BAKU, Azerbaijan, February 3. Mastercard has announced the development of its own generative artificial intelligence (AI) model designed to detect and counteract fraudulent transactions in the banking sector, Trend reports.
According to Mastercard, the processing of the data will take only 50 milliseconds, allowing banks to assess suspicious transactions in real-time.
"The new advanced artificial intelligence model, Decision Intelligence Pro, is developed as a patented recurrent neural network and is a key part of Mastercard's generative AI. The model is built from scratch within the company using data from over 125 billion transactions annually," said President of Mastercard's Cyber and Intelligence Solutions business division Ajay Bhalla.
He pointed out that the new transaction decision technology can improve fraud detection rates by an average of 20 percent, leading to significant savings for financial institutions.
The Mastercard algorithm is expected to be also capable of predicting future types of fraud that are currently unknown, which is one of the key advantages of this technology.
To note, over the last five years, Mastercard has invested more than $7 billion in the development of cybersecurity and artificial intelligence technologies, focusing on creating effective tools to combat fraud in the payment ecosystem.