BIS Agora vs mBridge · 2026 Update — Three-Pole CBDC Map & Regulatory Friction

TL;DR

The 2025-10 BIS exit from mBridge (covered in mBridge vs Agora · BIS's 2025 two-track contrast of a same-year switch) was not a project sunset — it was a handover to the participating central banks while BIS doubled down on Agora. By 2026-05, the post-handover picture is clearer: mBridge has continued under PBoC + HKMA + BoT + CBUAE + SAMA stewardship and is now actively negotiating an eighth member (Brazil’s BCB is the most-discussed candidate via DREX interop; Indonesia and Russia have been named as plausible). Agora has expanded its private-sector layer with BlackRock, Citi, and Morgan Stanley all participating in tokenization workstreams, while US Treasury sanctions concerns have introduced new political friction over which counterparties can participate. Together with Hong Kong’s Project Ensemble and Singapore’s Project Guardian, the CBDC/tokenized-money landscape has cleanly settled into a three-pole geography: Agora (Fed-anchored, tokenized-deposit architecture), Ensemble + Guardian (Asia-financial-center, hybrid wholesale), and mBridge (non-US central banks, pure CBDC architecture).

Wiki route

This entry sits under fintech index. Read it against BIS Project Agorá Overview for the Agora technical baseline, Agorá vs mBridge · Strategic Comparison of the Dual-Track Cross-Border Payment System for the prior framing, mBridge · BIS Multi-CBDC Cross-Border Payment Bridge (Overview) for the mBridge fundamentals, mBridge vs Agora · BIS's 2025 two-track contrast of a same-year switch for the 2025 inflection event, mBridge Non-USD Settlement Ring Scale Pattern · GDP Coverage as the Upper Bound for Wholesale Settlement for volume context, mBridge: List of 6 Central Banks and Their Roles (2026-Q2) for the membership map, BIS Project Ensemble / Ensemble vs mBridge dual-track for the HK leg, and BIS Project Guardian / Asia Tokenisation Infrastructure: Three Poles for the SG leg.

What’s new in 2026 vs the 2025 baseline

The 2025 narrative was “BIS picks a side.” The 2026 update has five new structural facts:

mBridge persists post-BIS exit — it is now a 5-central-bank consortium operating without BIS branding but with full operational continuity. PBoC + HKMA jointly hosted technical workstreams in early 2026.
mBridge eighth-member negotiations are active — public confirmation is limited, but Brazil’s BCB (via DREX interop) is the most-discussed candidate. If concluded, mBridge expands to 6 (Brazil) and active discussion for 7 (Indonesia BI) and 8 (Russia CBR or Iran, more politically fraught).
Agora private-sector roster expanded — BlackRock joins on the asset-manager side, Citi on the cash-management side, Morgan Stanley on the institutional-tokenization side. The original 41 commercial banks roster has grown to ~50+.
US Treasury sanctions friction has emerged — questions whether Agora-participating Russian, Chinese, or Iranian-counterparty bank branches can settle on the same ledger as US-supervised entities. OFAC has not issued specific guidance but the chilling effect on participation is real.
Three-pole map crystallizes — Agora (US/EU/JP/KR/MX/CH), Ensemble + Guardian (HK + SG, asset-tokenization-led), mBridge (China + GCC + non-US EMs).

mBridge post-BIS — operational continuity

When BIS exited in October 2025, the immediate question was whether mBridge would continue. The 2026 evidence: yes, with no apparent operational degradation.

Function	Pre-BIS exit (2024)	Post-BIS exit (2026)
MVP technical platform	BIS Innovation Hub HK Centre hosted	PBoC Digital Currency Research Institute + HKMA technology team co-hosted
Governance	BIS-coordinated, 5 central banks + observers	5-central-bank consortium, rotating chair
Compliance framework	BIS-templated AML/KYC	Each participant’s domestic framework with cross-recognition
Settlement volume	(not publicly disclosed)	Reportedly growing 2x H2-2025 vs H1-2025 (public statements by HKMA)
Onboarding pipeline	Not actively recruiting (BIS branding constraint)	Actively recruiting new central bank members
External press posture	BIS-coordinated communications	Each participant communicating bilaterally

The political read: BIS leaving was the exact constraint that mBridge needed to remove — it can now recruit Russia, Iran, Indonesia, and other politically-fraught participants without BIS member-country veto. The five remaining central banks have stronger incentives to expand than BIS had.

