American Express Relies On AI To Fend Off Fraudsters

Financial fraud is a major cause for concern to banks. Credit and bank cards are considered the major targets. And now that most transactions are conducted online, cybercrime will be on the rise as well. It is estimated that this type of criminal activity costs the global economy $600 billion annually or 0.8 percent of the worldwide GDP.

That is why American Express, which handles 8 billion transactions yearly, employs AI deep learning on NVIDIA GPU’s computing platform. In fact, according to NVIDIA CEO Jensen Huang, American Express has now deployed deep-learning-based models optimized with NVIDIA TensorRT and running on NVIDIA Triton Inference Server to detect fraud.

Photo by Daria Shevtsova from Pexels

In case you wanted to know, Here’s what NVIDIA’s tech does:

• NVIDIA TensorRT is a high performance deep learning inference optimizer and runtime that minimizes latency and maximizes throughput.
• NVIDIA Triton Inference Server software simplifies model deployment at scale and can be used as a microservice that enables applications to use AI models in datacenter production.

American Express has benefited greatly from such technology. Despite having more than 115 million active credit cards, the company has maintained the lowest fraud rate in the industry for 13 years in a row, according to The Nilson Report. Its systems are able to monitor transaction anomalies and flags ones that stand out using recurrent neural networks.

According to NVIDIA, “American Express was able to implement this enhanced, real-time fraud detection system for improved accuracy. It operates within a tight two-millisecond latency requirement, and this new system delivers a 50x improvement over a CPU-based configuration, which couldn’t meet the goal.”

This is just one of the areas where AI has proven to be a major asset. It ensures that that the backbone of the financial system remains in tact, and help banks and their customers protect themselves from cybercriminals and fraudsters.