NextFin News - Visa is repositioning itself in India not as a card network defending turf, but as a trust layer for every kind of payment — from plastic to QR codes to AI agents acting on a consumer's behalf. The shift comes as the Unified Payments Interface, India's state-backed real-time payment rail, processes more than 24.5 billion transactions in a single month and captured nearly half of all real-time payments worldwide last fiscal year.
The central question is no longer whether UPI has won everyday payments in India. It has. The question Visa's leadership is answering in 2026 is whether the company can profit from a payments market whose center of gravity has moved away from the four-corner card model it was built on.
The Playing Field: A Card Giant in a UPI Nation
India's payments landscape has split into two economies, and the numbers show just how wide the gap is. The National Payments Corporation of India reported 24.51 billion UPI transactions in August 2026, worth 29.82 trillion rupees. Credit cards, by contrast, recorded 537 million transactions in January 2026 — roughly one card payment for every 46 on UPI.
That arithmetic matters because it defines the battlefield. UPI connects bank accounts directly through QR codes and phone numbers, bypassing card networks and their interchange fees entirely. For Visa, which commands a 43.5% share of India's credit card service-provider segment according to a market research report, the growth story in everyday retail payments has effectively been outsourced to a public utility.
Yet the card business is far from dead, and that nuance is where Visa's strategy actually lives. India's credit card base grew from 118.63 million cards in March 2026 to 122.86 million by July 2026, according to Reserve Bank of India data relayed through financial media. Monthly spending has settled near 2 trillion rupees — average monthly spend reached 2 trillion rupees between January and July 2026, up from 1.93 trillion in 2025, with July alone hitting 2.08 trillion rupees. The India credit card market, valued at $20.1 billion in 2025, is projected to reach $39.5 billion by 2034, a compound annual growth rate of 7.49%.
The concentration is striking. HDFC Bank, India's largest card issuer with a 28.4% share of the spending market as of January 2026, runs its premium Infinia, Regalia and Millennia cards on the Visa network, according to the same market research report. Visa's Indian strategy therefore rests on a narrow but lucrative slice: affluent urban spenders, premium rewards, and cross-border transactions where a global network still beats a domestic rail.
Visa is not pretending it can out-volume UPI at the merchant counter. Instead, it is playing a different game — one fought in higher-value segments, in value-added services, and in the infrastructure layer beneath the transaction itself.
The Pivot: From Network Owner to Trust Infrastructure
Visa's India plan is best understood as a vertical migration. Where the company once made money by owning the rails, it now aims to make money by owning the assurance wrapped around the rails — authentication, tokenization, fraud screening, and dispute resolution. This is the same logic driving its global agentic-commerce push, and India is where the logic faces its hardest test.
On June 10, 2026, at the Visa Payments Forum in San Francisco, Visa announced a strategic collaboration with OpenAI to enable secure Visa payments within agentic commerce. The company also rolled out Visa Intelligent Commerce, a platform designed to give AI agents the trust, controls and connectivity needed to discover, initiate and complete transactions.
"AI will transform commerce more profoundly than the internet or mobile technology ever did," said Jack Forestell, Visa's Chief Product and Strategy Officer. "As AI agents become active participants in the economy, Visa's focus is to ensure transactions are trusted, secure and seamless. That's the infrastructure we're building with partners like OpenAI."
The India relevance is not that Indian consumers will soon have AI agents shopping for them at scale. It is that the same capabilities — tokenized credentials, agent scoring, a large transaction model trained on billions of transactions, real-time fraud monitoring — are the building blocks of a payments infrastructure that can sit on top of any rail, including UPI.
Tokenization is the clearest example. The Reserve Bank of India's card-on-file rules, effective October 1, 2022, forced merchants to delete stored card numbers and shifted credential storage to the networks themselves. The central bank permits only authorized card payment networks to operate tokenization services — in practice Visa and Mastercard. Every time a card is now used online, on a wearable, or inside an app, the network is already inside the transaction even if the consumer never sees a Visa logo.
That structural shift turns tokenization into a wholesale infrastructure service. It also gives Visa a foothold in a transaction flow it does not own end to end: the bank and the merchant may settle over UPI or a card rail, but the credential lifecycle runs through the network's token vault.
Visa has also introduced passkey-based authentication for card payments in India, replacing passwords with device-bound cryptographic keys. The move is defensive and offensive at once: it protects card transactions from account takeover, and it keeps Visa's security stack embedded in the checkout flow as wallets and super-apps try to own the customer interface.
Why This Is Structural, Not Cyclical
The most important judgment in this story is whether UPI's rise is a cyclical wave that will crest, or a structural regime change that will not revert. The evidence points decisively to structural.
A cyclical shift requires a self-correcting mechanism — a point at which rising costs, congestion, or consumer fatigue push activity back toward the incumbent. UPI shows none of these. Its marginal cost to merchants sits near zero under the regulator's zero-fee framework for most consumer transactions. Its QR codes require no point-of-sale hardware. And its network effects compound: more users attract more merchants, which attract more users.
