The UPI AI growth strategy outlined by Dilip Asbe, MD and CEO of the National Payments Corporation of India (NPCI), sets AI at the centre of everything from fraud detection to credit access, as the payments body pushes toward a billion daily transactions.
Asbe made the case during an interview with TechCrunch at Mumbai Tech Week 2026. His argument: AI is not a feature add-on for UPI, it is the mechanism that gets the next half-billion users onto the platform.
The UPI AI Growth Strategy: Voice, Credit, and New Users
‘AI will be used very effectively when we look at the next wave of UPI, and that includes all aspects, including reaching new users. We must use AI effectively to protect our current citizens, to find fraud, and to find mules. AI must also be used to provide credit to all the users and merchants who have digital footprints,’ Asbe said.
Voice interfaces are part of the picture, though Asbe is clear-eyed about where things stand. ‘We must use AI to look at the voice and multilingual solutions to make onboarding simpler,’ he said, before noting that voice models will need to be considerably more accurate before adoption moves. NPCI launched a voice assistant-based interactive system in 2023 and it has yet to take off at scale.
The language dimension matters here. NPCI’s AI dispute model, FiMI (Finance Model for India), already supports English, Hindi, Telugu, and Bengali, with plans to add further Indian languages within six to eight months of its February 2026 launch, according to Asian Banking and Finance. FiMI was formally launched at the India-AI Impact Summit on 17 February 2026 and powers NPCI’s UPI Help Assistant, handling transaction disputes, mandate lifecycle management, and regulatory queries through an agentic AI framework. Asbe noted it is already serving over 1 million users and scaling fast.
On the model-building side, Asbe sees a structural opportunity for Indian finance. ‘We believe that the models will differentiate from each other based on the data sets that are made available to them. We have a very rich data set in our ecosystem. 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,’ he said.
NPCI has already moved in that direction: it has partnered with NVIDIA to develop what the two organisations describe as a payments-native AI foundation model, using NVIDIA Nemotron in the build. The collaboration is designed to sit within India’s data sovereignty requirements, according to Electronic Payments International. The goal is a foundational AI layer for the payment ecosystem, not a collection of isolated agents.
The Scale UPI Is Already Operating At
The ambition makes more sense when you look at UPI’s current footprint. According to the Press Information Bureau, UPI processed 24,161.69 crore transactions worth ₹314 lakh crore in FY2025-26, representing 30% year-on-year volume growth. The daily average across 2025 was 660 million transactions; the snippet’s earlier reference to over 750 million reflects a more recent peak rather than the annual average confirmed by the PIB. UPI accounted for 85% of India’s digital payments in FY2025-26, with 703 banks live on the platform and acceptance extended across 8 countries. The system represented 49% of global real-time transaction volume in 2025.
Monthly momentum is consistent. In November 2025, UPI recorded over 19 billion transactions valued at Rs 24.58 lakh crore, a roughly 23% year-on-year volume increase from November 2024’s 15.48 billion, according to the Economic Times.
One structural wrinkle is where that volume actually sits. Research from IIM Bangalore found that in June 2025, person-to-merchant (P2M) transactions made up 64% of UPI’s total volume but only 28% of total value, pointing to a payment mix dominated by small-ticket purchases, according to the IIM Bangalore UPI analysis. Expanding credit access via AI, as Asbe envisions, would directly address that value gap.
The Market Concentration Problem
For all UPI’s scale, PhonePe and Google Pay together hold over 80% of transaction volume. NPCI’s proposed 30% cap per app has been extended twice before; the current deadline of 31 December 2026 is the third iteration of that target, originally floated in November 2020, according to PTI via Rediff Money.
Asbe’s view is pragmatic rather than regulatory: ‘The moment we see the commercial model being available to the ecosystem, I believe newer players will start investing very heavily.’ He pointed to BHIM UPI, spun off in 2024, as a sovereign and secure reference option rather than a market-share play. Its current share sits at around 1%.
The AI-driven UPI growth strategy, then, is doing double duty: it is both the pitch for the next wave of users and the implicit argument that a richer, more capable platform creates the commercial headroom newer players need to compete. Whether the 31 December 2026 cap deadline actually holds this time is the more immediate test.
