From NVIDIA: 2025-06-03 02:00:00

Financial losses from credit card transaction fraud are projected to exceed $403 billion over the next decade. NVIDIA’s new AI Blueprint for financial fraud detection, launched at the Money20/20 conference, offers advanced algorithms to combat fraud by improving accuracy and reducing false positives. Leading financial organizations like American Express and Capital One have already been using AI to enhance fraud detection and customer protection. The blueprint, optimized for credit card fraud, can also be adapted for other fraud use cases. NVIDIA RAPIDS and CUDA-X Data Science libraries enable payment companies to accelerate data processing and integrate graph neural networks for more effective fraud detection. The AI Blueprint enhances the XGBoost ML model with GNNs to reduce false positives and improve model accuracy. Major financial institutions like American Express and BNY Mellon are already leveraging AI to enhance fraud detection and customer protection. Systems integrators, software vendors, and cloud service providers can integrate NVIDIA’s new fraud detection blueprint to enhance their financial services applications and safeguard customers’ assets.



Read more at NVIDIA: New NVIDIA AI Blueprint Detects Fraudulent Credit Card Transactions