Synthetic Identity Fraud in Fintech: How to Catch a Customer Who Doesn't Exist

The most dangerous account in a fintech portfolio rarely looks risky. It passes onboarding, pays on time, requests a credit-line increase after six months, and gets one. Then it maxes out every product it has and disappears. There's no chargeback dispute, no angry customer, no identity-theft report. Collections can't find the borrower because the borrower never existed.

That's synthetic identity fraud, and it's now one of the costliest problems in fintech. TransUnion put U.S. lenders' exposure to suspected synthetic identities at $3.3 billion at the end of 2024, an all-time high. The Deloitte Center for Financial Services expects synthetic identity fraud to drive at least $23 billion in losses by 2030. Most of those losses are first written off as bad debt, so they never appear on a fraud report.

This guide covers how synthetic identities are built, why fintechs are a favorite target, which signals actually work for synthetic ID detection in payments, and how to layer controls so a fabricated customer is caught at onboarding instead of at write-off.

What Synthetic Identity Fraud Is (and Why It Isn't Identity Theft)

The Federal Reserve defines synthetic identity fraud as "the use of a combination of personally identifiable information (PII) to fabricate a person or entity in order to commit a dishonest act for personal or financial gain." In practice, a fraudster takes one real data point, usually a Social Security number, and adds an invented name, date of birth, address, phone number, and email.

Synthetic identities usually come in one of two forms:

• Manipulated identities. A real person's details are slightly altered, such as a changed date of birth or a transposed SSN digit, so the record no longer matches the original owner but still clears loose verification.
• Manufactured identities. An identity built almost entirely from scratch, often around an SSN that has never been used for credit. SSNs belonging to children, recent immigrants and deceased people are common choices because nobody is watching those credit files.

This is where synthetic fraud differs from traditional fintech identity fraud:

Because no real customer exists to file a complaint, synthetic identity fraud removes the feedback loop most fraud programs depend on. That's why it is so often misclassified.

The Synthetic Identity Lifecycle: Build, Cultivate, Bust Out

Synthetic fraud follows a predictable path, and each stage gives you a different chance to catch it.

1. Build. The fraudster assembles PII and applies for credit. The first application is usually declined, but the inquiry itself creates a credit file at the bureau. The "person" now exists on paper.
2. Establish. The identity opens low-risk products: a secured card, a credit-builder loan, a prepaid or neobank account. Fintechs that market to thin-file and credit-invisible customers are exactly where these first approvals tend to land.
3. Cultivate. For months, sometimes years, the identity behaves like an ideal customer. It makes small purchases and on-time payments. Fraud rings speed this up through piggybacking: adding the synthetic as an authorized user on a real account with strong history so its score jumps. Deloitte found more than half of synthetic identities carry credit scores above 650.
4. Bust out. Once limits are high enough, the identity draws down everything at once, including credit lines, BNPL, cash advances, and instant payouts. Sometimes it also pays with a bad check or reversible transfer to free up limit one more time before disappearing.

‍

There's also a growing non-credit use: synthetic identities opening money mule accounts at digital banks and payment apps, used to receive and move proceeds from scams and other fraud. For these accounts, the loss shows up as regulatory and reputational exposure instead of bad debt.

Synthetic Identity Fraud by the Numbers

The payments figure matters most for fintech operators. In AU10TIX's Q1 2026 data, payments companies had a confirmed fraud rate more than double that of banks (5.37% vs. 2.11%). Speed of onboarding and speed of money movement are exactly what synthetic fraudsters exploit.

Why Fintech Is the Softer Target

Traditional banks aren't immune, but fintech business models give synthetic identities several advantages:

• Onboarding is built for speed. Approval in minutes is a growth feature, and every verification step removed for conversion is a step a fabricated identity doesn't have to pass.
• Thin-file customers are the target market. Credit-builder products, secured cards and "no credit history required" lending serve real underserved consumers. On paper, a brand-new synthetic identity looks exactly like them.
• Money moves instantly. Instant payouts, real-time transfers and fast settlement shorten the time between a bust-out and an irreversible loss.
• Products are stacked. One identity can open a card, a BNPL line, a cash advance and a wallet across several apps at once, and those apps don't share data with one another.
• Generative AI makes fabrication cheaper. AI tools now produce convincing ID documents, selfies and synthetic voices at scale. Datos Insights found 40% of financial institutions have already seen higher attack rates linked to AI. AU10TIX's CEO described attackers generating identities "from scratch" and deploying them across many institutions simultaneously.

High-risk and regulated verticals face the same pressure from a different direction. Gaming, crypto, telehealth, and marketplaces that pay out to users are attractive cash-out points for synthetic identities, and they carry stricter KYC and AML expectations from processors and regulators.

Synthetic ID Detection in Payments: The Signals That Give It Away

No single check catches a well-built synthetic identity. Each fabricated element is internally consistent; the inconsistencies only appear when you compare data points against each other and against the wider network. These are the signals worth building around:

Quick check: would your onboarding catch a synthetic identity?

→ Do you verify the SSN, name and date of birth combination against SSA records, or only that the SSN is in a valid format?

→ Can your system see when two "different" applicants share a device, phone or payout account?

