickers. The exchange is testing systems that analyze data about trading activity against what traders say on their corporate chat and email accounts, in an effort to spot potential insider trading, market manipulation and other crimes faster and more accurately than current surveillance systems can.
Parsing the chatter of traders in the time before, during and after transactions -- and matching those findings with trading data -- provides “holistic surveillance,” said Bill Nosal, vice president of business development for market technology at Nasdaq NDAQ, -0.38% . “This can show what was happening in the trader’s head,” Nosal said.
The work comes as New York Attorney General Preet Bharara and other state and federal officials step up efforts to prosecute financial crimes. Since the financial crisis in 2008, big banks have been trying to improve methods for detecting not just potential crime but overly risk behavior by traders, said Anshuman Jaswal, a senior analyst at Celent, a research and consulting firm for the financial services industry.
| }
|