
IBM’s newest invention uses cloud analytics in order to determine if app traffic is legitimate or not. This technology works by analyzing the data entry points on your computer or mobile device. This data collection set is stored in the cloud and analyzed each and every time you authenticate to a service. For example, let’s say you are using your tablet to access your online banking. When you do this, there are typically a couple key areas that you tap or swipe that help you authenticate into the service. IBM’s invention detects if these input points are different than what others commonly tap or swipe on their mobile devices. You’re probably wondering, “How does this keep me safe?”
Malicious users aren’t typically clicking, tapping or swiping manually. In fact, hackers and fraudsters have applications built that simulate many of these activities. If you’ve logged into your online banking the same way 100 times previously and the 101st time, you login completely differently, the algorithm detects this and prompts the user for additional credentials. Much like a signature or a finger print, everyone has their own style of clicking and browsing the internet. IBM’s invention is smart enough to realize that a new set of clicks, taps and swipes are not like previous clicks, taps and swipes that IBM’s analytics algorithm is used to seeing.
IBM Master Inventor Keith Walker gives a better example by saying, “If an individual suddenly changes how they interact with an online bank or store, such as due to a broken hand or using a tablet instead of a desktop computer, I want these web sites to detect the change, and then ask for extra identity confirmation before accepting a transaction. Our experience developing and testing a prototype, which flawlessly confirmed identities, shows that such a change would more likely be due to fraud, and we all want these sites to provide more protection while simultaneously processing our transactions quickly.”
Although internet merchants and banks have taken significant steps to shore up security and authentication, online fraud is still a major concern. Just like with all IBM inventions, the fraudster prediction and recognition algorithm received its patent and all of the details can be found by searching for U.S. Patent #8,650,080. This invention is practical for both consumers and merchants and you should expect to see this technology pop up in many of your favorite web apps in the near future.