Worldpay, Inc. announced that it is releasing FraudSight a new solution designed to prevent fraud and optimize approval rates potentially boosting revenue across all channels. Capitalizing on the power of machine learning that looks at billions of transactions, FraudSight makes transaction validity decisions for merchants of all sizes based on a matrix of data points and strategies. Fraud poses an ever-growing threat to merchants as worldwide card fraud losses reached $28.8 billion in 2018 according to The Nilson Report. That sum is projected to reach $42.3 billion by 2026, with most of that loss stemming from card not present (CNP) transactions, where the merchant often carries more liability. Traditional rules-based fraud tools are ill-equipped to keep up with the speed and sophistication of modern day attacks. Unlike rules-based systems, FraudSight's machine learning uses advanced technology to assign deep contextual awareness to each transaction. FraudSight is uniquely engineered to connect in-store and online transactions, which it uses to build behavioral models from card-specific activity, regardless of channel. The system looks at these behavioral and device biometrics, as well as contextual anomalies, chargeback reputation, and other factors that can help authenticate transactions. Worldpay merchants in North America looking to implement FraudSight can likely turn on the service with no integration or development necessary. Merchants can then elect to turn up the level of threat protection with additional features such as behavioral and device javascript integrations and additional third-party data feeds. With eCommerce set to grow by 55% in the U.S. by 2022,2 an easy-to-integrate solution is especially important to small and mid-sized merchants who are becoming more vulnerable to fraud liability online. On May 14 at 2:00 PM ET, Worldpay will be hosting a webinar titled "Payment Innovation Should Not Fuel Fraud: How Advanced Analytics and Data Strategies Outsmart Attacks" with guest speaker Rivka Gewirtz Little.