tools
Ocrolus Detect
Ocrolus Detect checks bank statements, pay stubs, and W-2s for tampering, anomalies, and document-origin inconsistencies.
Ocrolus Detect analyzes supported financial documents for digital manipulation, metadata anomalies, altered values, suspicious layouts, screenshots, and synthetic-document indicators. It combines forensic analysis with Ocrolus algorithmic checks.
The product returns an Authenticity Score, an Authenticity Status, structured reason codes, evidence, and visual overlays through the Ocrolus Dashboard or API. Fraud analysts, underwriters, and lending operations teams use it during document review. Detect currently processes bank statements, pay stubs, and W-2s; pricing varies by document type, volume, extraction needs, and contract terms.
Features
- Detects metadata anomalies, digital modifications, suspicious PDF structures, and tampering indicators
- Returns an Authenticity Score ranging from 0 to 100
- Provides High, Medium, and Low Authenticity Status classifications
- Generates structured reason codes with confidence and human-readable explanations
- Highlights suspicious document regions with visualization overlays
- Detects screenshots, AI-generated artifacts, altered text, and re-rendered files
- Provides Book-level and document-level fraud signal APIs
- Supports webhooks for fraud signals found, not found, or unable to process
Use cases
- Review borrower bank statements, pay stubs, and W-2s for possible manipulation
- Route documents automatically for approval, manual review, or escalation
- Investigate altered transaction amounts, balances, tax values, or account details
- Integrate fraud signals into underwriting and internal review applications
- Audit flagged documents using reason codes, evidence, and visual overlays