Synthetic Identity Fraud Is Rapidly Rising: More Than 1 in 10 Frauds Use Invented Identities

# What happened

# Why this matters

Traditional fraud defences focus on stolen credentials and downstream detection. Synthetic identities undermine that approach because they often pass initial onboarding checks. There may be no real victim to trigger alarms, so detection that relies on complaints or downstream anomalies comes too late or becomes expensive.

# How synthetic identities slip past controls

The article highlights three specific weaknesses:

• Point-in-time checks are blind to long-term data provenance. They can't see whether an identity has a real, messy history.
• Third-party models miss many synthetic cases. The cited research found 85% of synthetic identities examined were not flagged by those models.
• Siloed tooling creates seams and blind spots. A synthetic profile can look clean in one system and be accepted, creating exposure across the business.

A quoted observer, Kimberly Sutherland of LexisNexis Risk Solutions, stresses that deepfakes complicate verification and that even small gaps in end-to-end controls act like open windows for fraudsters.

# Signs that an identity may be synthetic

The article suggests looking beyond single-point validation to assess trust. Practical indicators include:

• Unusually clean or "too perfect" data footprints with limited variation.
• Unrealistic density of corroborating data that lacks the normal imperfections and changes found in real identities (for example, prolonged consistency in address history without noise).
• Lack of behavioral history or long-running transaction patterns that are typical of genuine customers.

Research referenced also notes that over half of synthetic identities examined had credit scores above 650, meaning credit-risk signals alone can be misleading.

# Where risk sits now

Exposure has shifted upstream — onboarding is the primary attack surface. When a manufactured identity is accepted at account opening, the fraudster can use it to make withdrawals, launder proceeds, exploit new-customer incentives or create linked accounts that facilitate further abuse.

# What to do instead

• Assess attribute coherence: do name, address, phone, device and other elements fit a plausible life story?
• Check corroboration across sources and over time rather than relying on a single snapshot.
• Add behavioural and design signals that reveal whether an identity shows patterns consistent with synthetic construction or manipulation.
• Ensure capture and liveness checks are part of an end-to-end flow so that gaps can't be exploited.

# Bottom line