India's Payments Chief Hints at Commercial Models to Solve UPI Monetization Woes

As India’s Unified Payments Interface (UPI) pushes toward a target of one billion daily transactions, the National Payments Corporation of India (NPCI) is looking to artificial intelligence to define the next era of the network. According to TechCrunch, NPCI CEO Dilip Asbe believes AI will be instrumental in scaling user growth, bolstering fraud prevention, and facilitating credit distribution. While the network currently processes over 750 million transactions daily, Asbe suggests that the next phase of development relies on effectively leveraging AI to simplify onboarding through voice and multilingual solutions, and identifying financial mules to protect the ecosystem.

Despite the network's ubiquity, the Indian payments landscape remains heavily concentrated. TechCrunch reports that Walmart-owned PhonePe and Google Pay together command over 80% of the market share. Asbe attributes this persistent concentration risk to the current absence of a viable commercial model within the ecosystem. He argues that the moment a sustainable commercial framework becomes available, new entrants will be incentivized to invest heavily, potentially challenging the current duopoly. This perspective shifts the focus from simple transaction volume to the underlying economics that currently favor incumbents who have already poured millions into attaining their positions.

The regulatory environment remains a looming factor for these platforms. The regulator’s plan to cap an individual app’s market share at 30% is currently scheduled to take effect on December 31, 2026, barring any further deferrals. Asbe noted that switching costs for UPI apps are currently very low, and core features are largely shared across platforms. The challenge for new entrants, therefore, is not technical parity but the ability to build a business model that justifies the capital expenditure required to compete with the established players.

Beyond payments, Asbe sees a significant opportunity for Indian banks and fintechs to develop small language models. According to TechCrunch, he believes these models will differentiate themselves based on the rich datasets available within the Indian ecosystem. Asbe advocates for models that are sharp, specific, and as deterministic as possible. This strategy aligns with NPCI’s broader efforts to integrate agentic commerce, though Asbe acknowledged that a wider rollout of such capabilities is still pending, noting that NPCI’s previous demos with partners like Razorpay have yet to see broad adoption.

Meanwhile, the NPCI continues to manage its own sovereign offering, the BHIM app. After spinning off the application in 2024 to enhance its competitiveness, the platform has grown its transaction volume, though it currently maintains roughly 1% of the market share. Asbe clarified that NPCI is not targeting a specific market share for BHIM; rather, the goal is to position it as a secure, sovereign alternative within the ecosystem. As the industry awaits the implementation of market share caps, the focus remains on whether AI-driven efficiencies can provide the necessary leverage for a more competitive, decentralized market.