Smart Engines: AI-based PAC to protect deals from fraudulent attacks

2024: Announcement of the creation of AI-based PAC to protect transactions from fraudulent attacks

On August 27, 2024, Smart Engines announced the creation of firmware complex on the basis AI to protect transactions from. fraudulent attacks The technology is capable of verifying the authenticity of a set of documents provided by several parties in a few seconds. The solution will allow employees, banks notary chambers and other organizations to protect themselves from swindlers and optimize customer service.

Smart Engines

According to the company, ON several documents can be placed in a multispectral scanner with Smart Engines at the same time - for example, two passports. RUSSIAN FEDERATION The built-in AI will perform parallel authentication of all of them. When concluding a purchase agreement sales real estate , the technology will provide both safety parties. In addition, the system is able to simultaneously check a set of documents of one person - for example, a general civil passport, a driver's license, SNILS.

The developer's algorithms identify 66 signs of forgery in three ranges: optical, ultraviolet (UV), infrared (IR). The system analyzes the originality and protection elements of the document form (protective fluorescent fibers, fluorescent ink, etc.), detects mechanical interventions, including photo switching and other manipulations with the document. In addition, the AI also cross-validates the main spread data with the data of the machine-readable area (MRZ), checks the chip and the validity of the document.

Previously, in order to verify the authenticity of a client's passport, his driver's license or, for example, SNILS, employees of financial or insurance companies had to separately attach each of the documents to the scanner. Thanks to Smart Engines, it is possible to optimize this procedure and save time.

With the increase in the number of forged documents, many organizations have become interested in devices for detecting fakes. However, there was a problem that if the client presented several documents, checking all of them led to delays and queuing. Now, by optimizing neural network algorithms, we have managed to solve this problem. commented Vladimir Arlazarov, CEO of Smart Engines, Doctor of Technical Sciences

The solution was implemented with the financial support of the Fund for Assistance to the Development of Small Forms of Enterprises in the Scientific and Technical Sphere (Innovation Assistance Fund) and the Russian Information Technology Development Fund. Smart Engines document analysis and recognition systems are used in government agencies, industry and business.