Module A — What Actually Shipped

The 30-second answer

Production tokenisation is real but confined to cash and short-term money: bank-deposit tokens, tokenised money-market and Treasury funds, repo, and stablecoins. Tokenised bonds, equities and trade finance are still pilots or have failed outright. The split that matters is that funds and stablecoins chose public chains, mostly Ethereum, while banks and market infrastructure chose permissioned ledgers and the permissioned camp is not one stack. Multi-party securities and repo settlement went to Canton; single-operator cash rails went to permissioned EVM or bank-proprietary chains, JPMorgan's Kinexys among them. The highest-volume production system in the world (Broadridge's repo platform) is permissioned, not public. Ethereum leads the tokenised-real-world-asset layer by value, but that is the asset-tokenisation layer rather than the bank-settlement layer, and the distinction decides everything downstream for India.

The 5-minute summary

What is genuinely live. JPMorgan's Kinexys moves more than $5bn a day in tokenised deposits (Dec 2025, company figure). Broadridge's Distributed Ledger Repo settled on the order of $8tn in a single month in mid-2026, the largest tokenised-asset settlement venue anywhere. BlackRock's BUIDL and Franklin Templeton's BENJI have run tokenised money-market funds on public chains for years (BUIDL since 2024 on Ethereum; BENJI since 2021, mostly on Stellar). Stablecoins (USDC, USDT, PYUSD) function as production settlement rails, and Visa settles a small but real volume in stablecoins across nine chains. These are not pilots; they are operating systems with volumes and dates.

Which asset classes moved first, and why. Money-market and Treasury products led because their credit and legal simplicity made the token a thin wrapper over an uncontroversial asset, rather than because the technology fit them better. A $1.00-NAV Treasury fund needs no price discovery and carries near-zero credit ambiguity. Tokenisation went where the law and the credit were easy, not where the technology was uniquely suited. This is the single most useful pattern for a skeptic, because it deflates the "tokenise everything" pitch using the adoption record itself.

Public versus permissioned, resolved by the production record. Of the systems genuinely in production, funds and stablecoins sit on public chains (mostly Ethereum and its layer-2s); banks and core market infrastructure sit on permissioned ledgers. The stated reason is confidentiality: institutions will not put position-sensitive activity on a ledger the whole world can read. But "permissioned" is not a single choice, and the split inside it falls along the number of counterparties in one settlement. Multi-party securities, repo and depository settlement went to the Canton Network, where each party must see only its own leg of an atomic trade (Goldman, Broadridge, DTCC, HSBC). Single-operator cash rails went to permissioned EVM or bank-built chains, where the privacy boundary is the institution's own perimeter (Kinexys, Citi, the RBI's e₹-W): JPMorgan's Kinexys, at over $3tn cumulative, descends from Quorum, an Ethereum client, so the largest bank-run cash rail in this module is Ethereum-lineage rather than Canton. Treating the permissioned camp as uniformly Canton is the error a well-briefed banker will catch first. So the reconciled reading is that Ethereum leads the global asset-tokenisation layer by value while permissioned ledgers carry the highest settlement volumes. These are two different layers, and conflating them is the most common error in writing on this subject. (Ethereum's exact share of tokenised-RWA value is quoted anywhere from a third to two-thirds depending on the source and the measure; treat it as "the largest single share, roughly half by the best-tier data," never as a precise percentage.) For India the practical consequence is that the infrastructure choice is three-way rather than two-way: Canton-class architecture for multi-party confidential settlement, a permissioned EVM for single-operator registers and rails (the lower talent risk, given India's EVM-concentrated developer pool), and public Ethereum for fund tokens and cross-border reach. Two caveats before quoting any of this: Kinexys' current stack is unverified beyond the documented Onyx-era Quorum lineage, and JPMorgan has announced, but not shipped, JPM Coin natively on Canton, which if delivered would start migrating the cash layer to Canton too.

