Document Forgery Explained: Techniques, Detection, and Law

A forged document only needs to survive one review. A fake signature needs to clear one bank teller. A doctored contract needs to pass one signature match. An altered ID needs to get past one bouncer or one HR desk. Once it clears that single moment, the forger has what they wanted, and the person or business on the other side is left holding the loss.

Document forgery is not new, and it is not going away. What has changed is the tooling. A crime that once required a steady hand, a bottle of ink, and years of practice now sometimes takes nothing more than a photo editor or a generative AI prompt. This article covers what document forgery actually is, the techniques forgers use, how forensic examiners and modern detection tools catch it, and what US law says about punishing it.

What Is Document Forgery

Document forgery is the act of creating, altering, or signing a document with intent to deceive, so that it appears to be something it is not. That definition covers a wide range of conduct: a signature copied by hand, a diploma printed from a template, a bank statement edited in a PDF tool, or an invoice generated by AI to look like it came from a real vendor.

Legal definitions generally require two things to be present at once. First, the document has to misrepresent its own origin, meaning it claims to be made or signed by someone who did not actually make or sign it, or it claims facts that are not true. Second, whoever made or presented the document has to have meant to deceive someone with it. A careless error or a prank does not clear that bar. Without that intent element, an altered document might still be a problem, but it is not forgery in the legal sense.

It helps to separate forgery from two terms people often use interchangeably with it.

Counterfeiting usually refers to copying a protected item, like currency, a trademark, or a branded product, so that it passes as the genuine article. Forgery is broader and applies to any document or signature, branded or not.

Fraud is the umbrella term. Forgery is one method of committing fraud. A person can commit fraud without forging anything, for example by lying verbally to get a loan. Forgery specifically involves a false or altered document or signature used to carry out that deception.

The Main Types of Document Forgery

Forgery does not look the same across every case. Investigators and forensic examiners generally sort it into three buckets, because each one leaves a different kind of evidence behind.

Traced or simulated forgery: The forger copies a real signature or document by tracing it, freehand copying it after studying it, or using a light box to overlay the original. This category targets signatures most often, since a signature is the easiest single element to fake convincingly.
Alteration forgery: A genuine document already exists, and the forger changes part of it after the fact. A real invoice with an inflated total, a real bank statement with an edited balance, or a real prescription with an added line all fall here. The base document is authentic; the specific content is not.
Fabrication forgery: The document is built from nothing. There was never a genuine original. A fake diploma from a school the person never attended, an invoice from a vendor that does not exist, or a counterfeit ID for a person who has no legitimate one are all fabricated rather than altered.

Each type leaves a different fingerprint. Traced signatures tend to fail under stroke and pressure analysis. Alterations tend to fail on internal consistency, since editing one number rarely updates every other number connected to it. Fabrications tend to fail on structure and detail, since building an entire document from scratch means guessing at formatting, security features, and issuing conventions that the forger has usually never seen up close.

Common Document Forgery Techniques

Signature Forgery

Signature forgery covers freehand simulation, where the forger studies a real signature and reproduces it by memory and practice, and tracing, where the forger places the real signature underneath the target document and copies the outline directly. Freehand simulation tends to look natural but drifts from the true signature under close comparison. Tracing captures the shape accurately but often shows unnatural pen lifts, inconsistent pressure, and a slower, more deliberate stroke than a genuine signature written at normal speed.

Physical Alteration

Before digital tools, forgers relied on physical methods: erasure, chemical bleaching to lift ink off paper, correction fluid layered under new text, or cut-and-paste work where a genuine element from one document is physically transplanted onto another and re-photocopied to hide the seam. These methods still show up today, particularly on paper receipts and printed forms, and they remain detectable through close physical inspection.

Digital Editing

Image editing software turned document alteration into something almost anyone can attempt. A number on a scanned invoice gets replaced. A name on an ID photo gets swapped. A total on a receipt gets bumped up before it is submitted for reimbursement. Digital edits often introduce texture mismatches, blurred edges around the altered region, misaligned pixels, or a font that is subtly different from the rest of the document, since matching an institution's exact typeface takes more effort than most forgers put in.

Template and Blank Forgery

Forgers also build documents from scratch using templates that mimic a real institution's layout, logo, and formatting. Blank check stock, fake letterhead, and downloadable ID templates fall into this category. These forgeries can look convincing at a glance but often fail on details the template creator got wrong: an outdated logo version, a security feature that does not exist on the real document, or formatting that does not match how the actual institution issues its paperwork.

