Proving the authenticity of a document is an onerous task. It requires the latest techniques combined with the in-depth knowledge of experts to distinguish whether a document is genuine or fake and, thus, help prevent fraud. Fraudsters are evolving forgery techniques and challenging the forensic analysis of documents, also known as the document verification process.
To enhance your knowledge about document verification, we take an in-depth look into the subject, including the main points of an analysis and the method to be used. Continue reading!
What is document verification?
As the word suggests document verification refers to the process of observing and scrutinizing documents to identify whether they are genuine or fake. In other words, document verification is an expert forensic analysis of a document, where several characteristics are analyzed to prove a document’s authenticity.
The subdivisions within document verification
To prove the authenticity of a document, analysis is conducted at various levels and there are several parameters that are scrutinized. Some of these include:
- Graphotechnics: the analysis of handwriting;
- Mecanography: the study of mechanical writing;
- Document alterations: detection of erasures, additions, substitutions and/or obliterations;
- Examination of seals: analysis with the aim of validating the authenticity of seals and other security papers;
- Examination of inks: analysis of physical and chemical characteristics of the inks used in a given document;
- Supporting examination: verification of supporting characteristics of documents, such as paper or polymer;
- Examination of graphic instruments;
- Examination of metal coins and paper money.
The combination of several document analysis techniques is critical as there may be situations where only one parameter may be enough to prove that the document is manipulated, while there may also be cases that require deep investigation and focused attention to the smallest details.
Which documents can be analyzed?
The documents that can be submitted for verification are broadly divided into two types:
- those that have security features;
- those that do not have security elements.
Official documents, such as identity cards and passports usually have security elements. These elements may be incorporated in the materials that make up the document, such as paper and printing ink, a hologram or even the printing of microletters, for example.
Among the documents that do not have security elements, such as a property deed, the analysis of authenticity involves observing other characteristics, such as support, printing and writing.
Currently, document verification is commonly conducted for the following documents:
- ID card,
- Work card,
- Bank and Credit cards,
- Birth certificate,
- Driving license,
- Vehicle license certificate,
- Vehicle registration certificate,
- Trust deed,
- Voter registration,
- School records.
Although any document may be subject to verification, we will focus on the analysis of personal identification documents, which matter the most in fraud detection.
The main points in a document analysis
It is always better to have a clear idea about the main points to be analyzed in a document as it improves the effectiveness of the process and to correctly classify documents as authentic or fraudulent.
The main points in an analysis of security documents are as described below:
- knowledge of product specifications, particularly the security features incorporated in paper and printouts;
- the search for a pattern (a similar authentic piece);
- verification that the document matches with the standard;
- checking for erasures, which consists of mechanical changes to remove entries. The erasure is usually produced by the friction of an eraser, a scraper or a similar instrument;
- verification of chemical washing, total or partial. Chemical washing is a modality of document alteration that consists of making part or all of a text disappear through the use of chemical reagents;
- verification of additions, which consists of inserting dashes, letters, numbers, symbols, syllables, words or phrases in an already prepared document;
- verification of clippings, that is, cuts in the document support made with the aid of scissors, stilettos and/or similar instruments;
- graphic analysis (handwritten entries); and
- shipment analysis.
When necessary, an expert can extend the analysis to ensure the legitimacy of a security document and verify that the document was indeed issued by the relevant ‘Issuing Body’. The extended analysis may include scrutiny of facsimile impressions of stamps, mechanical seals, electronic/mechanographic impressions of variable data, signatures of those responsible for the expedition, among others.
Why is document verification important?
Document fraud continues to increase. From large organizations to the common user, everyone is facing the risks, with businesses incurring financial losses and damage to brand reputation.
According to the National Council on Identity Theft Protection, 2.5 million identities are stolen¹ every year, and every four seconds there is someone who falls prey to identity theft.
In Brazil alone there are 16 million counterfeit identity documents (RG) in circulation, as per the study on identity fraud conducted by Sumsub – a global identity verification platform – which cites data from Serasa Experian.