The graveyard is the most persuasive section. Every bank-consortium trade-finance blockchain of the last decade failed (TradeLens, we.trade, Marco Polo, Contour), and the Australian exchange's blockchain settlement rebuild was scrapped after a write-off of roughly a quarter-billion dollars. Almost none failed for cryptographic reasons. They failed on consortium governance, competitor distrust, weak product-market fit, and delivery complexity. India's own bank consortium (IBBIC/IBDIC) is structurally the same model and carries the same latent failure modes. That a module researching this subject returns a large graveyard is the signal that the analysis is not advocacy.

The full report

This is the module as originally researched. Where a figure or framing was revised in the reconciliation pass, the summary tiers above reflect the revision; the most important case is the Ethereum share-of-RWA figure, which should be read as "largest single share, roughly half" rather than any single percentage. See the Reconciliation tab and the Figure Ledger for specifics.

Overview

The production reality is narrow but real: tokenised cash and short-term money (bank-deposit tokens, tokenised money-market/Treasury funds, repo, stablecoins) is live at scale, while tokenised bonds, equities and trade finance remain pilots or have failed. JPMorgan Kinexys moves >$5bn/day (Dec 2025, T4/promotional); Broadridge DLR settled $8.0tn in July 2026 (T1); tokenised RWAs (ex-stablecoins) totalled ~$33.5bn (July 2026, T2 RWA.xyz).

Ethereum is the leading venue by value but not a monopoly. RWA.xyz shows Ethereum hosting the largest single share of tokenised RWA value (~$17.3bn of the tracked total); the most defensible chain-level estimates put Ethereum mainnet at roughly 58–63% of tokenised RWA value (Aug 2026, T5), with at least one outlier estimate as low as ~34%. The highest-volume bank systems (Kinexys, Citi Token Services, Broadridge DLR, HSBC Orion, Goldman GS DAP) run on private permissioned ledgers, not public Ethereum.

The graveyard is the most useful part for a skeptic: TradeLens, we.trade, Marco Polo, Contour and the ASX CHESS replacement all failed, almost none for cryptographic reasons. They failed on consortium governance, competitor distrust, weak product-market fit and delivery complexity. The identical failure modes are latent in India's IBBIC/IBDIC consortium.

Asset management and funds

BlackRock — BUIDL (USD Institutional Digital Liquidity Fund). Status: IN PRODUCTION. A tokenised money-market-style fund holding US Treasuries, cash and repo, issued as an ERC-20-style token by Securitize acting as transfer agent. One token targets a stable $1.00 value; yield is distributed on-chain. Institutional cash-management vehicle, qualified purchasers only. Launched March 2024 on Ethereum (T4 Securitize/PRNewswire). AUM crossed $1bn March 2025; peaked ~$2.9bn mid-2025; on-chain AUM shown ~$1.73bn on RWA.xyz (June 2026), and the figure varies materially by source and date (T2 RWA.xyz; T3 Yahoo Finance; T5 Messari). Expanded from Ethereum to Aptos, Arbitrum, Avalanche, Optimism, Polygon, Solana (Mar 2025) and BNB Chain (Nov 2025), reaching nine networks, cross-chain via Wormhole. Custody/access: Securitize transfer agent; allowlisted, KYC-gated wallets; qualified purchasers only. Accepted as exchange collateral (Binance, Crypto.com, Deribit) in 2025.

India translation. The analogue is a tokenised liquid or money-market mutual-fund unit: an SBI, HDFC or ICICI Prudential liquid fund with units on a ledger rather than only at the RTA (CAMS/KFintech). It would be owned by an AMC's digital/product team plus its RTA, supervised by SEBI. What blocks it: SEBI mutual-fund units are recorded through RTAs and held in demat via NSDL/CDSL, and no rule permits a public-blockchain token to be the legal register of a mutual-fund unit. What would change it: SEBI recognising a DLT record as a valid register for MF units (amendments to the SEBI (Mutual Funds) Regulations 1996 and the RTA framework), or a tokenised feeder within an AIF wrapper. Confidence: high that BUIDL is in production; medium on the current AUM, which ranges $1.7bn–$2.9bn across sources.