AI-Generated Forgery

The newest category does not start with a real document at all. Generative AI tools can produce a synthetic receipt, invoice, or identity document based on patterns learned from thousands of genuine examples, without ever copying one specific original. These documents can look structurally polished while containing entirely invented details: a vendor that does not exist, a transaction that never happened, an address with no real business behind it.

AI-generated forgeries behave differently from traditional forgeries in review. There is often no obvious splice, no mismatched font, no visible edit. The document can look internally consistent while still being completely fabricated, which means the strongest signals tend to come from cross-checking content against outside records rather than from a purely visual scan. The Docklands article on how to detect AI-generated receipts and synthetic invoices covers this shift in more depth.

How Document Forgery Is Detected

Forensic document examination has existed for over a century, and most of its core methods still apply today, alongside newer tools built for digital and AI-generated fakes.

Handwriting and signature analysis: Examiners compare stroke direction, pen pressure, letter formation, spacing, and the natural variation that appears across multiple genuine signatures from the same person. A traced signature usually shows slower, more hesitant strokes and less natural variation than a signature written from memory.
Physical and material examination: Paper stock, ink chemistry, and printing method can all be tested. Ink that has not fully aged compared with the rest of a document, paper that does not match the claimed printing era, or toner inconsistent with the printer an institution actually uses are all evidence that a document has been altered or fabricated.
Optical and magnification tools: Ultraviolet light, infrared imaging, and high-magnification photography reveal details invisible to the naked eye, including erased text, security fibers embedded in official paper, and layering from correction fluid or added ink.
Digital forensics and metadata review: For files that never existed on paper, metadata tells its own story. A native PDF carries a creation date, a modification date, and the name of the software last used to touch it. A statement that claims to be a direct bank export should not carry metadata showing it passed through a design or editing tool after its claimed issue date. The Docklands post on why fraud detection should start with original files and the deeper dive on metadata forensics for receipts both cover how much detail gets lost the moment a document is flattened into a screenshot or a compressed image, which is often exactly what a forger wants.
Content and mathematical consistency checks: Numbers that should reconcile but do not are one of the most reliable tells in the entire field. A subtotal, tax, and total that do not add up correctly. A running balance that does not match the sum of listed transactions. A pay period that overlaps with another submitted document. The Docklands article on doctored invoices walks through how altered totals leave arithmetic evidence behind, even when the visual edit itself is clean.
Cross-referencing external records: A document can be internally flawless and still be fraudulent if the business, person, or transaction it describes does not exist. Confirming that a vendor is registered, that an address is real, or that a claimed employer actually operates at the listed location catches fabrications that pass every internal check.
AI-assisted detection at scale: Manual forensic review does not scale to hundreds or thousands of documents a week. Machine learning models trained on large sets of genuine and forged documents can flag pixel-level tampering, unnatural texture patterns, and formatting inconsistencies far faster than a human reviewer, and they can apply the same standard to every single document rather than a random sample. The Docklands guide on how to catch fake documents breaks down how layering these checks together catches forgeries that would slip past any single method on its own.

Document Forgery and the Law in the United States

Forgery is prosecuted at both the state and federal level, and the charge that applies depends on the document involved and who was harmed.

State law: Most forgery cases are charged under state penal codes. Elements typically include making, altering, or using a false document or signature, with intent to defraud or injure another person. State forgery is frequently a felony, especially when the document involved has financial value, such as a check, deed, or contract. Sentencing ranges vary widely by state, but felony convictions commonly carry a range of one to several years in state prison, along with fines and court-ordered restitution to the victim.
Federal law: Several federal statutes address forgery depending on the type of document. Forging currency or government securities falls under 18 U.S.C. § 471, carrying penalties of up to 20 years in federal prison. Fraud involving identification documents, including driver's licenses, birth certificates, and other forms of ID, falls under 18 U.S.C. § 1028, with a tiered penalty structure that can reach 15 years or more depending on the offense and whether it connects to another crime like drug trafficking or terrorism. Passport-related forgery is charged separately under 18 U.S.C. § 1543. A case moves from state to federal jurisdiction when a forged document crosses state lines, involves a federal agency, or targets a federally regulated institution such as a bank.
Intent is the deciding factor: A document can be inaccurate, outdated, or even physically damaged without being forged. What separates an honest mistake from a crime is intent: the person creating or presenting the document has to know it is false and mean to deceive someone with it. This is also the most common defense raised in forgery cases, alongside arguments about the authenticity of the handwriting comparison or chain of custody for the physical document.
Civil exposure runs alongside criminal liability: A forgery conviction does not close the door on a lawsuit. Victims can pursue civil claims for damages, contract rescission, or restitution beyond what a criminal court orders, and businesses that accepted a forged document, even unknowingly, can face their own liability if they failed to apply reasonable checks before relying on it.

