What the Unified Markets Interface is

At the Global Fintech Fest on 8 October 2025, RBI Governor Sanjay Malhotra announced that the Reserve Bank had “conceptualised the Unified Markets Interface (UMI), as a next-generation financial market infrastructure” that “will have the capability to tokenise financial assets and settlements using wholesale CBDC.” The RBI’s annual report for 2025–26 describes it as “a multi-layer platform to facilitate tokenisation of financial assets while leveraging wholesale CBDC to enhance settlement efficiencies.”

Two ideas are combined. Tokenisation records a financial asset — a certificate of deposit, a bond, eventually perhaps gold — as a digital token on a shared ledger. Wholesale CBDC, the e₹-W, is central-bank money in digital form, available to banks and institutions rather than the public. The RBI has piloted it since November 2022 for settling government-securities trades, and since October 2023 in the interbank call money market with nine banks. Putting the asset and the money on the same platform lets a trade settle as one indivisible step: the token moves only if the rupees move, and vice versa. The RBI’s executive director P. Vasudevan called this “atomicity of settlement”, with no settlement gap.

The name echoes the Unified Payments Interface and, more recently, the Unified Lending Interface. The ambition is similar: a shared public rail that private institutions plug into, this time for wholesale money and securities markets rather than retail payments or credit.

What has actually run on it

CD figures as stated by RBI Executive Director P. Vasudevan at the Global Fintech Fest 2026 and reported by the press; bond figures from press reports of the SEBI pilot. The RBI has not published pilot-level data.

The bond pilot is the more novel design. According to reports of SEBI’s announcement, the tokenised bonds sit inside investors’ existing demat accounts, with ownership still recorded by the statutory depositories, so no new account or KYC is needed. An investor needs two things: an account enabled for “Demat 2.0”, SEBI’s upgraded framework for holding bonds as tokens, and a wholesale CBDC wallet with a participating bank. Settlement is delivery-versus-payment in e₹-W; interest and redemption are automated by smart contract; and SEBI says the bonds’ legal standing — repayment obligations and investor protections — is unchanged. Secondary trading is the next step, and retail access is to follow once the pilot has been assessed.

Alongside, the RBI has signed a memorandum on digital-asset collaboration with the Monetary Authority of Singapore, worked with the UAE’s central bank on a cross-border CBDC pilot, and joined the BIS Innovation Hub’s Project Rialto and the second phase of Project Mandala. Retail e-rupee circulation, by contrast, was only ₹1,016 crore in March 2025: the RBI’s digital-currency effort has moved decisively from shoppers to markets.

What atomic settlement changes — and what it costs

In today’s bond market, trades settle a day or more after they are agreed, with the security moving through the depository and the money through RBI’s payment systems separately. In the gap, each side is exposed to the other failing. Clearing houses such as the Clearing Corporation of India manage that risk by netting: thousands of trades are offset so that each bank pays or receives only its net position at the end of a cycle.

Atomic settlement removes the gap but also removes the netting. Every trade must be fully funded at the moment it settles, so each participant must hold enough wholesale CBDC in its wallet in advance. For a bank active in money markets, that can mean parking more liquidity than netting requires today. This is a known trade-off in central-bank work on tokenised settlement, not a flaw peculiar to India; the BIS has framed the design question as how to keep the benefits of atomicity without losing the liquidity savings of netting. How UMI handles it at scale is one of the things the pilots have not yet reported.

How India’s approach compares

Other central banks are building the same bridge between tokenised assets and central-bank money, with different designs.

• Eurosystem, Pontes. Launched on 21 September 2026, Pontes connects private DLT platforms to the Eurosystem’s TARGET services so that tokenised bonds and funds settle in central-bank money rather than in private stablecoins. Deutsche Bank, Santander and Clearstream were among the first onboarded. Later versions are to bring settlement onto a Eurosystem-run DLT platform with smart contracts. Unlike UMI, Pontes is a bridge to existing settlement rather than a single platform for both asset and money from the start.
• BIS Project Agorá. Seven central banks and more than 40 financial institutions are testing tokenised commercial-bank deposits alongside tokenised central-bank reserves for cross-border payments, keeping the two-tier banking system. Press reports of its 2026 results describe real transactions of about US$1 million across six currencies, settling in as little as 80 seconds.
• Singapore, Project Guardian and Global Layer One. MAS is working with banks including BNY, DBS, JPMorgan and MUFG on a shared, multi-purpose ledger for tokenised assets across jurisdictions, governed by common standards for assets, smart contracts and digital identity.

