Introducing Jumio 360° Fraud Analytics at Money20/20 USA, leveraging AI to detect and combat fraud linked to account sharing.
There is no denying the capability of artificial intelligence (AI), but it can be disastrous in the wrong hands. Jumio, an automated end-to-end identity verification, risk assessment, and compliance solutions provider, has introduced its Jumio 360° Fraud Analytics at Money20/20 USA in order to counter this danger.
Jumio’s research indicates that 25% of fraud is linked, meaning it is either committed by fraud rings or by individuals opening new accounts using the same details or login credentials. These may be found on sharing economy websites, banking websites, e-commerce sites, and other websites. The new Jumio function employs AI-driven predictive analytics to more accurately and sophisticatedly identify fraud trends in an effort to counteract this.
How it functions
It examines more than just connections, such reporting someone for being associated with an established scammer. Rather, it uses Jumio’s cross-industry network’s billions of data points to find patterns based on indicators and behavioural commonalities.
Identity transactions are grouped into clusters across the network using graph database technology with a layer of machine learning, and each cluster’s fraud risk is calculated. This offers a multifaceted perspective on every transaction and the overall cross-customer ecosystem.
Additional attributes comprise:
- Using cluster association, AI-driven predictive analytics calculates each identify transaction’s probability of fraud risk.
- Identity transactions may be automatically rejected by the predictive fraud score. Moreover, it can instantly prevent scammers from gaining access to your site.
- Dashboards facilitate integrated data visualisation and offer transparency.
Deserving inclusion
Today, Jumio runs a number of sophisticated fraud checks when a user goes through the ID verification procedure. The identification transaction will also be compared to the clusters with the inclusion of Jumio 360° Fraud Analytics, which will produce a prediction fraud score that may be used to immediately reject the transaction if it over a predetermined threshold.
According to Jumio’s preliminary research, this method increases the company’s already very accurate fraud detection rate by at least 30% while lowering the incorrect rejection rate. Through APIs via the Jumio Portal, business users can audit the decision’s justifications.
A limited number of customers can presently get Jumio 360° Fraud Analytics through an early release. It will go on sale to the general public in early 2024.
According to Philipp Pointner, chief of digital identity at Jumio, “we provide insights on why the identity transaction was rejected because explainability is a key requirement when using automation.” Additionally, we offer a graphical depiction of the high-level connections we discovered. This facilitates the visualisation of relationships amongst individuals, papers, gadgets, and more. This new tool is revolutionary in the fight against fraud because it combines advanced analytics with easily digestible insights.
According to Jumio CEO Robert Prigge, “Jumio revolutionised online identity verification when it pioneered the ID + selfie approach.” “Today, by utilising predictive analytics to find fraud trends throughout our extensive network, we are pushing the industry to change once more. This will provide us greater accuracy than ever before in identifying and disrupting fraud rings and other coordinated attacks.
