Socure, a provider of Day Zero digital identity verification, today announced the launch of Sigma Synthetic Fraud, its latest addition to the Socure ID+ platform to address the complex and rapidly growing problem of synthetic identity fraud. Synthetic ID fraud, which involves fictitious, created identities that have been cultivated over months and even years, represents the next wave of complex fraud and financial theft online.
As early as 2016, synthetic ID fraud accounted for 20 per cent of total credit losses for U.S. lenders. Socure’s Sigma Synthetic Fraud offering – now available via a single API within Socure’s ID+ product suite – achieves auto-capture of greater than 90 per cent of synthetic ID fraud in just the riskiest three per cent of users, for both new users and scrubbing existing customer portfolios.
“As the volume of online applications for accounts and credit has grown exponentially, so has the opportunity for synthetic ID fraudsters who prey on the fact that this is a victimless crime and is extremely difficult to detect using conventional, 3rd party fraud detection methods,” said Johnny Ayers, Founder & Chief Product Officer of Socure. “Tackling this growing problem requires a highly sophisticated machine-learning platform that is easy to deploy, easy to manage, and a team that is maniacally focused in eradicating identity fraud from the internet.”
To deliver Sigma Synthetic Fraud, Socure invested heavily in feature engineering and data source analysis, and used both supervised and unsupervised machine learning models to derive a common definition of synthetic identity fraud across many of its largest banks, lenders and issuers. Socure leveraged this definition to develop classification models that have proven effective in combating this elusive type of fraud. Sigma Synthetic has achieved AUCs up to 97.44 per cent with auto-capture rates as high as 91 per cent in the top three per cent of riskiest users. Customers of Sigma Synthetic receive the benefit of leveraging insights from Socure’s multi-industry customer network as well as feedback file data sets, which continuously enhance the synthetic ID fraud engine, resulting in the most accurate and performant solution on the market.
Socure offers Sigma Synthetic Fraud both standalone and as part of the integrated ID+ platform alongside Sigma Identity Fraud, KYC Compliance, DocV, and more. This new module effectively manages synthetic fraud risk without degrading customer experience, slowing down growth, or turning a blind eye to potentially risky, sleeper accounts in a company’s current portfolio of customers. When applied at account enrollment or to an existing customer portfolio, Sigma Synthetic quickly assesses the risk of synthetic ID fraud and provides a decision in a fraction of a second. Sigma Synthetic and all Socure products are accessible via one single API that powers the entire Socure ID+ platform.
“Our clients must address synthetic ID fraud for a number of reasons, but most importantly to avoid the typically severe monetary losses from unpaid balances often in the tens of thousands of dollars per incident,” Mr. Ayers said.
Sigma Synthetic Fraud use cases and verticals include credit cards, lending applications, telecommunications, healthcare, insurance, eCommerce and marketplaces, government benefits, and real estate.
The launch of Sigma Synthetic corresponds with Socure’s participation in Money 20/20’s MoneyFest event. Socure is the Official Live Streaming Partner for MoneyFest, and CEO Tom Thimot is scheduled to speak to attendees on the opening and closing days of the event. The company will also be participating in a number of thought-leadership sessions including, “Top COVID era fraud trends,” to be presented by Socure’s SVP of data science, Pablo Abreu, on October 27 at 12:05 p.m. EDT and “FinTech innovation: What will fuel the evolution?,” led by Mr. Ayers on October 28 at 12:20 p.m. EDT.
For additional resources on synthetic ID fraud, Socure’s whitepaper “Anatomy of an Alleged PPP Synthetic Identity Fraud Deal” can be downloaded here.
