Search Icon
Back to Blog

Synthetic identity fraud — or when criminals blend genuine and falsified information to create new, fictitious identities to fraudulently apply for loans, credit, government benefits, or move illicit funds — shows no signs of slowing. In fact, the Deloitte Center for Financial Services expects synthetic identity fraud to generate at least $23 billion in losses by 2030.

To combat this pervasive threat, Socure recently launched the latest model of its Sigma Synthetic Fraud solution, which uses advanced machine learning and diverse, third-party as well as transaction and feedback data from its vast consortium to uncover patterns linked to insidious synthetic identity fraud.

How do we do this? One way is by using “proof of life” data sources to help to determine if an identity is real or fabricated. Proof of life refers to records and data trails that are difficult for a criminal to fake and that tend to accumulate naturally over time for genuine people.

The majority of best practices shared publicly on how to stop fabricated and manipulated identities relate to credit history or trade lines.

So what does a company do when tradeline data isn’t available?

Some other examples of proof of life sources include:

  • Address history spanning 5+ years
  • Email history spanning 3-5+ years
  • School records
  • Legal records like bankruptcies or liens
  • Obituaries and death records
  • Driver’s license records
  • Property ownership history

For younger applicants without decades of records due to their age, Socure looks for proof of life such as:

  • School enrollment history
  • Social media accounts with long-term, consistent activity
  • Sports team rosters and competition records
  • Driver’s license records

A (Genuine) Life Well Lived

The logic is that real people naturally generate this type of data over time as they go through life. While the fraudsters creating synthetic identities may sign up for retail loyalty cards or utility accounts to appear real, they can’t necessarily fabricate robust, lifelong records such as a driver’s license or school records.

When evaluating a potentially suspicious application, Socure’s fraud investigation team first confirms that the core personal details like name, date of birth, and Social Security Number, correlate to the presented identity. Then we look for proof of life connected to that identity, such as long-term address history spanning 5+ years. If they find it, it’s likely to be a real identity. And what about the new-to-country population who doesn’t have a Social Security number? Socure can determine fabricated identities versus true identities even if we can’t find proof of life sources that match ID names, DOB, and location.

On the other hand, thin credit files only active for a year or two — and especially those new-to -country — raise red flags, as do address and email histories that only go back to 2018 or 2020. This fits the pattern of synthetic identities that criminals invent for temporary use before moving on to the next scam.

Proof of life data helps to establish trust for younger and, especially, immigrant demographics who would otherwise appear to look like synthetic identities. Because proof of life sources are incorporated into the latest Sigma Synthetic Fraud model, it enables customers to approve more people in this group as the good customers they are.

Connecting the Dots of Synthetic Identities

In short, we all generate a trail of digital breadcrumbs as we move through life. While no single record constitutes definitive proof of life on its own, the accumulated weight of evidence over 5, 10, or 20 years makes it increasingly difficult for even skilled fraudsters to fake genuine, lifelong identities. By analyzing historical proof of life records layered with other identity checks, Socure can help our customers stay a step ahead of synthetic identity fraud.

Learn more about Sigma Synthetic Fraud v4 here.

Bre Reimer

Bre Reimer works as a fraud investigations manager at Socure, specializing in detecting identity and synthetic fraud.