India’s virtual digital asset industry is still discussed as a market of users, volumes and prices. The quieter change this year is operational. Participation is being asked to show the systems underneath it, and the clearest proof came from the bond market, not from a trading screen.
In September, SEBI and the Reserve Bank of India put three tokenised corporate bonds on a permissioned ledger and settled them against wholesale e-rupee. REC, L&T and IIFL raised ₹1,025 crore between them, issued as native digital tokens on a ledger run by India’s statutory depositories, NSDL and CDSL, with the funds leg settled through the RBI’s Unified Market Interface. The bonds kept their ISIN, coupon, rating and investor rights. What changed was the settlement, atomic delivery-versus-payment, where the securities leg and the cash leg move at the same moment instead of across separate systems with a reconciliation gap. The asset attracted the announcement, and the infrastructure decided whether the trade could close.
The same test now sits under ordinary VDA participation. Identity verification, continuous due diligence, transaction monitoring, and the ability to reconstruct a transfer across rupees and virtual assets are no longer paperwork at the edge of the business. They are the conditions on which participation is allowed to continue. An industry can add users and still be incomplete, because scale is not the same thing as maturity. Maturity is whether the pipes can hold capital when the order size, and the scrutiny around it, rises.
Trading is the first pipe
A retail ticket can live with friction that a larger book will not. What that book needs is execution, visible liquidity, a price that can be explained, surveillance, risk limits and a system that stays up when the tape is violent. Digital-asset platforms have historically held functions that listed markets split across exchange, clearer and custodian, concentrated in a single technology stack. That concentration makes governance a design choice, not a policy afterthought, since order matching, risk checks and logging carry more systemic weight here than on a listed exchange where those functions sit with different, independently regulated entities. A CIO should look for what happens to an order when the market gaps, and whether it can be reconstructed the next morning. The credible answer is a limit that held and a log that still exists.
Custody is where confidence stops
Holding the asset is a key-management problem. Trusting the holding is a governance problem: ownership records, access control, recovery, segregation from the platform’s own assets, and a cyber response not dependent on one person.
This is also where banking comfort stops. A bank or custodian counterparty does not need a view on the token. It needs to know who can move the asset, how that authority is split, typically through multi-signature or threshold-based key management rather than a single controlling key, and how the holding would be evidenced under stress. The Reserve Bank’s June 2026 Financial Stability Report treats digital assets as a question of market functioning and operational links with the existing system, not a product story.
The file and the rail
Monitoring obligations turn compliance into a standing function, not a one-time check. Screening, transaction-trail data, and a file that can be rebuilt across the fiat leg and the virtual-asset leg are now a systems requirement as much as a regulatory one. A platform that cannot show the path of a transfer will be treated as a venue, not infrastructure.
Value still has to move from the rupee system into a VDA network and back, carrying its own compliance overhead at every crossing. Reconciliation, reporting and a withdrawal that can be explained are harder than connectivity itself. The riskiest transaction is often the last one off the platform. UPI, IMPS and RTGS already move rupees cheaply within the regulated system. New settlement rails belong where the old ones are slow, in wholesale and cross-border settlement, not as a replacement for a domestic system that already works.
A bond that stayed a bond
Tokenisation is the cleanest proof that infrastructure, not the wrapper, is the subject. A tokenised bond is still a bond. Demat 2.0 did not improve the credit of REC or L&T. It tested whether ownership could be recorded and cash could move at the same moment, inside institutions that already exist, using the same depositories and the same legal instrument investors already trust. If the legal claim is unclear, or the record cannot be produced, the technology has not matured the market. It has only changed the screen.
The indicators that matter are operational, and they translate directly into what a CIO should look for in a platform. Order execution that holds up when liquidity is thin, degrading gracefully rather than failing silently. Unusual activity surfaced before it becomes a report, not after the fact. Assets demonstrably segregated, with an audit trail independent of the platform’s own operators. Transfers reconstructed for a bank, regulator or auditor without manual reassembly. A path back to rupees that does not depend on a manual exception every time.
India’s virtual digital asset opportunity is not a demand story waiting for a product. It is an infrastructure story waiting for a standard. The industry will mature only as far as the pipes allow a serious counterparty to stay.