Razorpay launches Vulcan, an AI foundation model designed specifically for payments, as the fintech company looks to improve payment success rates, fraud detection and checkout experiences. Built in collaboration with NVIDIA and AWS, the transformer-based model has been trained on nearly 3 trillion data points across 4 billion payments.

Key Highlights

• Razorpay's Vulcan AI model improved payment success rates by 8-10% and detected 8x more international card fraud.
• Built with NVIDIA and AWS, Vulcan is trained on 3 trillion data points across 4 billion payments.

Razorpay said early components of Vulcan have already delivered an 8-10% improvement in payment success rates. The company also reported that the technology detected eight times more international card fraud and identified five times more fraudulent or disputed transactions, without increasing the overall number of alerts generated. On Razorpay's Magic Checkout, 40% more shoppers are now seeing their preferred UPI app, helping complete an additional 1-2 lakh purchases every month.

A Single Intelligence Layer Across Payments

The model is designed to function as a unified intelligence layer across payment routing, fraud, risk and checkout, replacing the separate machine-learning models Razorpay previously used for individual functions. Vulcan learns from roughly 3,000 signals per transaction and evaluates payment routes in real time to identify the route most likely to succeed for a given payment.

Razorpay said the model is already being used in live payment environments by customers including Blinkit, Bachatt and redBus. An internal study covering 1.5 million shoppers and more than 51,000 businesses found similar payment friction patterns across both metro cities and smaller towns, the company noted.

Beyond improving success rates, Vulcan can also detect fraud across Razorpay's broader network, flag potentially risky Cash on Delivery orders, and recommend the payment method most likely to succeed for a specific customer.

Razorpay CEO on Building Trust in Digital Payments

Harshil Mathur, CEO and Founder of Razorpay, said India's appetite for digital payments is real, but not yet universal - for a large part of the country, going digital ultimately comes down to whether it works, every single time. He said Vulcan was built for the customer still deciding whether to trust a screen over cash in hand, adding that an AI-led payments foundation model doesn't just solve today's problem and stop there - every payment teaches the system something that makes the next payment better. He described this as less of a product launch and more the starting point for how payments in India will continue improving on their own for years to come.

Razorpay clarified that Vulcan is not a large language model like ChatGPT. Instead, the company describes it as a foundation model trained specifically on payments data to understand patterns in the movement of money. The system was built using NVIDIA AI Models and infrastructure to train and run the model, while AWS infrastructure, including Amazon SageMaker, supports its ongoing development, training and deployment.

NVIDIA and AWS Executives Weigh In

Pahal Patangia, Head of Global Industry Business Development and Payments at NVIDIA, said India's rapidly evolving digital economy is creating an opportunity to make payments more intelligent, reliable and secure. He said NVIDIA  collaboration with Razorpay and AWS on AI payments foundation models has opened up a new frontier, turning complex payments data into real-time contextual intelligence through a proprietary, purpose-built semantic AI layer.

Kiran Jagannath, Head of FSI and Conglomerates at AWS India and South Asia, said Razorpay is reimagining payments intelligence at India scale with an AI foundation model built on Amazon SageMaker that consolidates billions of transaction insights into a single, continuously learning intelligence layer. He added that this replaces fragmented ML models with unified AI that delivers higher payment success rates, rapid iteration and enterprise-grade security for mission-critical payment flows.

Also Read: Razorpay Ties Up with Replit as India Payments Partner for AI Platform

Looking Ahead

Razorpay said it eventually wants Vulcan to power a wider range of payment decisions, extending from authentication and routing to fraud detection and beyond, as the company continues expanding the model's role across its payments ecosystem.