Joel R. McConvey
In an article in the Financial Times, Onfido’s chief product officer says AI does better than people at verifying customer identity through facial recognition, specifically for banks and financial institutions.
Yuelin Li says that “humans are not particularly great at recognizing humans,” especially when making a 2D to 3D match using an ID photo that could have been taken years ago. She cites a study that compared the accuracy in facial recognition by forensic examiners and other “professional facial examiners” against face recognition algorithms (which, in fact, concluded that “optimal face identification was achieved only when humans and machines worked in collaboration.”)
Regardless, the question of human versus automated onboarding for financial institutions is on its way to becoming moot. Increasingly, says Li, banks and other businesses are conducting verification online, through apps or portals, increasingly using biometrics.
“One of the things people are very nervous about is things like call centers,” she tells the Financial Times. “A lot of attacks that are going on now – a phishing attack or a text message or something – they get you because you’re coming out of the app.”
Although some research shows that a high percentage of customers abandon digital bank onboarding, biometrics has the potential to expedite existing onboarding processes that are inefficient and clumsy. In addition to familiar tools such as fingerprint or face biometrics, firms such as the French fintech Revolut are exploring behavioral biometrics that analyze gestural factors such as device positioning and typing speed as an addition to their verification toolkit.
-
0 Comments