Where Document Forgery Causes the Most Business Damage

Forgery is not just a criminal law topic. It is a recurring operational risk for any business that relies on documents to make financial decisions.

Accounts payable teams face forged and altered invoices submitted for payment, sometimes for vendors that do not exist at all. The Docklands guide on invoice fraud covers how these schemes typically run.
Insurance carriers encounter forged or altered receipts, repair estimates, and medical bills submitted to inflate or fabricate claims. The comparison of fraud detection models against document forensics explains why claims teams need both approaches working together.
Employee expense programs deal with altered or duplicated receipts submitted for reimbursement, a pattern covered in the Docklands piece on receipt fraud manipulations.
Lenders and onboarding teams review forged income documents, bank statements, and identification during account opening and credit decisions.

The common thread across every one of these use cases is volume. A single forged document is an isolated crime. A thousand documents a week, reviewed by a small team under time pressure, is where forged documents most often slip through and turn into recurring financial loss.

Building a Detection Process That Actually Works

A reliable defense against document forgery layers several checks rather than depending on any one of them.

Compare against known genuine samples for layout, fonts, and formatting whenever a reference document exists.
Recalculate the math on every invoice, statement, or receipt rather than trusting the printed total.
Pull metadata from native digital files to check edit history and the software used to create or modify them.
Cross-check external details, confirming that the business, person, or institution named in the document is real and matches the claim.
Request original files, not screenshots or compressed copies, since originals carry far more forensic evidence.
Escalate on combined signals rather than a single flag, since a mismatched font plus a math error plus a metadata anomaly is a far stronger case than any one issue alone.
Apply the same standard to every document, not a random sample, since forgers count on volume overwhelming manual review.

How Docklands Helps Catch Forged Documents

Docklands AI applies exactly this kind of layered review automatically, at the pixel level, across every invoice and receipt that comes through a business rather than a spot-checked sample. The platform combines digital edit detection, metadata forensics, mathematical verification, physical tampering detection, and AI-generated content detection into a single pass, so the signals that a forgery would otherwise scatter across multiple checks get caught together.

These checks apply directly across the areas where forged documents cause the most financial exposure: insurance claims, accounts payable, and employee expenses.

FAQ

What is the difference between forgery and fraud?

Fraud is the broader crime of deceiving someone for gain. Forgery is one specific method of committing fraud, involving a false or altered document or signature. Not all fraud involves forgery, but forgery is almost always a step toward fraud.

Is forgery always a felony?

Not always, but it frequently is, especially when the forged document carries financial value or legal weight, such as a check, deed, contract, or government ID. Some states allow misdemeanor charges for lower-value or first-time offenses, while federal forgery statutes generally carry felony-level penalties.

Can someone be charged with forgery if the document was never actually used?

In many jurisdictions, yes. Creating or possessing a forged document with intent to defraud can be enough to support a charge, even if the document was never presented to anyone. Actually using it to obtain money, property, or a benefit typically increases the severity of the charge.

How do forensic examiners tell a traced signature from a genuine one?

They look at stroke speed, pen pressure, and natural variation. A genuine signature written at normal speed shows some inconsistency across multiple samples. A traced signature is often too consistent, with slower, more deliberate strokes and unnatural pen lifts where the forger paused to follow the original.

Are AI-generated documents legally treated the same as traditional forgeries?

Generally yes. Most forgery statutes focus on whether a document falsely represents its origin or content and whether the person acted with intent to defraud, not on the specific tool used to create it. A synthetic invoice generated by AI to deceive a business into paying it fits the same legal definition as a hand-altered one.

The Takeaway

Document forgery has moved a long way from ink and light boxes, but the underlying goal has not changed: get a false document past one review before anyone looks closely. Traditional forensic techniques, from handwriting analysis to material testing, still catch a meaningful share of forgeries. Digital metadata, mathematical consistency checks, and AI-assisted detection close the gap left by scanned images, edited PDFs, and synthetic documents that were never real to begin with.

Businesses that review invoices, receipts, and claims at volume can see how Docklands AI applies these checks automatically by visiting Docklands AI or booking a demo.