Mastercard has announced a new proprietary generative artificial intelligence model. The company claims that it has built its own AI model that will help thousands of banks in its network to detect and root out fraudulent transactions.
In a statement to CNBC, the company said that its new advanced AI model is named Decision Intelligence Pro. This AI model will allow banks to better assess suspicious transactions on Mastercard’s network in real time and determine whether they’re legitimate.
Mastercard Decision Intelligence Pro: How it works
Mastercard’s proprietary algorithm is trained on data from nearly 125 billion transactions that go through the company’s card network annually. The company claimed that this data helps Mastercard’s AI model to understand relationships between merchants and predict where fraudulent transactions are taking place.Instead of textual inputs, Mastercard’s algorithm uses the history of a cardholder’s merchant visit as the prompt. This helps the model to determine whether the business involved in a transaction is a place the customer would likely go.
The algorithm then generates pathways through Mastercard’s network (which works like a heat-sensing radar) to find the answer in the form of a score.
A higher score would follow the pattern of what’s the usual kind of behaviour expected from the cardholder. Meanwhile, a lower score is out of that pattern. Mastercard claims that this process all happens in just 50 milliseconds.
What Mastercard CEO said about Decision Intelligence Pro
Mastercard’s president of the cyber and intelligence business unit, Ajay Bhalla, said that the new AI solution is a proprietary recurrent neural network (which is a core part of generative AI). He also claimed that Mastercard has built this network from scratch by the company’s cybersecurity and anti-fraud teams.
In an interview, Bhalla said: “We are using the transformer models which help get the power of generative AI. It’s all built in the house and 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 which comes to us from the ecosystem.”
Bhalla noted that Mastercard relies on open source “whenever needed,” in some cases. However, the “majority” of the technology is created in-house, he added.
He said that the new transaction deciding technology from Mastercard can help financial institutions improve their fraud detection rates by 20%, on average. In some cases, though, the model has led to improvements in fraud detection rates of as much as 300%, Bhalla mentioned.
Mastercard also said that it has invested more than $7 billion in cybersecurity and AI technologies over the last five years.
