The AI-based solution from the company BIT, developed by the technological partner of the state corporation Rostech, NtechLab, prevented thefts in Russian retail stores by more than 2.2 billion rubles in January-August 2025. The system analyzes video streams from surveillance cameras in fractions of a second, recognizes suspicious activity, and alerts store personnel. According to statistics, the items of greatest interest to shoplifters are alcohol, cosmetics, and electronics. The thefts are primarily committed by men with low-income levels.

BIT estimates that, on average, the number of attempted store thefts has been growing annually by 5-7% during the period from 2020 to 2025. Based on anonymized data, an average profile of a shoplifter has been compiled. About 60% of offenders are men. Typically, violators are around 34 years old and have a low level of income.

In recent years, the share of thefts among people with medium and high-income levels has increased. The primary motive for attempting to take something from a store without paying is for subsequent resale. Sometimes crimes are committed under the influence of alcohol or due to kleptomania. A small percentage of thefts are even committed "on a dare."

Regarding product categories, the most popular targets have become expensive food items — alcohol, cheeses, coffee, and meat delicacies. Cosmetics, perfumes, various electronics, and branded clothing and accessories are also popular among shoplifters.

The most common methods used for shoplifting are traditional ones — concealing merchandise under clothing or in a bag, as well as attempts to cut off RFID tags or simply remove an item from its packaging.

"Professional shoplifting groups, posing the greatest danger to retail, however, use more cunning methods. Fortunately, artificial intelligence cannot be outsmarted. It analyzes video streams from cameras in fractions of a second and alerts staff to violations. Essentially, the neural network prevents thefts from store shelves and doesn't even give a hypothetical opportunity to steal something," noted Alexey Palamarchuk, General Director of NtechLab.

The increase in thefts is recorded primarily in poorly protected stores that are not equipped with video surveillance and facial recognition systems, have "blind spots," weak staff administration, and a high proportion of outsourced management. In places where a consistent set of anti-theft measures is in place, the indicators practically do not grow.

The system has been in use since 2017. The software is already operating in more than 10,000 stores from Vladivostok to Kaliningrad.