US driver’s license verification is now powered by AAMVA DLDV 3.0

With the rise of agentic fraud, fraudsters are now increasingly using Gen AI tools to create synthetic identities to impersonate people online. With high-risk business transactions like bank account opening, verifying a person’s identity is more important than ever.

The most common way to verify identities is to cross-check driver’s license information against the state’s issuing authority. The American Association of Motor Vehicle Administrators (AAMVA) works with US state DMVs and manages the Driver's License Data Verification (DLDV) service for this purpose.

This service helps catch synthetic identities by validating driver’s license information against the state DMV record. For years, these government data checks only returned a pass or fail result. A mismatched detail, such as a slight deviation in name or address, resulted in the same response as a genuinely incorrect value. Businesses regularly declined good applicants based on that lack of distinction.

But no more. AAMVA recently announced it has launched Driver’s License Data Verification (DLDV) 3.0, which brings additional context to government data checks.

This post covers what's new: inconclusive answers separated from failures, address matching and better name matching, a new signal confirming driving privilege, and correct handling of licenses issued without an expiration date. It also explains how Incode customers can take advantage of these improvements today through the GovMatch suite.

Improvements made by DLDV 3.0

Government data checks now return more context rather than a binary pass or fail result. Here’s what has changed.

Inconclusive answers are separated from failures

Where the state holds no record for an attribute, or can't answer, the response now says so, and a missing attribute no longer has to mean a decline. This is the change most likely to improve your true positive match rates.

Name and address matching improves

Addresses used to be all or nothing. If the state's record said Street and your applicant typed St., that was a mismatch. Now they read as the same address, along with Ave and Avenue, and N and North. A misspelled street name is forgiven too, so Gruve Street still finds Grove Street. House and unit numbers are the one exception. Those still have to match exactly.

Names match better, too. For example, Stephen against a record reading Steven won’t automatically fail. That tolerance extends to transposed letters (Jonhson → Johnson), missing letters (Sara → Sarah), maiden names on state records, and alternate names the DMV recognizes, so verification succeeds even when names don't line up perfectly.

A new signal confirms someone is allowed to drive

A suspended license is still a genuine document that passes every check, yet the holder can't legally drive. That distinction now comes back in the response.

Jurisdiction coverage for this signal is still expanding, so check availability for the states you onboard in.

Credentials with no expiration date are read correctly

Some states stop putting expiration dates on licenses past a certain age. The previous service read those as missing or malformed values. They now come back identified as what they are: valid credentials with no expiration date.

How Incode customers can access DLDV 3.0

Incode customers have access to DLDV 3.0 via GovDataMatch, an offering within the GovMatch suite:

• GovMatch: Incode's suite of identity verification solutions built on government records, the hardest source for a fraudster to fake.
• GovFaceMatch: Part of GovMatch. Matches a live selfie directly against the DMV's biometric record.
• GovDataMatch: Part of GovMatch. Matches the personal details someone submits directly against the DMV's data record, with no selfie required. This is now powered by DLDV 3.0.

Why this matters for your business

If your rules treat every government non-match as a decline, you're losing legitimate users to formatting differences and fields the state simply doesn't hold.

Nobody could see the difference before. Now you can.

Two limitations are worth knowing. A nickname only matches where the agency holds it, so Robert doesn’t become Bob automatically. And driving privilege confirms someone can legally drive a standard passenger vehicle; violations, points, and commercial licenses are outside its scope.

Several benefits with no integration work

With DLDV 3.0, government data checks now include context that can lift your approval rates. Existing GovDataMatch customers will see these changes applied by default in the coming weeks. If you want the updates sooner, contact your account team.

Ready to see how GovDataMatch and identity verification run together in a single onboarding session? Request a demo today.