Franklin Templeton — FOBXX / BENJI. Status: IN PRODUCTION. BENJI is the on-chain share token of the Franklin OnChain US Government Money Fund, a US-registered '40 Act mutual fund whose blockchain is the fund's official system of record, the only one of the large tokenised money funds wrapping a public registered mutual fund rather than a private vehicle. Live since April 2021 (T4 Stellar/Franklin; T3). BENJI suite AUM $1.98bn as of 29 April 2026 (T4, promotional). Runs on up to nine public chains, Stellar primary, with an allowlist model that blocks transfers between non-allowlisted addresses at the contract level.

India translation. Same analogue and same blocker as BUIDL: no SEBI recognition of a DLT register for MF units, and FOBXX's standing rests on US SEC transfer-agent rules with no Indian equivalent. Confidence: high on production status; medium on precise current AUM.

WisdomTree, Janus Henderson, Invesco, Apollo, Hamilton Lane. WisdomTree's Treasury Money Market Digital fund (~$765m, T5) and Janus Henderson's and Invesco's tokenised Treasury funds are live and visible on RWA.xyz on Ethereum (Aug 2026, T2): IN PRODUCTION. Apollo and Hamilton Lane run tokenised private-market feeders, drawn to Polygon for lower fees; AUM not confirmable at T1–T3, so LIMITED PILOT / small-scale production. The India analogue is debt/liquid fund and AIF (Cat II) feeder units on DLT; the cleanest current route is an AIF wrapper with blockchain-based unit tracking inside an already-regulated structure. Confidence: medium, since several products are confirmed live but AUM figures for private-market products are thin at T1–T3.

Banks and market infrastructure

JPMorgan — Kinexys (formerly Onyx). Status: IN PRODUCTION. A private, permissioned platform for institutional digital payments and tokenised deposits (JPM Coin), plus a Tokenized Collateral Network and, from 2025, a tokenised money-market fund on Ethereum. Live commercially since 2020. Per Kinexys chief product officer Arif Khan, over $3tn in cumulative volume has been processed and the platform averages more than $5bn daily (Dec 2025, T4/promotional; ~$7bn/day cited April 2026, T3). For context, J.P. Morgan Payments processes >$10tn/day overall, so Kinexys is a small fraction (T3 framing). Primarily its own permissioned chain; in 2025 deployed JPM Coin on Base (a public L2) and launched a tokenised MMF (~$779m) on Ethereum. Named clients include Siam Commercial Bank, Qatar National Bank, Mitsubishi Corporation and Trafigura (T3). Stated reason for permissioned: deposits move within JPMorgan accounts, entirely inside the regulated banking system.

India translation. The analogue is tokenised bank deposits for 24/7 corporate treasury movement, which is what the RBI's October 2025 tokenised-CD pilot on the wholesale CBDC (e₹-W) layer is testing. It would be owned by a large bank's transaction-banking division (SBI, HDFC, ICICI, Axis), with the RBI as supervisor and settlement-layer provider. India already routes this through RBI infrastructure and uses a permissioned CBDC layer, not public chains; the blocker to a Kinexys-style cross-border rail is FEMA and the absence of a live cross-border CBDC bridge. What would change it: the RBI extending the e₹-W tokenised-deposit pilot to production and opening cross-border corridors. Confidence: high.

Citi — Citi Token Services. Status: IN PRODUCTION. Tokenised deposits on a private permissioned blockchain owned solely by Citi, giving corporate clients 24/7 cross-border USD and EUR movement between Citi branches; clients hold no tokens themselves. Moved from pilot to commercial October 2024 (T1 Citi). Live in the US, UK, Singapore, Hong Kong and Dublin. Named clients: Mars, Siam Commercial Bank. "Billions of dollars in transaction value since launch," with no precise daily figure disclosed (T1, but vague). Private permissioned, Citi-owned, chosen to operate entirely within the regulated banking system.

India translation. The analogue is a single-bank tokenised-deposit liquidity network for a large Indian bank's global corporate clients, owned by the transaction-banking division of an internationally-branched bank (SBI, ICICI, Axis), RBI-supervised and FEMA-relevant. Cross-border movement of tokenised deposits engages FEMA and correspondent-banking rules, with no RBI framework yet outside the CBDC pilot. Confidence: high on production; the "billions since 2024" figure is real but too vague to chart.