What distinguishes India’s version is that it is run by the central bank itself, settles in wholesale CBDC rather than tokenised deposits, and keeps legal ownership in the existing depository system. SEBI has described the bond pilot as the first to issue bonds on distributed ledger technology while keeping ownership records with statutory depositories inside the existing regulated framework. That makes it slower to build, but less likely to need legal surgery later.

The question a ledger cannot answer: what happened off it

A tokenised asset is only as good as the information written onto the ledger about the world outside it. A ledger can enforce that a bond token and its payment move together. It cannot know, on its own, whether the issuer has defaulted, whether a coupon is due after a holiday shift, whether a rating has changed, or — for gold — whether the bars behind the tokens are in the vault. Something has to tell the ledger. In blockchain jargon, that something is an “oracle”.

Public blockchains have tried to solve this without trusted institutions. The best-known example is UMA’s “optimistic oracle”, which settles prediction markets such as Polymarket. Anyone can propose an answer to a question by posting a bond — US$750 on Polymarket — and if nobody challenges it within a short window, two hours on Polymarket, the answer stands. If someone disputes it, holders of UMA’s governance token vote in a commit-and-reveal ballot, and the losing side forfeits its bond. UMA says its oracle had processed more than 100,000 assertions securing about US$50 billion in volume by October 2025.

The weakness is that whoever holds the most tokens decides the truth. In March 2025 a US$7 million Polymarket market on whether Ukraine would sign a minerals deal with the US by a deadline was resolved “yes” though no deal had been signed; a single holder cast about 5 million UMA tokens across three accounts, around a quarter of the votes in the dispute. Polymarket called it “unprecedented” but did not refund bets. UMA has since moved to “managed proposers”, a set of approved parties who propose answers, and is experimenting with AI agents to propose resolutions. A May 2026 Wall Street Journal report, as cited in later coverage, found the ten largest UMA holders controlled more than half the voting power in most Polymarket disputes.

UMI takes the opposite approach. Off-chain facts reach its ledger from regulated institutions with legal responsibility — issuers, depositories, registrars, debenture trustees, the RBI itself — whose errors carry penalties and courts behind them. For certificates of deposit and corporate bonds, whose life events are few and already reported through regulated channels, that is a sensible design. It is also why the RBI’s own list of concerns for gold — legal clarity, data privacy, concentration — matters. A gold token needs continuous proof that the physical metal exists, is of the stated purity and is not pledged twice. Neither a token vote nor a central-bank ledger can supply that; it takes vault audits and an accountable custodian, with the results written to the ledger.

What would fix this

• The Reserve Bank of India should publish UMI pilot data each quarter: number and value of transactions by instrument, participants, primary and secondary shares, settlement times and any failed or reversed transactions, so that claims of efficiency can be checked.
• The RBI and SEBI should publish a rulebook for off-chain events on UMI before adding new asset classes, naming which institution writes each life-cycle event — coupon, redemption, default, corporate action — to the ledger, how errors are corrected, and who is liable for them.
• The RBI should publish an assessment of the liquidity that atomic settlement requires compared with netted settlement through the Clearing Corporation of India, so that banks and the market know what moving more volume onto UMI would cost them in pre-funded e₹-W balances.
• Before tokenising gold, the RBI and SEBI should require independent vault audits, published at least quarterly, with bar-level holdings reconciled to tokens in issue, and name the custodian legally liable for any shortfall.
• The RBI should amend its master directions on certificates of deposit, and SEBI its debt regulations, to recognise the tokenised form explicitly, so that the legal status of these instruments rests on the rules themselves rather than on pilot announcements.

These are proposals from this blog, not recommendations any agency has adopted.

Sources and caveats

About this article: Researched, written and edited by Umashankar Triplicane Dwarakanathan, with AI research assistance; every figure is meant to trace to the primary source cited. See the Editorial Policy for how sourcing, AI use and corrections work. Nothing in this piece is investment advice.