Razorpay launches Vulcan AI model to lift checkout conversion and fraud detection
Built with NVIDIA and AWS, Razorpay’s proprietary payments model is live with merchants including Blinkit. The company says Vulcan improves payment routing, fraud screening and UPI-app personalisation, with reported gains in payment success and completed purchases.
What happened
Razorpay launched Vulcan, a proprietary AI payments model built with NVIDIA and AWS. It supports routing, fraud detection, risk assessment and checkout
Key facts
- Nearly 3 trillion data points
- 4 billion payments
- Around 3,000 signals per transaction
- 8-10% improvement in payment success rates
- 8x more international card fraud detected and stopped
- 5x more fraudulent or disputed transactions identified
- 40% more shoppers shown preferred UPI app
- Additional 1-2 lakh purchases completed monthly
- 1.5 million shoppers studied
- More than 51,000 businesses studied
- India digital ecommerce market projected at $350 billion by 2030
Why this matters
Payments, commerce and cloud players should view Razorpay’s NVIDIA/AWS-backed model as a partnership opportunity and a signal that proprietary AI is becoming a key payments-platform differentiator.
What to watch
- Independent merchant data confirming sustained 8–10% payment-success improvement after controlling for mix and seasonality.
- Changes in merchant GMV, take rate, retention and enterprise wins attributed to Vulcan.
- Fraud-loss outcomes versus the claimed 8x increase in international-card fraud detection, including false-positive rates.
- RBI, NPCI or consumer-protection guidance on UPI-app personalization, consent and payment-routing transparency.
- Competitor launches from PayU, Cashfree, PhonePe, banks, payment orchestrators or global processors.
- Evidence that improved checkout conversion materially boosts repeat purchase rates and merchant marketing ROI.
- Razorpay packages Vulcan capabilities into premium enterprise checkout, fraud and international-payments offerings.
- Large quick-commerce, marketplace and travel merchants run A/B tests and shift more transaction volume to AI-driven routing.
- Razorpay expands integrations with issuers, card networks, UPI apps and merchant CRM systems to improve model inputs and closed-loop attribution.
- Competitors announce comparable AI fraud, payment orchestration or checkout-personalization products.
- Merchants rebalance checkout design toward preferred UPI-app prompts, increasing pressure on apps to secure default-position and distribution partnerships.
Also reported by
- Entrackr — Same time