Synthetic identity fraud is one of those problems most consumers never think about until it shows up indirectly as tighter credit, higher interest rates, or another frustrating verification step when applying for a loan. On Thursday, Equifax said it has a new answer for that problem, announcing the launch of a product called Synthetic Identity Risk that is designed to help lenders spot fake identities before the damage is done.
At a high level, synthetic identity fraud happens when criminals mix real and fake personal information to create a brand new person on paper. A real Social Security number might be paired with a made-up name, address, and date of birth. That fake person then slowly builds credit, opens accounts, and eventually walks away from the debt. Because nothing looks obviously wrong at first, these cases can sit undetected for years, quietly turning into losses that lenders have to absorb.
Equifax says this new product uses artificial intelligence and machine-learning models to look for patterns that traditional credit checks can miss. Instead of relying only on static data points, it looks at how identities behave over time, how information changes, and how accounts are used. The goal is to flag suspicious identities early, either when an account is opened or later while it is active.
The company is not shy about framing this as a financial problem first and a security problem second. According to its own data, the average charged-off loss tied to a known synthetic identity is around $13,000. Multiply that across large portfolios and the numbers add up quickly. For lenders, that loss eventually flows downhill in the form of higher costs for everyone else.
What Equifax is really selling here is timing. Traditional fraud tools often catch problems only after payments stop and the account goes bad. Synthetic Identity Risk is positioned as something that can work both at onboarding and throughout the life of an account, giving lenders a chance to intervene before a loss is locked in. That continuous-monitoring angle matters, since synthetic identities often look clean at the beginning and only reveal themselves later.
The company says the system analyzes identity data, credit history, and behavioral signals together, rather than in isolation. That sounds sensible, especially since synthetic fraud is all about passing surface-level checks. Still, as with any AI-powered product, the real question is accuracy. Flag too much and legitimate customers get blocked or delayed. Flag too little and the losses continue. Equifax did not share detailed performance metrics in the announcement, which is not unusual but does leave some open questions.
There is also a broader trend at work here. Fraud detection has become one of the most aggressive areas for AI adoption in financial services. Banks and lenders are under pressure from rising losses, stricter compliance requirements, and customers who expect instant approvals. AI promises speed and scale, but it also introduces new concerns about transparency. When an application is rejected because an algorithm thinks something looks off, lenders still need to explain that decision to regulators and customers.
Equifax seems aware of that tension and is pitching this as a way to shift from reactive cleanup to proactive prevention. The company’s leadership says the product is meant to help lenders protect legitimate customers while reducing exposure to hidden risks in their portfolios. That sounds good in theory, but the real test will be how well the system works in messy real-world data, where people move, change jobs, and sometimes look strange on paper without being criminals.
To clarify how the system is meant to be used, Equifax also shared some practical guidance. A synthetic identity, according to the company, is a fake identity created by combining pieces of real personal data such as names and addresses with fabricated information. These identities are typically used to acquire goods or services before eventually busting out, leaving lenders with losses.
Equifax says its synthetic identity checks can be delivered in multiple ways depending on a customer’s needs. The solution supports both API-based delivery and batch processing, and synthetic checks can also be delivered through ACRO and certain partner integrations. That flexibility matters for lenders running older systems that cannot easily be modernized overnight.
The company also stressed that these checks are for fraud-risk alerts only. They are not meant to determine a person’s eligibility for credit or to be used for any other purpose covered by the Fair Credit Reporting Act. In other words, this is a warning system, not a credit-decision engine.
Equifax says the alerts can also be applied retroactively. Lenders can use the system as a back-book cleanup tool to scan existing portfolios and pinpoint accounts that may have been opened using synthetic identities. For banks sitting on years of accumulated risk, that could be just as valuable as stopping new fraud at the front door.
From a market perspective, this launch shows how credit bureaus are evolving beyond raw data providers into ongoing risk platforms. For Equifax, that means deeper integration with clients and a more permanent role in lending workflows. For lenders, it means relying more heavily on external models to tell them when something feels off.
Synthetic identity fraud is not going away anytime soon. If anything, it is getting easier as more personal data leaks online and digital onboarding becomes the norm. Tools like this are likely to become standard, even if they are not perfect. The real challenge will be reducing losses without making the credit system even more frustrating for legitimate borrowers.
For now, Equifax is betting that lenders are ready to accept smarter alerts, more complexity, and more AI in exchange for fewer painful surprises down the line.
Support independent tech journalism
NERDS.xyz is independently owned and operated. If you enjoy my coverage of Linux, AI, hardware, cybersecurity, and tech culture, consider supporting the site on Ko-fi.
Support NERDS.xyz