Signifyd COO and Co-Founder Michael Liberty discusses machine learning and fraud prevention with PaymentsSource
Liberty begins by clarifying Artificial Intelligence as more than automated, decision-making processes consisting of both traditional, rules-based screening systems and machine learning tools that acquire knowledge and determine new outcomes and results. Signifyd’s real-time machine learning screens and offers a viable, convenient, guaranteed alternative to ecommerce businesses who simply can’t provide the same level of attention to each transaction.
Also interviewed is Jay Sung, CEO of EcoReco Scooter (a Signifyd customer), who design, produce and sell top-of-the-line electric scooters. Sung describes his experience when EcoReco began and how they suffered a high amount of fraudulent orders. Signifyd helped the company recover 5% to 10% of lost sales and completely eliminated fraud for the company by accepting liability on any fraudulent chargebacks.
Liberty calls this path to freedom from financial burden a “two-way street” dependent on customer feedback for machine learning to really succeed. Feedback flows freely with Signifyd’s customers because Signifyd covers any losses from fraudulent orders and thus all negative results are reincorporated to improve their machine learning. “That is a critical step to close that loop,” says Liberty, “if you want to apply machine learning to this domain. Plus, it’s an interesting positive effect of our business model.”
The full article is available here on PaymentsSource.
David Heun regularly provides coverage of breaking stories, industry trends and interesting companies in the payments space. PaymentsSource provides comprehensive coverage of the global payments industry with breaking news and in-depth analysis by industry experts.