HSBC — Orion and Gold Token. Status: IN PRODUCTION. Orion is HSBC's private permissioned platform for issuing and settling digitally native bonds end-to-end, operated from Luxembourg and integrated with the HKMA's Central Moneymarkets Unit. It has been used for the HKMA/Hong Kong multi-currency digital green bond, Luxembourg's digital treasury certificates, and the UK government's Digital Gilt (DIGIT) pilot. HSBC Gold Token tokenises physical gold in HSBC's London vault, sold to Hong Kong retail since April 2024, with >$1bn traded (T4, promotional). Private permissioned, chosen for integration with regulated settlement and ICSD linkages.

India translation. For bonds, the analogue is tokenised G-Sec or SDL issuance, but India's G-Sec settlement (CCIL, E-Kuber, NDS-OM) already delivers near-atomic DvP in central-bank money, so the marginal benefit is small, which is why the RBI is testing tokenised CDs rather than G-Secs. For gold, a tokenised Sovereign Gold Bond faces issuance-authority ambiguity. Confidence: high on Orion/Gold Token production; medium on Indian applicability, since existing plumbing already solves much of what Orion solves.

Goldman Sachs — GS DAP. Status: IN PRODUCTION (platform); spin-out ANNOUNCED ONLY. Goldman's production platform for issuing and settling tokenised bonds and MMF tokens, built on the Canton Network using Daml, with atomic T+0 DvP settlement. Launched November 2022 via a €100m EIB digital bond. Goldman announced (Nov 2024) intent to spin GS DAP into an independent, industry-owned company, targeted mid-2026, with regulatory approvals pending as of April 2026. Canton was chosen explicitly for the sub-transaction privacy institutions demanded, and for atomic DvP.

India translation. The analogue is a tokenised corporate-bond issuance-and-settlement platform for Indian debt capital markets, owned by NSE/BSE, NSDL/CDSL, or a bank consortium, SEBI-supervised. India's corporate-bond settlement runs through NSDL/CDSL and clearing corporations, with no SEBI framework for DLT-based issuance, and privacy sits in tension with SEBI's transparency expectations. Confidence: high on platform production; the spin-out is announced-only.

Broadridge DLR, DTCC, Canton — market plumbing. Broadridge's Distributed Ledger Repo is IN PRODUCTION and the single largest institutional tokenised-RWA settlement venue by volume: ~$8.0tn settled in July 2026 (~$365bn/day, +28% YoY), on a permissioned Canton-lineage ledger, with UBS a named client and DLR data now on the Bloomberg Terminal via Kaiko (T1 Broadridge releases; T2 Kaiko/Bloomberg). DTCC received an SEC No-Action Letter and chose Canton to tokenise DTC-custodied US Treasuries, joining the Canton Foundation as co-chair with Euroclear (Dec 2025): LIMITED PILOT moving toward production. A first cross-border intraday repo using tokenised UK gilts ran in Feb 2026 with LSEG, Euroclear, DTCC, Tradeweb, Citadel Securities, Societe Generale, Archax and Cumberland DRW: LIMITED PILOT.

India translation. The analogue is tokenised tri-party repo / collateralised borrowing (India's CROMS/TREP and CCIL-operated repo), owned by CCIL and RBI-supervised. CCIL already provides guaranteed central-counterparty repo settlement, so the incremental benefit of a DLT layer is unproven and the RBI has not identified repo as a priority (it chose CDs). Confidence: high on Broadridge production volume; medium on Indian applicability.

Payments and stablecoin infrastructure

Stablecoins (Circle USDC, Tether USDT, PayPal PYUSD). Status: IN PRODUCTION as settlement infrastructure. Total stablecoin market cap ~$316bn as of 12 June 2026 (DefiLlama, T2): USDT ~$187bn (~59%), USDC ~$75bn (~24%), together ~83%. PYUSD peaked ~$4.2bn (March 2026) then contracted to ~$2.7–2.9bn as incentives tapered. Note: the US GENIUS Act (2025) prohibits interest-bearing stablecoins, which is why tokenised MMFs occupy the on-chain-yield niche instead. Stablecoin "market cap" is a supply-in-circulation measure, not an investment valuation; sources that treat it as the latter conflate the infrastructure question with the asset question, and it is treated here as infrastructure only.

