Sift Launches ThreatClusters: Transforming Fraud Detection with Industry-Specific Consortium Models

Sift, a leading AI-powered fraud prevention platform that secures digital trust for global businesses, has launched ThreatClusters, a pioneering data science innovation in fraud detection. This new feature enhances the accuracy of fraud decisions by integrating industry-specific model insights, combining the precision of customer-specific risk models with the extensive intelligence of a global model to generate unique risk signals for each industry.

As fraudsters deploy increasingly sophisticated tactics, including AI-driven threats that can outmaneuver many traditional fraud prevention strategies, conventional detection models often fall short. These models either focus too narrowly on a single organization’s data or apply insights too broadly across various industries. ThreatClusters overcomes these challenges by grouping companies with similar fraud patterns into cohorts, accounting for nuances in risk patterns, and enabling more accurate fraud decisioning.

Through Sift’s proprietary technology, customers can utilize a detection model fine-tuned to their specific cluster while also benefiting from models that highlight emerging fraud vectors from other clusters.

Key Features and Benefits of ThreatClusters:

• Increased Accuracy: ThreatClusters enhance fraud detection accuracy by up to 20%, reducing the risk of false positives and negatives by incorporating insights from industry-specific fraud patterns.
• Quicker Implementation: The integration of global and cohort models accelerates accuracy, leading to faster adoption and quicker realization of benefits for businesses.
• Improved User Experience: Industry-specific fraud patterns allow for better differentiation between legitimate users and fraudsters, enabling step-up friction without compromising the customer experience or conversion rates.

“ThreatClusters marks a major advancement in our mission to help businesses stay ahead of fraudsters,” said Raviv Levi, Sift’s Chief Product Officer. “With industry-specific consortium models, we can offer our customers unparalleled insights into fraud patterns unique to their industry while also protecting them against emerging threats from other sectors. This allows our customers to better assess risk, protect revenue, and grow with confidence.”

In addition to ThreatClusters, Sift’s latest release includes other significant innovations that optimize score accuracy and enable fraud and risk teams to more effectively detect sophisticated fraud behaviors across various use cases, such as payment fraud and account takeover.