Mastercard has taken a significant step forward in enhancing fraud detection capabilities for banks within its network by launching its proprietary generative artificial intelligence (AI) model, Decision Intelligence Pro.
This development represents a groundbreaking use of generative AI technology to improve the identification and prevention of fraudulent transactions in real-time.
Designed by Mastercard’s cybersecurity and anti-fraud teams, the model is built upon a proprietary recurrent neural network, a key component of generative AI.
Ajay Bhalla, the head of Mastercard’s cyber and intelligence division, revealed to CNBC that their latest AI innovation is a unique recurrent neural network, a fundamental element of generative AI, meticulously developed in-house by their cybersecurity and anti-fraud experts.
“We are using the transformer models, which basically help get the power of generative AI,” Bhalla told CNBC in an exclusive interview earlier this week. “It’s all built in-house. We’ve got all kinds of data from the ecosystem. Because of the very nature of the business we are in, we see all the transaction data that comes to us from the ecosystem.”
This technology is trained on the vast data generated from approximately 125 billion transactions processed annually across Mastercard’s card network. Unlike traditional methods, this model focuses on understanding the relationships between merchants to predict fraud more accurately.
Mastercard’s approach to fraud detection with Decision Intelligence Pro is innovative, using the history of a cardholder’s merchant visits as a basis for evaluation.
This method functions similarly to a heat-sensing radar, navigating through the network to assess the legitimacy of transactions. The process assigns scores to transactions in just 50 milliseconds, significantly improving the speed and efficiency of fraud detection.
Implementing this AI model has enhanced fraud detection rates by 20% on average, with some instances showing improvements of up to 300%.
As the news hit the internet, readers seemed disappointed. Critics argue that generative AI is unrelated to cybersecurity, suggesting that Mastercard might be leveraging a trendy buzzword to generate unwarranted excitement.
Yawn…….
Just.abother company who wants to latch onto the AI buzzwords to juice their stocks.
AI credit card fraud detection? Puhlez…. It would have been better if they said 75% of the workforce has been laidoff thanks to AI, saving the company billions. 😂🤣
— Alby (@albyva) February 1, 2024
fraud detection isnt even genAI
— algomax (@algomax06) February 1, 2024
The company’s dedication to advancing financial security is further demonstrated by its investment of over $7 billion in cybersecurity and AI technologies over the past five years. This includes strategic acquisitions like purchasing Swedish cybersecurity firm Baffin Bay Networks in March 2023.
With these advancements, Mastercard not only aims to bolster its fraud detection capabilities but also to reduce the costs associated with managing illegitimate transactions, potentially saving banks up to 20% of these expenses.
Introducing Decision Intelligence Pro by Mastercard marks a significant milestone in using AI within the financial sector. This model’s ability to predict and identify fraudulent patterns and trends offers a promising outlook for fraud prevention in the future.