History no longer applies. In mature card markets, Visa's playbook was to subsidize issuance, buy merchant acceptance, and let interchange fees fund rewards that locked in the affluent. In India, the state built the acceptance network first and priced it at zero. The rewards arms race that once defended card share now partly runs on a rail Visa does not control: RuPay credit cards can be linked to UPI, letting issuers offer points and credit while settling over NPCI's infrastructure. NPCI lists "RuPay Credit Card on UPI" as a live product, and from April 2026 it reduced the fees that consumer-facing UPI apps earn on such transactions — a small adjustment in basis points, but one that tilts incentives toward the domestic scheme across billions of monthly payments.
The counter-argument deserves space. Card spending in India is still growing — the base expanded by more than 4 million cards between March and July 2026 alone, and monthly spend has held near 2 trillion rupees. Premium cards remain profitable, and cross-border spend by Indian travelers is a natural Visa stronghold given its acceptance across more than 200 countries and territories. A bull case for the old model says India is simply following the development curve of other emerging markets, where cards eventually reclaim share as incomes rise.
That case fails on the mechanism. In other markets, cards won because they solved a problem UPI now solves better: instant, interoperable, low-cost digital payment. Once that problem is solved at national scale, the incentive for a consumer to carry a card for everyday purchases disappears. Cards become a credit and rewards product, not a payment product. The volume economics of the network business — which depend on billions of small transactions — do not recover.
The structural call has a falsifying signal. If UPI's share of consumer spending stops climbing and card spend per active card accelerates materially above inflation for four consecutive quarters — indicating consumers are voluntarily choosing cards over QR for routine purchases — then the regime-change thesis is wrong. The current data points the other way: NPCI has projected 1 billion daily UPI transactions by fiscal 2026-27, and the UPI market is forecast to grow from $34.67 billion in 2026 to $603.17 billion by 2034, a compound annual growth rate of 41.47% according to the same market research report that tracks card share.
The Second-Order Play: Who Owns the Layer Above the Rail
The first-order read of Visa's India strategy is straightforward: defend premium cards, grow tokenization, sell security. The second-order question is more consequential — who owns the layer above the rail, where the customer relationship and the data live?
UPI's design makes the payment rail a commodity. The value migrates to the apps that sit on top of it — PhonePe, Google Pay, Paytm — and to the banks that hold the accounts. PhonePe alone handled 48.07% of UPI transaction volume in June 2026, with Google Pay at 33.96% and Paytm at 8.25%, according to NPCI data relayed through industry reporting. Three apps control nine of every ten UPI payments. These platforms own the interface, the merchant relationships, and the behavioral data.
Visa's agentic-commerce bet is an attempt to leapfrog that layer entirely. If commerce increasingly happens through AI agents rather than apps, then the "front end" of commerce is up for grabs again. Forestell has framed the shift in global terms: artificial intelligence is transforming the front end of commerce, while stablecoins are reshaping the back end. Visa's play is to insert its trust infrastructure — credentialing, authorization, fraud monitoring — into that new front end before any single super-app consolidates it.
The risk is timing and fit. Agentic commerce is real but early. UPI is mature, ubiquitous, and regulator-backed. A strategy that banks on a future commerce paradigm cannot afford to lose the present one. That is why Visa's India moves are deliberately dual-track: build the next-generation trust layer while defending the cash-generative premium card franchise that funds it.
There is also a geopolitical dimension. India has made its payments infrastructure a strategic export — UPI is live for cross-border payments in markets including Singapore, the UAE, Bhutan, Nepal, Sri Lanka, France, Mauritius and Qatar, according to a government press release. A domestic network that goes global reduces the need for Visa's cross-border rails over time. Visa's counter is its own global footprint: the company operates across more than 200 countries and territories, and for Indian travelers and merchants selling abroad, that acceptance remains a moat that no domestic scheme can yet match.
What to Watch
The thesis that Visa can thrive as a trust layer rather than a volume network rests on three observable signals over the next 12 to 18 months.
First, tokenization economics. If Visa can show that value-added services — tokenization, fraud screening, passkeys — are growing faster than core processing volumes in India, the migration story is working. The RBI's card-on-file mandate guarantees the credential base keeps expanding, and every tokenized transaction runs through the network's security stack.
Second, premium card resilience. Watch HDFC Bank's card metrics and the spending share of the top issuers. If premium-card spend per card keeps rising while UPI takes more of the mass market, Visa's narrow-but-lucrative positioning holds. If premium spend flattens as RuPay credit links on UPI mature and fee incentives shift toward the domestic scheme, the franchise is under genuine pressure.
Third, agentic adoption. The OpenAI partnership announced in June 2026 is a global product, not an India-specific one. Its relevance to India depends on whether AI-driven checkout gains traction in Indian e-commerce before PhonePe or Google Pay lock in agent-facing interfaces of their own.
The base case is that Visa remains highly profitable in India but grows more slowly than the payments market as a whole, trading volume share for margin quality. The upside case is that agentic commerce arrives faster than expected and Visa's trust infrastructure becomes the default layer for AI-initiated payments. The downside case is that RuPay credit links on UPI erode the high-margin card segment faster than value-added services can replace it.
Visa's India plan, in the end, is a bet that payments will always need a referee even when the game moves to a different field. The company no longer expects to own the field. It expects to own the whistle.
Explore more exclusive insights at nextfin.ai.