Card networks and Stripe. Visa's stablecoin settlement reached a $7bn annualised run-rate, up 50% quarter-on-quarter, across nine chains (April 2026, T1 Visa), skewing to Solana and Ethereum; US launch December 2025 with Cross River Bank and Lead Bank settling in USDC over Solana: IN PRODUCTION. Mastercard announced settlement support for USDC, PYUSD and RLUSD (June 2026), early production. Stripe acquired Bridge for $1.1bn (closed Feb 2025), launched Stablecoin Financial Accounts in 101 countries, and co-launched Tempo, a payments-focused EVM-compatible L1, with Paradigm: Bridge and accounts IN PRODUCTION, Tempo LIMITED PILOT.

India translation. Stablecoins are used as back-end settlement, not consumer checkout; the Indian analogue is cross-border remittance and B2B settlement, not domestic payments, since UPI already provides instant 24/7 retail settlement. It would be owned by NPCI, banks and remittance providers, RBI-supervised. India has no legal framework for INR stablecoins, the RBI consistently favours the CBDC over private stablecoins, and FEMA restricts foreign-currency stablecoin use, while UPI removes most of the domestic pain stablecoins solve elsewhere. Confidence: high that global stablecoin settlement is in production; high that India's domestic case is weak because of UPI.

Corporate and sovereign

The European Investment Bank has issued repeat digital bonds since a €100m Project Venus issuance (Nov 2022) on GS DAP/Canton, settled T+0 with experimental wholesale CBDC: IN PRODUCTION. Hong Kong's government has issued multi-currency tokenised green bonds via HSBC Orion (>200 investors in the 2025 issuance): IN PRODUCTION. The UK government chose Orion for its Digital Gilt (DIGIT) pilot: ANNOUNCED/pilot.

Pattern analysis

Recurrent primitives, ranked by number of institutions adopting. Tokenised cash, deposit tokens and stablecoin settlement is the most-adopted primitive (JPMorgan, Citi, Visa, Mastercard, Stripe, Circle, Tether, PayPal, RBI e₹-W), because cash is the leg everyone needs first. Then tokenised money-market/Treasury funds (BlackRock, Franklin Templeton, WisdomTree, Janus Henderson, Invesco, JPMorgan); atomic settlement / DvP (Goldman, HSBC, Broadridge); tokenised collateral and intraday repo (Broadridge, JPMorgan TCN, Canton), which carries the highest volume but the fewest distinct owners; tokenised bonds (EIB, HKMA, UK, HSBC, Goldman), with many issuances but mostly one-off; privacy and verification without disclosure (the Canton adopters), the reason permissioned ledgers keep winning at banks; and trade finance / shared registry, the most-attempted and most-failed.

Why money-market products moved first. The hypothesis that they moved first because credit quality was uncontroversial, rather than because the technology fit better, is well supported. A $1.00-NAV Treasury fund needs no price discovery, no complex corporate actions, and carries near-zero credit ambiguity, so the token is a thin wrapper over an uncontroversial asset. Tokenisation went where the law and credit were easy, not where the technology was uniquely suited.

Public versus permissioned, quantified. Of the initiatives found IN PRODUCTION: funds and stablecoins chose public chains (BUIDL, WisdomTree, Janus Henderson, Invesco, JPMorgan's MMF, PYUSD/USDC, Visa partly on Ethereum; BENJI on Stellar; PYUSD on Solana); banks and market infrastructure chose permissioned ledgers. The permissioned book then splits in two, and the split is not cosmetic. Canton/Daml lineage took the multi-party venues, where many counterparties settle one atomic transaction and each may see only its own leg: Goldman GS DAP, Broadridge DLR, DTCC, HSBC Orion. Ethereum-lineage or bank-proprietary chains took the single-operator cash rails, where the privacy boundary is the institution's own perimeter: Kinexys core (Quorum-derived EVM), Citi CTS (stack undisclosed), RBI e₹-W. The highest-volume systems are permissioned; the highest-value tokenised-asset pool sits on public Ethereum.

