Stopping AI-Created Synthetic Identities Starts with Email Verification

Before biometrics, before ai: trust still begins with the data. Synthetic identities thrive on the borrowed, but email address intelligence exposes the history they can’t fake.

Before there were biometrics, before AI-powered fraud detection, there was the inbox. The email address was the first true digital identity marker, tethering people to the online world. Today, synthetic identities built by artificial intelligence are testing every assumption we’ve made about trust. Yet the irony is this: the most advanced fraud begins in the same place it always has, with simple datapoints.

The latest Cifas report Fraudscape 2025 Six Month Update reveals a deeper truth: fraud isn’t just multiplying, it’s mutating. Identity fraud filings still top the charts, with over 118,000 cases recorded in the first half of this year. And increasingly, those cases are fueled by high-quality fakes: synthetic identities engineered by AI to look indistinguishable from the real thing.

“We’re watching fraud change in real time,” notes Diarmuid Thoma, Head of Fraud and Data Strategy from AtData, the leader in email address intelligence. “And AI has given fraudsters something they’ve never had before: the ability to generate identities with the polish of legitimacy, and to escalate at a pace defenses can’t keep up with.”

The tactics may be evolving, but the signals that expose the lie haven’t.

Why Email Is the First Red Flag

Every synthetic identity needs a front door. For years, that entry point has been the email address. It’s one of the most consistent credentials required to apply for loans, open a bank account, claim promotions, or access just about any account. Fraudsters know this. That’s why disposable, recycled, and freshly created inboxes are the scaffolding of synthetic identity at scale.

The Cifas report highlights how criminals pivot between tactics, shifting from communications-based impersonations to account takeovers in mobile products. But whether fraud manifests as a takeover, a false application, or misuse of a facility, it often traces back to an email address that doesn’t behave like a real person.

Think about it: a synthetic identity may pass credit bureau checks and present convincing documentation, but the history of the email supporting it tells another story. Is it brand new, created yesterday to apply for a credit card? Is it tied to a cluster of high-risk activity across multiple applications? Has it been sitting dormant only to come alive and set up hundreds of accounts across the web? Or does it show years of consistent use, tied to real engagement and predictable human patterns?

The Synthetic Identity Problem Is Really a Data Integrity Problem

Cifas reported that fraud now accounts for 41% of all crime, with AI amplifying the scale and the believability. What that statistic doesn’t bring to light, however, is that many of these so-called “identities” aren’t identities at all. They’re constructs — data stitched together by algorithms with no human on the other side.

That’s why email address intelligence matters so profoundly. Unlike static data points, email addresses carry history. They accrue trust over time, exposing whether the inbox belongs to a person navigating daily life or to a botnet generating accounts by the thousands.

AtData’s approach utilizes an advanced email signal network to surface behavior and history: the age of the email, whether it has legitimate activity, and if it maps to a network of suspicious sign-ups. This kind of intelligence is what separates a living, breathing customer from an AI-generated placeholder.

What the Future Requires

What we know, and the Fraudscape report makes clear, is that fraud is dynamic. Identity fraud may have dipped 7% year-over-year, but only because criminals shifted to other avenues. The problem isn’t going away; it’s mutating. AI is democratizing fraud, making synthetic identities routine in application workflows.

That means the future of fraud prevention will hinge on building systemic resilience, starting at the point of entry. If an email address doesn’t carry the digital footprint of a real person, the application should never advance.

It also means collaboration is non-negotiable. Fraudsters share playbooks across borders and industries. Defenders must do the same. The urgent call from Cifas for cross-sector intelligence sharing is one AtData echoes: the only way to expose scalable fraud is to scale our trust signals just as widely.

From Illusion to Integrity

Synthetic identities are, at their core, illusions. They thrive when businesses equate “submitted data” with “trustworthy identity.” But illusions collapse under scrutiny, if you know where to look.

Email address intelligence is a magnifying glass. It’s not glamorous, but it is the first, most pragmatic line of defense against an AI-enabled fraud economy. By treating the email address as the anchor of digital identity, we move closer to a future where fraud doesn’t masquerade as growth.

Stopping synthetic identities doesn’t start with AI detection arms races or futuristic biometrics. It starts with an inbox. It starts with data integrity. And it starts now.

Learn How AtData Stops Synthetic Fraud