logo
Free PDF Check

AI-Generated Document Detection: What the PDF File Shows

Built for fraud ops at lending, insurance & compliance teams

AI now writes convincing fake pay stubs, bank statements, and invoices on demand – perfect on screen, indistinguishable from real exports to the human eye. HTPBE? doesn’t grade the prose. It reads the PDF’s binary structure and flags the seams where AI-generated content was rasterized, stitched, or stamped into the file.

~3 sec
per document
63
forensic checks per document
1,500+
docs / month on Growth
Plans
subscriptions or one-off packs – see pricing
Scope

HTPBE? is a structural PDF forensics tool, not an AI-content classifier. We detect what happens to a PDF when AI output is assembled into it – edits after a real export, scanned or rasterized pages mixed with inserted content, documents reworked in online editors or design tools, documents built from a reused template, modifications after a digital signature.

What we do not promise: reading a PDF and saying ‘this was written by ChatGPT.’ If a fraudster invents a document outright, with a stolen logo and a fabricated IBAN, there is no editing history for a structural tool to find – including ours. That is content forgery, a different product category (Inscribe, Resistant AI, GPTZero). We sit alongside content-AI detectors as the structural layer they don’t analyze.

The problem

AI didn’t invent document fraud. It scaled the assembly line

Until 2023, fabricating a convincing pay stub or bank statement required either a real export to edit or a designer’s hour. Generative AI collapsed that to seconds. A fraud-ops team that saw a dozen forged documents a quarter now sees them weekly – and the rendered output is cleaner than what a human editor produced.

But the rendering layer is not where forensic evidence lives. AI generates pixels; a PDF is a structured container. To put AI-generated content into a PDF that fools a human, the fraudster has to either rework a real institutional file or stitch AI output into a real export. Both leave structural evidence in the file, and that is what HTPBE? examines.

For risk and fraud-ops teams, the AI-fraud question isn’t ‘was this written by a model?’ – that’s a content question. The operative question is ‘was this PDF assembled by a human or a tool, after the real document was generated?’ That’s the structural question, and it’s answerable in under three seconds per file.

AI-fraud patterns we actually catch

  • AI-generated text printed and rebuilt as PDF in an online editor
  • AI-rasterized pages swapped into a real bank statement export
  • Design-tool rebuild of an institutional template (Illustrator, InDesign)
  • Mass template forgery – same fraud kit run repeatedly through AI
  • AI-generated overlay applied to a stolen template via Photoshop
  • AI modifications appended to an already-signed document

What this looks like

Document fraud in 2026 – three concrete patterns

Three real fraud mechanics we catch at the structural PDF layer.

01

AI-generated text printed and rebuilt as PDF in an online editor

02

AI-rasterized pages swapped into a real bank statement export

03

Design-tool rebuild of an institutional template (Illustrator, InDesign)

04

Mass template forgery – same fraud kit run repeatedly through AI

05

AI-generated overlay applied to a stolen template via Photoshop

06

AI modifications appended to an already-signed document

63
Forensic checks per document
~3 sec
Median analysis time, end to end

The detection gap

KYC platforms check the document. HTPBE? checks the file.

Two different checks – both matter.

KYC & identity platforms

Plaid · Persona · Alloy · Jumio

  • Is this a real bank statement template?
  • Does the account number match the identity?
  • Is the document format consistent with the issuing bank?

Detects fake documents. Does not detect edited real documents.

HTPBE? tamper detection API

Structural PDF integrity

  • Was this specific PDF file modified after it was generated?
  • Does the file carry evidence of editing after export?
  • Were digital signatures valid at the time of signing?

Catches edits invisible to visual review and template checks.

Results in under 3 seconds30 to 1,500+ documents/month

What HTPBE? checks

What the API detects in AI-assembled PDFs

Six structural patterns that betray AI assembly – results in under 3 seconds

Scans mixed with inserted content

When AI-generated images replace pages in a real PDF, HTPBE? detects the scanned or rasterized page inserted into an otherwise digital document.

Page assembly from different sources

HTPBE? detects documents assembled from pages of different sources. A bank statement where the first pages are a real export and a later page was rebuilt with AI is flagged.

Edits after the real export

When AI content is pasted into a real PDF and saved again, HTPBE? reports that the file was edited after it was created – whoever or whatever wrote the pasted text.

Editing-tool traces

AI-generated content rarely lands in a PDF directly – it goes through a tool. When a file carries the fingerprints of a known PDF editing tool that worked on it after it was first produced, HTPBE? reports this as a named marker.

Design-tool rebuilds

HTPBE? flags institutional templates rebuilt in Illustrator, Photoshop or InDesign – a document that should have come straight from a banking system but was produced in a design workflow instead.

Modifications after digital signature

AI-generated content added to an already-signed contract or institutional document triggers the highest-confidence verdict in the system: ‘certain’. HTPBE? detects edits made after signing and signatures that were removed.

Share with engineering

Wire this into your intake pipeline in under a day

Two API calls – one POST to submit the PDF, one GET to retrieve the verdict. Forward this page to your engineering team; the full API reference, quotas, and copy-paste examples in cURL, JavaScript, Python, PHP, Go, and Ruby are one click away.

Full API reference for engineering

Pricing

Self-serve plans, no sales call

All plans include the same forensic checks. Pick the quota that matches your monthly document volume.

manual

Starter

30 credits/mo

Manual spot-checks and integration testing

most common

Growth

350 credits/mo

Active document processing pipelines

high volume

Pro

1,500 credits/mo

High-volume automation and API integrations

Enterprise (unlimited, on-premise available) — see full pricing

Secure your workflowView API docs

API key on signup. Free test environment on every plan. No card required.

