Generative artificial intelligence in banking: reducing fraud

Abstract

Recently, the number of financial crimes in banking organizations has been growing significantly, which in turn has a negative impact on the image component of banks, and also leads to huge losses not only on the part of clients, but also on the part of banks in terms of compensation for this damage. This article reflects the solution of current issues of reducing the number of fraudulent transactions in relation to the funds of clients of banking institutions. The work proposes an improvement of the existing fraud monitoring system designed to assess financial and non-financial events for their suspiciousness from the point of view of possible fraud. An analysis of the typical structure of the bank's anti-fraud system was carried out, its shortcomings were identified. As a result of the studies, a new structure of the anti-fraud system with generative artificial intelligence was proposed, and the main risks of using this approach, as well as measures to prevent and correct them, were formulated.