Home/Guides/Detect Fake Bank Statements

Document Fraud

Fake bank statements cost lenders billions. Here's how to catch them.

Altered balances, inserted transactions, and AI-generated statements are the most common form of income fraud. Manual review catches less than 20% of sophisticated edits. TamperCheck runs 130+ forensic checks on every bank statement PDF — and returns a verdict in under 5 seconds.

The problem

Why fake bank statements are so hard to spot

Modern document fraud has moved well beyond crude Photoshop edits. Today's fake bank statements use correct institutional branding, matching fonts, and plausible transaction patterns that pass visual inspection. Some are generated entirely by AI from a text prompt.

The most dangerous fakes are partial edits — a genuine statement where only the balance figures or a few transaction rows have been changed. These are nearly impossible to catch by eye because 95% of the document is authentic.

Manual review doesn't scale. A human reviewer might catch an obvious font mismatch, but they won't notice that the running balance arithmetic is off by $12.40 in row 47, or that the PDF's internal font table lists a typeface that doesn't match what's rendered on screen.

What we detect

Forensic signals that expose fraud

Running balance arithmetic

TamperCheck verifies that every transaction row's running balance is mathematically consistent with the previous row. A single edited amount breaks the chain.

Font metrics and substitution

Edited fields often use a slightly different font weight, spacing, or rendering engine. TamperCheck compares font metrics across every text element in the document.

PDF text layer vs raster mismatch

When someone edits a PDF in a tool like Adobe Acrobat, the text layer and the rendered pixels can diverge. TamperCheck detects these mismatches at the character level.

Metadata and creation tool fingerprint

Genuine bank statements have consistent PDF producer metadata. Tampered files often show traces of editing tools (Acrobat, LibreOffice, online PDF editors) that banks don't use.

Table row insertion fingerprints

Inserted or deleted transaction rows leave structural artifacts in the PDF — inconsistent spacing, misaligned columns, or orphaned formatting objects.

AI-generation spectral analysis

Fully AI-generated statements exhibit noise patterns and compression artifacts that differ from scanner- or printer-produced documents. TamperCheck's CV layer detects these signatures.

The solution

How TamperCheck catches fake bank statements

TamperCheck analyses every bank statement across three layers — structural forensics, computer vision, and AI adjudication — and returns a plain-English verdict with specific findings your team can act on.

Digital PDFsPhone photosScanned hard copiesPhysical stampsHandwritten signatures

How it works

  • Upload via API or dashboard — single endpoint, single file, verdict in seconds
  • 130+ forensic checks run automatically on every document
  • Plain-English findings tied to specific regions of the document
  • Risk score (0–100) calibrated for document type
  • $0.50 per document — no subscriptions, no minimums
  • Zero document storage — files are processed ephemerally
  • BYOK — bring your own AI provider keys for model control

FAQ

Common questions

Related

Keep reading

See it working on your documents

Start with $5 in free credits — no contract, no card required. Upload your first document and get a verdict in seconds.