IBM has patented a technique that helps online and cloud-based businesses eliminate fraud by analyzing browsing behavior to determine whether customers are who they say they are after accessing a website or app. For example, when individuals access a banking or shopping site, they subconsciously establish characteristics of how they interact with the site, such as clicking certain areas more often than others; using the up and down arrow keys on the keyboard to navigate; relying solely on the mouse; or tapping or swiping the screen of a tablet or smartphone in a distinct manner. Similar to how individuals recognize changes in the behavior of a family member or friend on the phone - even when the audio is fuzzy - by the words they use, how they answer the phone, their mannerisms, etc., IBM's invention helps businesses analyze and identify sudden changes in online behavior.

If the invention detects a change in behavior, it triggers a secondary authentication measure, such as a security question. This helps businesses and website operators avoid unintentionally hindering legitimate customer activities or transactions.

IBM received U.S. Patent #8,650,080: "User-browser interaction-based fraud detection system" for the invention.