In such a scenario, document verification emerges as a powerful ally in fraud prevention. This is because a large majority of manipulated photos, data or even, the complete documents that are used for crimes such as ideological falsehood, embezzlement and numerous types of scams, resulting in huge financial losses.
Why is it easier to cheat today?
Gone are the days when document fraud was difficult for fraudsters and required high-quality printing, more sophistication in detail, and higher costs.
The digital transformation is facilitating this crime as fraudsters now have easy access to commoditized tools that make it easy to create fake documents using stolen consumer information and document templates available on the internet.
They can easily manipulate the stolen documents. Today, even an amateur can edit a document in Photoshop and/or other image editors. The ease with which fraudsters can create/manipulate documents makes fraud scalable and more frequent.
As if that were not enough, frequent incidents of data breach and data leaks further increase the risks of document fraud.
For these reasons, it is safe to say that security teams are no longer fighting the same types of crimes that they knew a decade ago.
As fraudsters continue to innovate, businesses are finding it increasingly difficult to predict all possible forms of fraud, and many organizations are facing challenges dealing with and resolving the problem efficiently.
How the documentation process works
Technological development increasingly means the replacement of manual analysis processes with new analysis methods that leverage artificial intelligence and other technologies to speed up the process and improve accuracy.
Manual analysis and its problems
Until a few years ago, the market standard for document verification was manual analysis.
As its name suggests, manual analysis entails numerous human experts analyzing each and every document physically to assess whether they are authentic or not.
They analyze the photo of the front and back of a user-submitted document on several parameters for verification.
This is time-consuming and meticulous work. Therefore, it is essential that the humans analyzing the documents are experts.
This is where the problems start to appear. As the volume of documents received for verification increases, this limited team of experts is overly burdened to accurately verify the documents.
While experts with years of experience in this area can verify the documents and decipher the story behind them, often just by looking at the document and the features that the document presents, the number of experts with such competence is low.
Further, skilled employees are expensive and businesses must pay high wages to hire them due to a shortage of experienced experts in the anti-fraud market.
Today, the reality is that people who perform document verification usually have a year or two of experience in the field.
This means in the initial years, they do not have as much direct experience with analysis, which increases the chances of human errors creeping in and causing losses for companies.
In addition to the problems of large volumes of documents and high costs of skilled professionals, manual analysis is quite technical and subjective; which makes it prone to biases and errors.
Integrated approach: the evolution of document verification
Designed to solve the problems arising out of manual analysis, technology-enabled and AI-powered document verification promises more efficiency and faster analysis of large volumes of documents at reduced costs.
Automation streamlines the document verification process by eliminating redundant steps and complements the analysis by referring critical/unusual documents to human experts for quick verification. Therefore, a combination of technology and human expertise together improve the speed and accuracy of documents being verified, enabling businesses to approve more users.
Leveraging artificial intelligence in the document verification process, makes it possible to compare documents with thousands of other documents from the same geographical location and the year of issue, and spot patterns or inconsistencies indicative of possible fraud.
For example: one of the signs that a document may have been forged is its alignment. Artificial intelligence algorithms can be programmed to read this information automatically and quickly detect and stop fraud.
An automated document verification system is also able to identify signs of document tampering.
For example, if a person uses Photoshop to edit the document, AI-enabled document verification can easily detect the edits, whereas a person, with the naked eye, may not be able to.
Automated document verification is, therefore, less subjective and makes it possible to reduce fraud significantly.
How document verification works at Caf
At Caf, document verification follows an integrated approach which combines artificial intelligence with a hint of human expertise, when needed.
A large volume of documents - more than 90% - are evaluated by an AI-driven system and only the other 10%, that too only for specific cases, are passed on for evaluation by an expert human eye.
The flow of Caf's document verification process has several stages - as described below - and each one of them evaluates a certain set of rules for authenticating the document.
By default, the analyses take place automatically. However, there is always a human expert available to perform document analysis for cases that are unusual or critical. The complete analysis-to-reporting process normally takes up to one minute.
A perfect balance of technology and human expertise can help you obtain the necessary protection against document fraud and avoid losses in your business.
To see how Caf’s ID Document Verification solution works in practice, click this link to speak with a Caf expert and request a free demo.