Industries

A wrong match can cost someone a job or a home. AVAAS certifies the system that makes it.

Background screening and consumer reporting increasingly run on AI that matches records and scores people for employers, landlords, and lenders. When the match is wrong or the score is unexplained, a person loses an opportunity they earned. AVAAS certifies the system at the point it reports on someone.

FCRA §607(b)FCRA §604(b)CFPB Circular 2024-06Private right of action
Assess your screening models →
Where AI acts on a person

The decision point in screening

In background screening, the AI acts when it matches records to a person and scores their risk. That output decides whether someone gets a job or an apartment.

When an algorithmic background score is wrong, FCRA treats it as a consumer-report accuracy failure, and that carries a private right of action.

What keeps you exposed

What keeps screeners exposed

Accuracy

A name-only match flags the wrong person

FCRA section 607(b) requires maximum possible accuracy. Mismatches and expunged records are documented, costly failures.

Consumer-report status

Your AI score is a consumer report

The CFPB has confirmed algorithmic background and surveillance scores fall under FCRA. The compliance duties attach to your model.

Adverse action

A denial without proper notice

FCRA section 604(b) requires specific notice and a copy of the report. An opaque score makes that hard to satisfy.

This is already happening
CFPB Circular 2024-06 confirms that algorithmic background and worker-surveillance scores are consumer reports, so accuracy failures such as name-only matches carry FCRA liability and a private right of action.
CFPB Circular 2024-06 · Kistler v. Eightfold AI, FCRA
How AVAAS adds value

Evidence the screen is accurate and fair

Does the matching model meet the accuracy standard?

AVAAS evaluates the system for the failure modes that drive inaccuracy, with causal attribution to what produced a hit.

Can a denial be explained with the required notice?

AVAAS tests whether the system can give specific, accurate reasons suited to an adverse-action notice.

Does the score disadvantage protected groups?

Five structurally independent validators test for demographic disparity in screening outcomes.

You get documented, third-party evidence that the screening model meets the accuracy and fairness FCRA requires, produced independently of the score vendor.

Related AVAAS coverage: Employment & HR · Housing. Or run the free Regulation Checker to see what applies to you.

See where your background-screening AI creates liability.

Tell us how AI matches records or scores applicants, and we will scope an AVAAS certification to the FCRA accuracy exposure.

Ready to start now? Certify Your AI →  or  email [email protected]