- Jumio underscores the growing risks of organized crime and AI-based fraud, resulting in significant financial harm. Jumio 360° Fraud Analytics provides businesses with a proactive way to protect themselves.
- Jumio 360° Fraud Analytics employs graph database technology and machine learning to group identity transactions into clusters across the network, assessing the fraud risk of each cluster.
Digital identity management and verification startup Jumio Corp. recently unveiled a new fraud-fighting technology that leverages predictive analytics powered by artificial intelligence to more accurately and sophisticatedly identify fraud patterns.
The new service, Jumio 360° Fraud Analytics, was created to address the problem of interconnected fraud, which is fraud committed by fraud rings or individuals who use the same information or credentials to open new accounts on banking, e-commerce platforms, and sharing economy websites.
According to Jumio, a single organized crime group or fraud ring can cause hundreds of millions of dollars in damages. With the availability of AI, fraudsters are becoming more sophisticated than ever, forcing companies to adapt in order to stay ahead of the game. Jumio 360° Fraud Analytics comes into play here.
Jumio 360° Fraud Analytics addresses the challenge of interconnected fraud using a novel approach that goes beyond basic connections, avoiding the mere identification of individuals based on their association with known fraudsters. Instead, using behavioral similarities and other indicators, the new service analyzes billions of data points from Jumio’s network to find patterns.
Jumio 360° Fraud Analytics classifies identity transactions into clusters across the network and calculates the fraud risk of each cluster using graph database technology combined with a layer of machine learning. The business claims this offers a multifaceted perspective of every transaction and the entire cross-customer ecosystem.
Jumio 360° Fraud Analytics also intervenes when a user completes the ID verification procedure. The new service will also assess identity transactions by comparing them to the clusters, producing a predictive fraud score that, if it surpasses a specific threshold, can lead to the rejection of the transaction.
In testing, Jumio reported that this method enhances its already precise fraud detection rate by a minimum of 30% while maintaining the same low false rejection rate. Business users can scrutinize the decision’s rationale within the Jumio Portal or through application programming interfaces.
Jumio’s Chief of Digital Identity, Philipp Pointner, said, “Because explainability is a key requirement when using automation, we provide insights on why the identity transaction was rejected. We also provide a graphical representation of the high-level linkages we found, which makes it easy to see connections between people, documents, devices and more.”
With a general release anticipated for early 2024, Jumio 360° Fraud Analytics is currently being made available in early release to a select group of Jumio customers.