Customer Stories

Teams that stopped document fraud

Compliance, finance, and risk teams use HTPBE? to catch manipulated PDFs before they become costly mistakes.

Caught an invoice where the total had been changed by less than a thousand dollars. Without this I would have approved it without a second look.

Sarah M.

AP Manager

United States

We had three applicants in the same week with bank statements that looked completely fine. Two of them were flagged as modified. You simply cannot see this by reading the document – it is in the file structure.

Lars V.

Risk Analyst, Online Lending

Netherlands

Salary slips were coming with altered figures. We identified two problematic files before the placement was finalised.

Priya K.

HR Operations Lead

India

Since we started checking documents this way, we stopped two applications early in the process that would have been very difficult to reverse later.

Julien R.

Fraud Analyst, Fintech

France

Some applicants were sending PDFs that looked authentic but had been edited in ways not visible to the eye. We now ask for checked originals when something is flagged. Already saved us from a few bad decisions.

Marta S.

Compliance Coordinator

Spain

One invoice was caught because there was a mismatch between the document dates and structure. That particular case would have cost us significantly.

Tariq A.

Finance Manager

United Arab Emirates

FAQ

Frequently asked questions

Does HTPBE? detect text written by ChatGPT or Claude?

No, and that’s deliberate. Text classifiers (GPTZero, Originality.ai, Inscribe) read the prose and decide whether it looks AI-written. We read the PDF’s structure and edit history and decide whether the file was assembled normally or stitched together. The two are complementary layers of an AI-fraud defense, not substitutes. Run text classifiers on the content; run HTPBE? on the file.

What about a fully fabricated PDF generated from scratch by an AI agent?

‘Born-synthetic’ PDFs are out of structural scope. A document invented outright, rather than edited after it was issued, has no editing history for a structural tool to find – including ours. That’s a content-forgery problem (checking that the account, the employer and the branding are real), and the right tools are content-AI detectors and KYC platforms. Our scope is post-creation tampering and assembly artifacts.

How is AI document fraud different from traditional document fraud?

In structural terms, less than you’d expect. The same forensic markers fire: revisions saved after creation, fingerprints of editing software, pages assembled from more than one source, image-based content mixed into a digital page. AI changed the volume and polish of fraudulent documents, not the underlying way they get built. A pay stub with an AI-rasterized fake amount triggers the same markers as a Photoshop-edited pay stub from 2018.

Does this replace content-AI detection tools like GPTZero or Inscribe?

No. They check the prose; we check the container. A bank statement that fools GPTZero on text can still betray itself in its file structure; a PDF that’s structurally pristine can still be a born-synthetic forgery their tools would catch. AI-fraud defense is a stack – identity layer (Persona, Onfido, Plaid), content layer (GPTZero, Inscribe, Resistant AI), structural layer (HTPBE?). Run all three on the same upload.

Won’t this trigger false positives on documents legitimately processed by AI tools?

It depends on what ‘processed by AI’ means. A bank exporting a real PDF and then routing it through an AI-OCR pipeline before sending to the customer would carry post-export markers – but banks don’t do that, they ship the original export. The patterns we flag – scans mixed with inserted content, rework in an online editor, pages from different sources – are not part of normal institutional document workflows. If you see a result you believe is wrong, the named markers tell you exactly what was found, so your reviewer can check it.

Can I test it on AI-generated PDFs we’ve already collected?

Yes. The web tool accepts any PDF up to 10 MB – upload your collected AI-fraud samples and see the verdict and named markers per file. New accounts get 5 free credits on signup; after that, top up with credit packs or a subscription – see pricing. For automated pipelines, the API uses the same engine and the same shared credit pool. Fraud-ops teams typically run a backfill of recent flagged applications through HTPBE? as a first-week exercise to size the AI-fraud baseline before integrating into the live queue.

What does an AI-made PDF look like at file level compared with a genuine export?

Often it looks like nothing special, and that is the point to understand. A document from a payroll system or a bank carries the traces of that system and looks like it came from where it says it did. A PDF that passed through other tools afterwards – an online editor, a design suite, a page swapped in from elsewhere – carries traces of those tools instead. HTPBE? reports which of these traces are present. A file generated from scratch in one clean pass can look like any other single-pass file, which is why a result is never a statement about who or what wrote the content.

What can the verdict say about a document that might be AI-generated?

It can say that the file was modified after it was created (modified), that no modification was found (intact), or that the file’s origin does not allow a conclusion (inconclusive), together with named markers that explain the result. It cannot say that a document is AI-generated, that its figures are true, or that it is authentic. An intact result means the file was not edited after export; it does not mean the content is correct.

What should I do when an AI-suspected PDF comes back inconclusive?

Treat it as a reason to look further, not as a pass. inconclusive means the document came from software that anyone can use, so the file alone cannot confirm where it came from. Ask the sender for the original from the issuing system, or check the details with the issuer directly – the account number with the bank, the employer with the payroll contact.

Is it worth running a check on every suspected AI document?

A check takes a few seconds and runs on the file only, so it fits as a first filter ahead of manual review or content-level tools. Plans and pay-per-check packs are on the pricing page.

Secure your workflow

Create your account – check PDFs on the web or with an API key, both ready on signup.
No sales call. Cancel any time.

Start Free – Close the Structural Fraud GapSee Pricing
Read API Docs →