→ Does your liveness check detect injected video and AI-generated faces, not just printed photos?

→ Do you re-score customers after onboarding, or does a passed KYC check stand for the life of the account?

→ Do write-offs with no contactable borrower get reviewed for synthetic patterns, or do they go straight to bad debt?

A Layered Prevention Stack for Fintech Identity Fraud

The teams that handle synthetic identity fraud well don't depend on one vendor or one checkpoint. They stack controls so each one covers the blind spots of the others.

Layer 1: Validate the data, not just the format

Start with authoritative sources. The Social Security Administration's electronic Consent Based SSN Verification (eCBSV) service confirms whether an SSN, name, and date of birth match SSA records, with the consumer's consent. It's one of the few checks a manufactured identity can't fake. Pair it with bureau-level checks for SSN issuance dates, deceased-person flags, and file age.

Layer 2: Prove a real human is present

Document verification plus liveness detection confirms that a physical person matches the ID being presented. In 2026, this layer has to defend against injection attacks, where a deepfake video is fed straight into the camera stream, and against AI-generated document templates. Checking for presentation attacks alone is no longer enough.

Layer 3: Read device and behavior signals

Device fingerprinting, IP and geolocation consistency, and behavioral biometrics (typing rhythm, navigation patterns, hesitation on fields a real person would know) expose fraud farms that run many identities from the same hardware or scripts. These signals add no friction for legitimate users.

Layer 4: Link identities across the network

Synthetic identities are rarely built one at a time. Graph analysis and consortium data connect shared phones, addresses, devices, and payout accounts across applicants and across institutions, so you can uncover whole rings instead of chasing single accounts.

Layer 5: Keep watching after onboarding

Most synthetic identities are approved, so post-onboarding monitoring matters as much as the front door. Watch for limit-increase velocity, new authorized-user additions, sudden utilization spikes, and new external funding sources, all classic signs of an approaching bust-out. Hold or step up verification on large instant payouts from accounts showing these patterns.

Layer 6: Classify losses correctly

If you can't measure synthetic fraud, you can't fund the fight against it. Review charged-off accounts with no contactable borrower, no right-party contact, and clean pre-default behavior for synthetic markers, and move confirmed cases out of credit loss and into fraud. The Federal Reserve's Synthetic Identity Fraud Mitigation Toolkit gives a shared framework for that classification.

Balancing Friction and Conversion

The usual objection is conversion: every check costs approvals. Risk-based orchestration avoids that trade-off. Low-risk applicants whose data validates cleanly, whose device is consistent, and whose digital footprint is established go straight through. Additional steps are reserved for applicants whose signals don't add up. Applied that way, synthetic ID detection in payments makes onboarding faster for real customers and slower only for fabricated ones.

Where Payment Infrastructure Fits

Identity checks at sign-up are only half the defense. The other half sits in the payment layer, where synthetic identities eventually try to move money. A payment partner built for regulated and high-risk industries treats KYC and AML as part of every transaction, not a one-time onboarding step.

Approvely builds AML/KYC underwriting into its payment infrastructure, alongside fraud prevention and chargeback protection, for verticals such as gaming, iGaming, crypto, telehealth and marketplaces. These are businesses where fast payouts and elevated fraud exposure make synthetic identities especially costly, and where a processor that understands the risk profile is worth more than one that freezes the account at the first spike.

FAQ: Synthetic Identity Fraud in Fintech

What is synthetic identity fraud?

Synthetic identity fraud is the creation of a fake identity by combining real personal data, usually a Social Security number, with invented details such as a name, date of birth and contact information. The fabricated identity is used to open accounts, build credit and eventually take out as much money as possible before disappearing.

How is synthetic identity fraud different from identity theft?

Identity theft uses a real person's complete identity, and that person usually notices and reports it. Synthetic identity fraud creates a new "person" who doesn't exist, so there's no victim monitoring the account. That's why it can run undetected for months or years and is often booked as bad debt instead of fraud.

Why are fintechs especially exposed to synthetic identity fraud?

Fintechs often prioritize fast digital onboarding, serve thin-file customers who look similar to new synthetic identities, and offer instant money movement through payouts and real-time transfers. Together, these make it easier for a fabricated identity to get approved and to cash out quickly.

What are the best methods for synthetic ID detection in payments?

The most effective approach combines SSN validation through SSA's eCBSV service, document and liveness checks that resist deepfake injection, device and behavioral analytics, cross-applicant link analysis, and ongoing transaction monitoring for bust-out patterns. No single method is enough on its own.

Can AI-generated identities pass KYC?

Increasingly, yes, if KYC relies on document images and basic selfie checks alone. In AU10TIX's Q1 2026 data, synthetic-pattern attacks made up nearly half of confirmed identity-verification fraud. Modern KYC needs injection-attack detection, data validation against authoritative sources, and network-level linking to keep up.

Does Approvely help with fraud prevention for high-risk fintech and payments businesses?

Yes. Approvely's payment infrastructure includes built-in fraud prevention, chargeback protection, and AML/KYC underwriting for high-risk and regulated industries, including gaming, iGaming, crypto, telehealth, and marketplaces.

‍