Three-way, not two-way, for India. Because of that split, the infrastructure choice facing Indian market infrastructure is not "public Ethereum or Canton." Multi-party confidential settlement (dealer-to-dealer bonds, tri-party repo, a depository settling across competing participants) points to Canton-class architecture. Single-operator registers and rails (a bank's deposit token, a single depository, a CCIL-adjacent collateral system) are served by a permissioned EVM at far lower talent risk, since India's developer pool is concentrated in Solidity/EVM while Daml and Corda skills are scarce in every market, and a permissioned EVM preserves the migration ladder from private chain to permissioned L2 to public L2 that JPMorgan actually climbed. Public Ethereum is realistic for fund tokens with an allowlist model, and for cross-border reach via GIFT City. The trade-off in the middle option, stated honestly: Canton is a network with cross-institution composability, whereas every permissioned-EVM deployment is an island needing bilateral bridges, but that composability is bought with a governance commitment to a foundation, the axis on which the graveyard's projects died.

Sourcing caveat on the split. The Canton/Daml attributions are sourced (Broadridge T1, GS DAP T3/T5, DTCC T3, HSBC T3). Kinexys' current stack is UNVERIFIED: the Quorum lineage is well documented for Onyx, but the Nov 2024 Onyx->Kinexys rebrand may have brought undisclosed changes, and Citi has never disclosed the technology behind CTS. Confirm with a JPMorgan or Citi disclosure before citing the lineage externally. The behavioural claim, that single-operator rails did not choose Canton, stands on the deployment record regardless of stack.

The live counter-trend, stated so the split is not overclaimed. JPMorgan announced (7 January 2026) a plan to issue JPM Coin natively on Canton. It is ANNOUNCED ONLY as of this cut, but if it ships, the cash layer begins migrating to Canton as well and the two-layer split described above is a snapshot of 2020–2026 rather than a stable structure. Check before publication whether JPM Coin is live on Canton, and whether Kinexys' own chain is being retired or kept alongside it.

Ethereum's specific share. RWA.xyz shows Ethereum with the largest single network value (~$17.3bn), ahead of BNB Chain (~$5.7bn), Solana (~$4.1bn) and Stellar (~$3.3bn). On the share question sources disagree and the number must be handled carefully: estimates run from ~34% (The Block, Feb 2026) to ~58–63% (yellow.com, Aug 2026, T5); an earlier ~68% figure is an outlier and should be avoided. The measure changes the answer: by value and issuer count Ethereum leads, but by tokenised equity BNB Chain overtook Ethereum ($1.2bn vs ~$800m, 31 Aug 2026). Reconciliation note: use "largest single share, roughly half by the best-tier measure (RWA.xyz)," never a single percentage; re-pull RWA.xyz on the publication date and name the measure. All RWA figures move monthly.

Source conflation warning. Several sources discuss Ethereum's RWA share alongside ETH price and "price follow-through," conflating the infrastructure question (where tokenised assets settle) with the asset question (ETH as an investment). They are separate; this report addresses only the infrastructure question.

The graveyard

TradeLens (Maersk + IBM). WOUND DOWN Q1 2023. A shared Hyperledger Fabric platform for shipping documents. Stated reason: not commercially viable as an independent business. Probable reason: rivals would not route sensitive data through a platform carrying a competitor's name, a failure of neutral governance rather than technology.

we.trade. WOUND DOWN 2022. SME trade-finance for a 12-bank European consortium on Hyperledger Fabric. Insolvency after transaction growth proved too slow for the network effect; lost >$8m on ~$4m revenue in 2020; only two shareholder banks fully deployed.

Marco Polo Network (TradeIX). WOUND DOWN 2023. Receivables/payables trade finance for ~30–45 banks on R3 Corda. Insolvency in Ireland; cumulative losses ~$85m by 2021; a $12m Bank of America deal collapsed after FTX's failure. Probable reason: no product-market fit in paper-heavy documentary trade finance.

