Fintech unicorn Razorpay has launchedVulcan, an AI foundation model built specifically for payments, with technology support fromNVIDIA and AWS. The Bengaluru based company said the transformer based model has been trained on nearly3 trillion data points across 4 billion paymentsand uses around 3,000 signals per transaction to make payment related decisions. NVIDIA GPUs were used to train and run the model, while AWS cloud infrastructure and Amazon SageMaker supported its development, training and deployment. Razorpay said Vulcan is designed as a common intelligence layer for payment functions such as routing, fraud detection, risk assessment and checkout personalisation, replacing separate models for individual functions. Components of the model were already being tested on Razorpay’s network ahead of the launch. Customers includingBlinkit, Bachatt and redBushave started using some of these capabilities in live payment environments. Razorpay claims the technology has improved payment success rates by8-10%and helped detect and stop8x more international card fraud. It has also identified five times more fraudulent or disputed transactions without increasing the number of alerts. Through Magic Checkout, the company said40% more shoppers are now shown their preferred UPI app, helping complete an additional1-2 lakh purchases every month. The model draws on signals across merchants, payment instruments, issuers and gateways to assess payment routes and select the one most likely to succeed. It can also identify fraud patterns across merchants, assess risks associated with cash on delivery orders and recommend payment methods during checkout. Razorpay said Vulcan has been developed from the ground up, with its architecture and training data being proprietary. The company’s internal study across1.5 million shoppers and more than 51,000 businessesfound payment friction such as failed transactions, delays and drop offs across metro and smaller markets. Razorpay plans to expand the model to other areas, including authentication and lending, as it looks to use a single AI layer across more payment decisions. The launch comes as India’s digital ecommerce market is projected to reach$350 billion by 2030, according to Razorpay. Bareback Media has recently raised funding from a group of investors. Some of the investors may directly or indirectly be involved in a competing business or might be associated with other companies we might write about. This shall, however, not influence our reporting or coverage in any manner whatsoever. You may find a list of our investorshere.
