New methods to combat fraud were a hot topic at Money 20/20 USA this year. In crypto and the traditional financial world, customers have lost millions from hacks, phishing, SIM card takeovers, and wasted time — all due to the difficulty of proving identity in the digital realm. Mastercard-acquired NuData is using some interesting techniques based on behavioral patterns and machine learning with its product NuDetect, so Bitsonline set out to find out more about how it can help.
Also see: Perry Farrell Talks About Kind Heaven, a Mythical and Immersive Arts Experience
NuData Can Understand Your Intent as well as Your Device
It’s clear the login and password model is no longer adequate. Even some two-factor authentication techniques have flaws, if hackers can gain access to customers’ phones and emails. Simple device-based metrics are losing favor as users move between multiple devices, and locations, to access financial services.
“Instead of looking at the car, we’re trying to understand the user behind the car,” says NuData’s Don Duncan. Understanding more about a user’s typical behavior can show their intent, and help prove whether that user is the real account holder, or an imposter.
Watch the video above to see NuData’s demonstration — the exact same login name and password typed by two different people results in a completely different data profile.
Trust Without Knowing Names or Even Identities
The way each person interacts with their device can be as unique as a biometric. Typing cadence, long and short key presses, right and left-handedness etc. — NuData/NuDetect can detect and analyze these actions to build a unique user profile that doesn’t even require a real name.
Duncan likens it to the relationship between a shopkeeper and a regular customer, who may build trust in each other over time without knowing each other’s name. The shopkeeper knows the face and actions of their customer, and vice versa. Companies can store this information in a form that wouldn’t be useful to any hacker, even in the event of a data breach.
NuDetect can even work when making payments, i.e.: when a credit card or other payment method is used without going through the login process.
Advanced machine-learning techniques will play a greater role in identifying individuals in the coming years. While there are genuine privacy concerns over technology like face and gait recognition in public places, there will also be times when people need to prove their identity quickly and securely. It’s not easy, but companies like NuData and others are working to tackle the problem.
Are behavioral analytics the best way to prevent online and financial fraud? Tell us what you think in the comments.
Images and video via Bitsonline, Jon Southurst, Tyson O’Ham
Source: Read Full Article