-

Extend Introduces Verify, a Document Fraud Detection API that Builds Trust into Document Pipelines

Verify detects signs of document tampering and AI-generated fraud, helping businesses trust the documents their AI agents act on and deploy those agents confidently in the real economy.

LAS VEGAS--(BUSINESS WIRE)--Extend today at Money20/20 USA announced Extend Verify, a document fraud detection API that helps developers and AI agents assess the integrity of documents before using them to make decisions.

"Businesses want to move beyond agents that search documents to agents that help get work done in the real economy,” said Kushal Byatnal, CEO of Extend. “That requires trust in the document evidence those agents act on."

Share

Much of the work businesses want AI agents to perform begins with a document:

  • A bank statement supports a lending decision
  • An invoice initiates a payment
  • Receipt substantiates an insurance claim

Today, companies rely on teams of human reviewers to catch document fraud. But AI has made forgeries that are imperceptible to the human eye extremely easy to produce. When fraud escapes review to inform a loan, insurance claim, or background check approval, it can cost a business millions of dollars.

Giving agents greater autonomy means enabling them to cross that trust barrier. With Extend Verify, we’re expanding our all-in-one document processing platform to assess document integrity, so your agents can flag suspicious inputs before acting on them.

“Businesses want to move beyond agents that search documents to agents that help get work done in the real economy,” said Kushal Byatnal, CEO of Extend. “That requires trust in the evidence those agents act on. Verify helps developers build document integrity into their pipelines to prevent fraud-related losses, mitigate the difficulty of human review, and scale up agent deployments.”

Document integrity built into the pipeline

Extend Verify examines visual content, file structure, and revision history, combining specialized models, forensic checks, and agentic analysis to identify suspicious signals. It checks for signs of:

  • AI-generated documents: Fabricated documents designed to resemble legitimate records.
  • Localized AI-tampering: Changes to specific fields or regions within an otherwise genuine document, down to single character edits.
  • Conventional digital editing: Patched text, image overlays, and other alterations made with PDF editing software.
  • Post-signature changes: Evidence of changes made after a document was digitally signed.

Verify returns detected signals, risk assessments and scores, explanations, and traceable evidence. Businesses can use those findings to determine which documents require further review.

Verify performance

Document fraud detection needs to catch suspicious inputs without overwhelming teams with alerts on legitimate documents. In a challenging benchmark evaluation, Verify accurately detected 96.3% of fraudulent documents at a 4.0% false-positive rate.

The evaluation measures both how often Verify identifies fraudulent documents and how often it flags legitimate ones. These measures help businesses assess the tradeoff between detecting potential fraud and the volume of documents requiring further review.

For a lender, Verify could flag suspicious changes to an income document before it informs a credit decision. For a payments team, it could surface an altered invoice before payment is authorized. For an insurer, it could identify a potentially fabricated receipt before a claim is settled.

See Verify at Money20/20 USA

Visit Extend at Booth 6085 for a demonstration of Verify and to play Fraud Arcade, an interactive game that challenges attendees to distinguish legitimate documents from manipulated ones.

Availability

Developers can use Verify independently via API or add it to pipelines that already parse, split, and extract documents with Extend. Bringing document processing and integrity checks together gives agents and applications both structured information and evidence about the source it came from.

Extend Verify is available today. Learn more and request access.

About Extend

Extend is a Document AI platform that provides models, APIs, and infrastructure to enable developers and agents to make every document process programmatic. Its all-in-one platform delivers models, tooling, and APIs to parse, extract, split, edit, run evals, and build end-to-end pipelines to turn unstructured documents into structured data that agents can reason over and act on. Companies including Zillow, Mercury, Opendoor, Brex, Amgen, First American Financial, C.H. Robinson, Flatiron Health, and more use Extend to power document ingestion in production.

Contacts

Media contact
Jing Reyhan
Extend
pr@extend.ai

Extend


Release Versions

Contacts

Media contact
Jing Reyhan
Extend
pr@extend.ai

Social Media Profiles
More News From Extend

Extend Launches Operator-1, an Autonomous Agent for Complex Document Extraction

NEW YORK--(BUSINESS WIRE)--Extend announced Operator-1, an autonomous document extraction agent that delivers state-of-the-art (SotA) accuracy on the most challenging extraction tasks. Operator-1 combines document processing tools with persistent memory in a dedicated workspace to extract data, validate its work, and reuse successful approaches. In evaluations, it set a new performance record across three document extraction benchmarks. Complex documents challenge extraction systems that rely o...

Extend Delivers Sub-Cent Document Parsing with Leading Cost–Accuracy Performance on Two Open-Source Benchmarks

NEW YORK--(BUSINESS WIRE)--Extend, a document processing platform for developers building AI applications and agents, today highlighted results for Light Parse on two open-source benchmarks. The generally available parsing engine achieved 90.5% field-level accuracy on RealDoc-Bench and 54.9% question-answering accuracy in Extend’s evaluation on Databricks OfficeQA Pro. At $0.00625 per page, Light Parse gives developers a cost-efficient way to process documents for AI applications. Light Parse i...
Back to Newsroom