Mastercard, a leading payments giant, has unveiled its proprietary generative artificial intelligence model designed to enhance fraud detection across its extensive network of banks. The exclusive AI solution, named Decision Intelligence Pro, empowers banks to assess suspicious transactions in real time, distinguishing between legitimate and fraudulent activities.

Ajay Bhalla, Mastercard's President of the Cyber and Intelligence Business Unit, shared in an exclusive interview that Decision Intelligence Pro is a proprietary recurrent neural network, a pivotal component of generative AI. Developed from scratch by Mastercard's cybersecurity and anti-fraud teams, the advanced AI model employs transformer models to harness the capabilities of generative AI.

Bhalla says, "It's all built in-house, and we've gathered diverse data from the ecosystem. Due to the nature of our business, we receive transaction data, allowing us a comprehensive view."

While Mastercard occasionally utilizes open-source technology, the majority of Decision Intelligence Pro is internally developed. The algorithm is trained on data from approximately 125 billion transactions processed annually through Mastercard's card network.

Unlike large language models such as OpenAI's GPT-4 and Google's Gemini, Mastercard's algorithm focuses on understanding relationships between merchants. By analyzing a cardholder's history of merchant visits, the algorithm predicts potential fraudulent transactions and generates pathways through Mastercard's network, resembling a heat-sensing radar.

The algorithm assesses transactions in just 50 milliseconds, providing a score that reflects the typical behavior expected from the cardholder. According to Bhalla, this technology can enhance financial institutions' fraud detection rates by an average of 20%, with some cases showing improvements up to 300%.

Mastercard, having invested over $7 billion in cybersecurity and AI technologies in the last five years, aims to reduce banks' costs associated with assessing illegitimate transactions by up to 20%. The company believes its algorithm's true potential lies in identifying patterns and trends to predict future types of fraud not currently known within the payments ecosystem.

Bhalla emphasized, "The beauty of Mastercard’s ecosystem is we see data from all our customers globally from these transactions. What that does is it helps us see fraud and patterns across the ecosystem globally."

As competitors like Visa also invest in AI technologies, Mastercard anticipates its algorithm will bring significant advancements in predicting and preventing fraud, ultimately revolutionizing the payments industry.

READ: Tech companies layoffs big in January

For more Updates Connect with us on

Facebook

Instagram

WhatsApp