Indian fintech companies like Pine Labs and Paytm are expanding into agentic payments, leveraging AI to automate and optimise transaction processes, signalling a new frontier in digital payments and operational efficiency.
Indian fintech groups are rapidly expanding into agentic payments, a new frontier in which software agents can act on behalf of users or merchants, because they see a chance to sell more services around their core businesses. Pine Labs said in June it launched P3P, an open protocol that lets AI agents initiate and complete payments on behalf of users with prior consent, while Paytm is developing similar tools for merchants. Cashfree Payments, Juspay and Razorpay have also introduced agentic interfaces aimed at helping merchants improve transaction success rates.
The shift is about more than marketing. These systems sit behind the merchant interface and can automate tasks such as chasing abandoned carts, retrying failed payments, issuing refunds and sending payment links when a business owner gives a plain-language instruction. Pine Labs has described P3P as a consent-driven model built on standard web interfaces, with support for delegated authorisation, order creation, payment execution and verifiable receipts. The company says the protocol is designed to work with merchants, AI agents and other autonomous systems that need secure machine-led payments.
Executives are already pointing to internal efficiency gains. Paytm president and group chief financial officer Madhur Deora said on an analyst call that AI was improving operating leverage and could support higher margins over time. Pine Labs chief executive Amrish Rau said on another analyst call that about 90 per cent of new code in at least four divisions was being written with AI, and that the company was seeing strong demand in India, especially for agentic payments. Rau added that the impact was showing up first in the company’s online business.
Still, the revenue case remains early. Fintech founders say many AI products are still in beta or testing and have yet to make a meaningful contribution on their own. At the same time, some firms are facing a cost squeeze as demand for memory and servers pushes up infrastructure costs, which may feed through to cloud bills later. For now, companies are leaning towards outcome-linked pricing, charging for measurable results such as successful cart conversions, dispute resolution or completed collections. In July, Business Standard reported that the National Payments Corporation of India was working on a standard for agentic payments on Unified Payments Interface, called Unified Agentic Protocol, which would allow AI agents to be registered, verified and authorised across the system without changing the underlying rails.
Disclaimer: This article is intended to inform and educate, not to recommend or endorse any financial product, investment or strategy. Please consider your own financial circumstances and seek professional advice where appropriate before making financial decisions.
