NPCI chief says AI will drive UPI from 750 million to billion daily transactions

NPCI targets ambitious growth for India's digital payments ecosystem

India's Unified Payment Interface has reached a significant milestone, processing over 750 million daily transactions. Now, the National Payments Corporation of India is setting its sights higher. Dilip Asbe, MD and CEO of NPCI, which oversees UPI, believes artificial intelligence will be the driving force behind reaching billion daily transactions1

1

. Speaking at Mumbai Tech Week 2026, Asbe outlined how AI integration could propel the next era of digital payment growth through collaboration between NPCI, India's central bank, and the government2

2

.

AI in digital payments to power multiple use cases

"AI will be used very effectively when we look at the next wave of UPI, and that includes all aspects, including reaching new users," Asbe said during his interview with TechCrunch1

1

. The applications span several critical areas. For fraud prevention, AI must be deployed to protect current citizens, detect fraudulent activities, and identify mules operating within the payment system3

3

. Beyond security, AI will enable credit distribution to users and merchants who have established digital footprints, opening new financial opportunities across India's digital payments ecosystem.

Multilingual voice onboarding represents another frontier where AI could simplify user access. NPCI launched a voice assistant-based interactive system in 2023, though Asbe acknowledged adoption has not taken off yet2

2

. Voice models need greater accuracy before they become a critical component in the payment ecosystem, but the potential remains significant for a country with diverse linguistic needs.

Small language models offer India a competitive edge

Asbe sees a substantial opportunity for India's fintech ecosystem to build small language models tailored to local contexts. "We have a very rich data set in our ecosystem," he noted. "I think there is a big opportunity for Indian companies—the banks, FinTechs, and the ecosystem—to create small language models which are sharp, specific, and as deterministic as possible"1

1

. This approach aligns with broader discussions around AI sovereignty and building models that understand India's unique linguistic and cultural landscape.

NPCI has already made progress with FIMI, the Finance Model for India launched earlier this year. This domain-specific AI model serves over a million users monthly to cancel mandates and resolve UPI-related disputes, with numbers scaling rapidly3

3

. Asbe indicated NPCI would continue leveraging open-source technology while developing additional sharp, tool-driven use cases for FIMI in collaboration with banks and the broader ecosystem.

Regulatory framework needed for AI-powered finance

As startups and public companies in the U.S. race to add AI to finance—with platforms like Coinbase and Robinhood allowing agents to trade on users' behalf—India faces questions about its own path forward. NPCI demonstrated agentic commerce and payments with Razorpay last year, but wider rollout has not followed2

2

. Asbe believes India can adopt AI-powered finance with robust regulations and frameworks that provide adequate user protection and risk mitigation. Critical to this is the ability to trace instructions and consent given by users to AI agents when things go wrong1

1

.

Market competition remains concentrated despite low switching costs

Market competition in UPI apps continues to be dominated by Walmart-owned PhonePe and Google Pay, which together control over 80% of transaction share2

2

. A regulatory market share cap limiting any single app to 30% is set to take effect on December 31, 2026, unless NPCI defers the deadline again3

3

. Asbe attributes the concentration not to high switching costs—which he says are very low—but to the absence of viable commercial models for newer entrants. "The moment we see the commercial model being available to the ecosystem, I believe newer players will start investing very heavily," he explained1

1

.

NPCI's own BHIM app, spun off in 2024 to boost competition, holds roughly 1% market share despite growing transaction volumes2

2

. Asbe clarified there is no specific target market share for BHIM, stating its purpose is to serve as a sovereign and secure alternative to other apps3

3

. As India's digital economy scales and investors worldwide watch the regulatory landscape, the success of AI integration and the development of viable business models will determine whether the next era of digital payment growth can deliver on its ambitious targets.