Synthetic Fraud – Nemesis of the Financial Institutions

Synthetic Fraud is the most recent Cybercriminal Activity that has become the fastest growing financial crime. Synthetic Fraud, fueled by “Frankenstein Identities” is the hardest to detect and quickly accounting for the highest percentage of fraud losses primarily involving credit cards and unsecured lending portfolios. Unfortunately, by the time the banks, credit unions and lending companies act on red flags and finally identify these borrowers do not exist, they are out of hundreds of thousands of dollars.

Synthetic Fraud is quite different from traditional identity fraud, Cybercriminals create “Frankenstein Identities”. Instead of stealing a real identity, scammers create a new fake identity using a real and/or unused social security number, combine the social security number with a fictitious name, driver’s license and physical address to form the “Frankenstein Identities”.

These Synthetic Fraud scammers play the long game, taking months to years to build up history with the credit bureaus. Once the credit profile is established, they apply for credit with various financial institutions leading till they get approval. Approval may start with a secured credit card or a product designed for high risk borrowers. Over time, the scammers establish credit worthiness by acquiring multiple credit cards and small loans. Scammers max out all credit and disappear forcing the financial institutions to write off significant losses.

Financial Institutions need to re-think their fraud protection strategy, particularly for digital on-boarding and lending transactions without adding friction to legitimate customers. Financial Institutions must ensure additional attributes beyond credit score is verified by evaluating third-party data. Ensuring there is high degree of consistency within additional data attributes by matching them to a proven identity.

Emerging technologies like Artificial Intelligence and Machine Learning, combined with Big Data Analytics and High-Performance Computing applications can process large volumes of data from various sources and data sets (Structured, Unstructured and Semi-structured).  Such technologies can achieve higher scalability by processing large scale data types containing media data, image data and object data to spot trends and patterns across disparate data sets. Financial Institutions can predict and capture “Synthetic Fraud” applicants early in the process by applying accelerated deep learning and statistical machine learning technologies.

Artificial Intelligence and Machine Learning technologies improve the efficiency of the credit decisioning process, while ensuring automated systems speed up the approval timeline to gain higher revenue and credit worthy customer market-share. Synthetic Fraud scammers get rejected, credit-worthy customers appreciate the benefits of less paperwork workload and faster decisioning process. Artificial Intelligence and Machine Learning will play an important role in protecting bona fide customers and financial institutions from “Synthetic Fraud”.