Contour. WOUND DOWN late 2023. Letter-of-credit digitisation with nine-bank backing; ceased operations unable to secure further funding, leaving Komgo as the sole major survivor of the 2019 "big four" trade-finance networks.

ASX CHESS replacement. WOUND DOWN November 2022. A Digital Asset/Daml rebuild of Australia's clearing and settlement system, announced 2016 as a "world-first industrial-scale blockchain." Scrapped after an Accenture review; A$245–255m (~US$165m) pre-tax write-off; software only ~63% complete. ASIC later took ASX to court. ASX is rebuilding without blockchain.

IBBIC / IBDIC (India). LIMITED PILOT. 15 (later 18) Indian banks incorporated the Indian Banks' Blockchain Infrastructure Company (June 2021), each investing ₹5 crore (~$670,000; total ~₹90 crore), on Infosys Finacle Connect, to digitise domestic letters of credit, GST invoices and e-way bills, aiming to cut LC processing from 4–5 days to hours and prevent the duplicate-financing fraud that enabled the Nirav Modi/Choksi case. Renamed IBDIC. Early pilots reportedly cut processing time ~75% (T5 for that figure, UNVERIFIED); its financing system passed the RBI Regulatory Sandbox in 2025.

Failure patterns. Consortium governance and competitor distrust (TradeLens, we.trade, IBBIC-risk); no product-market fit and weak network effects (we.trade, Marco Polo, Contour); delivery complexity and over-engineering (ASX CHESS); and weak commercial models (every trade-finance network ran losses far exceeding revenue).

India translation of the graveyard. Every bank-consortium trade-finance blockchain failed on governance and commercial viability, not cryptography. IBDIC is structurally the same model. Its mitigants: RBI involvement from the start, a domestic-only single-jurisdiction scope, and a genuine fraud-prevention driver the others lacked. What would have to be different: mandatory participation or an RBI nudge to reach network effect, a real per-transaction commercial model, and not positioning the platform around any one bank. Confidence: high that the global consortiums failed for these reasons; medium on IBDIC's current live status.

Module close

1. Findings hardest to dismiss. Broadridge DLR settled ~$8tn in a month on permissioned Canton (a listed-firm disclosure, now on Bloomberg); JPMorgan Kinexys does >$5bn/day but is a fraction of JPM's >$10tn/day total (the honest, deflating framing); BlackRock and Franklin run tokenised funds in production (five years, not a pilot); the entire first wave of bank-consortium trade-finance blockchains failed, mostly for non-technical reasons; and the RBI is doing what the evidence predicts, tokenising the simplest cash-like instrument (CDs) on a permissioned central-bank layer.
2. Strongest chart. Tokenised-RWA value by blockchain network, from the RWA.xyz overview (Ethereum ~$17.3bn against BNB ~$5.7bn, Solana ~$4.1bn, Stellar ~$3.3bn), paired with a second bar showing the highest-volume systems (Broadridge, Kinexys) sitting on permissioned ledgers. The value-versus-volume contrast is the whole module in one image.
3. Claims most likely to be challenged. "Ethereum hosts most tokenised RWA value" (best evidence RWA.xyz; present as "largest single share, roughly half," disclose the 34%–63% range); "Kinexys does $5bn+/day" (JPMorgan's own disclosure plus American Banker; caveat that it is company-reported and small vs JPM's total); "IBBIC cut LC processing 75%" (single-sourced at T5, present as "reported," fall back to the T3 design goal).
4. One-sentence summary. Tokenisation is genuinely in production for cash, money-market funds and repo, mostly on private bank-run ledgers with public Ethereum leading the smaller pool of tokenised funds, while every ambitious bank-consortium trade-finance project of the last decade has failed, and India's regulators are deliberately starting with the narrowest, safest instrument.
5. What this module could not establish. Named banks in the RBI's October 2025 CD pilot; IBDIC's current live transaction volume; precise current AUM for BUIDL and BENJI; independent corroboration of vendor volume claims (Kinexys, Citi); whether any tokenised-bond issuance beat conventional settlement on cost; and specifics for SocGen Forge's EURCV and UBS's uMINT, left UNVERIFIED pending issuer disclosures.