Chinese Banks Embrace Facial Recognition To Fight Fraud

China-Banks-Facical-Biometrics-fraud

Banks in China have embraced micro-expressions technology as they try to combat fraud earlier by studying the facial movements of customers.

The Financial Times, citing the Chinese banks, reported they are concerned that some of their would-be customers are lying about why they are taking out loans — so the banks have created systems using smartphone cameras to detect tiny facial expressions that can indicate if the person is being truthful. According to the Financial Times Ping An, the financial services company, has created the technology and is using it in its lending arm. It is among the first to embrace the technology commercially, noted the report. Western banks have been less eager to use the technology over concerns that it may be unreliable and unethical.

The technology developed at Ping An can pinpoint 54 brief, involuntary micro-expressions that a face often creates before the brain can control the movements of the face. They include eyeballs that move back and forth and rapid blinking.  “We use micro expression recognition technology to review loan applications. It captures subtle changes in customers’ facial expressions which help to identify and warn against fraud risks,” Lee Yuan Siong, one of Ping An’s deputy chief executives, told the Financial Times.  “We’ve reduced credit losses by 60 percent using this technology. It is more accurate than other approaches to fraud detection.”

Ping An isn’t relying solely on the tech to approve a loan, but if it does spot something amiss, the customers will be flagged and place under extra scrutiny, reported the FT, noting Ping An isn’t alone in testing the technology.  Kee Sun-Tuan, chief information officer in China at Standard Chartered, told the FT micro-expression technology “acts like a lie-detector, using the camera in a phone to check people’s responses to questions and scan their facial movements.” The report noted that the system can also enable banks to assess lending risk remotely, providing the opportunity to do business where they may not have physical branches.