The “eighth member” question

Public discussion identifies several candidate central banks:

Candidate	Likelihood	Rationale	Friction
Brazil BCB (DREX)	High	Active mBridge-DREX interop pilots; Brazil already in BRICS+, low political cost for BCB	Brazil’s banking ecosystem complexity, multi-stakeholder coordination
Indonesia BI	Medium-high	Indonesia already runs Project Garuda CBDC work; large EM with non-aligned posture	Indonesia’s careful US/China balancing; need to avoid signaling alignment
Russia CBR	Medium (politically high-cost)	Russia eager for non-USD settlement; CBR has rouble CBDC pilot	US/EU secondary-sanctions risk for any participating bank touching Russian counterparties
Iran CBI	Low-medium	Iran clearly motivated; CBI has worked on rial CBDC	Maximal US sanctions exposure; even China-linked participation faces secondary risk
Kazakhstan / Belarus / Uzbekistan	Medium	EAEU coordination, RUB/CNY corridor relevance	Smaller scale; less marquee value
South Africa SARB	Low	BRICS-+ membership, but SARB historically conservative	South Africa’s banking sector US-correspondent dependence
Egypt / Pakistan	Low	Belt-and-Road interest	Domestic regulatory immaturity

The probable 2026 outcome: Brazil joins (announced or confirmed), Indonesia joins in 2026-H2 or 2027-H1, and a Russia decision is deferred for political reasons. The optics matter — adding Russia would convert mBridge from “non-US settlement infrastructure” to “explicitly anti-US sanctions infrastructure,” triggering sharper US Treasury response.

For roster baseline see mBridge: List of 6 Central Banks and Their Roles (2026-Q2) (which already includes BIS-observer-tier roster; “six” includes the BIS prior role).

Agora private-sector expansion

The original 2024 Agora announcement included 41 commercial banks (JPMorgan, Citi, HSBC, MUFG, BNP Paribas, Santander, Deutsche, UBS, and others — see BIS Project Agorá Overview). The 2025-2026 expansion has added:

BlackRock — as a tokenized-asset issuer, integrating BUIDL (Treasury MMF) and potentially private credit tokenization. Connection point: BlackRock BUIDL · Tokenized MMF Benchmark · \"Stablecoin Yield Infrastructure\" and BUIDL stablecoin issuer adoption.
Citi — Citi Treasury Services has explicitly committed to tokenized-deposit experiments within Agora. Citi’s Token Services product line is the most-developed tokenized-deposit offering among US banks.
Morgan Stanley — institutional tokenization workstream, focused on tokenized fund shares and collateral mobility. MS Wealth Management’s direct-indexing infrastructure is the institutional anchor.
State Street / BNY Mellon — custodian-side expansion (originally observer-tier, now active participant).
Mastercard / Visa — payment-network-side participation as the consumer-credit-rail integration point.

The Agora architecture remains tokenized correspondent banking: central banks issue tokenized reserves (potentially via wholesale CBDC), commercial banks issue tokenized deposits backed by those reserves, asset managers issue tokenized fund shares, and all settle atomically on a unified ledger. This is the deposit-token thesis in production (see Institutional Market Stablecoins = Only Bank-Issued Deposit Tokens Are Structurally Viable).

Why this expansion matters

The 2024 Agora was “central banks + commercial banks try tokenized atomic settlement.” The 2026 Agora is closer to “complete capital markets stack from reserve → deposit → MMF → fund → equity exposure tokenized on one ledger.” This goes beyond replacing SWIFT correspondent banking and toward replacing the entire wholesale capital-markets plumbing.

If Agora succeeds at this scale, the question becomes whether private stablecoins (USDC, USD1, USDT) become secondary-tier assets only relevant in retail / EM / agent-economy contexts. The contrast is set up in The Blockchain Industry Has Split at the DNA Level into \"On-Chain Finance\" and \"Crypto\".

US Treasury sanctions friction

A new 2026 layer: Agora’s expanded multi-jurisdictional roster has created OFAC questions that did not exist when Agora was 7 central banks + 41 G7-anchored banks.

Specifically:

Korean and Mexican central banks in the Agora roster have correspondent relationships with banks that touch sanctioned counterparties (Russian, Iranian, North Korean).
Mexican peso settlement in Agora could in principle clear transactions originating from US-sanctioned cartel-adjacent entities (a stretch but legally argued in regulatory comment).
Japanese banking participants have Russian-exposure legacy.
Citi and HSBC themselves have historical OFAC settlements; their participation in any new multi-jurisdictional ledger heightens compliance risk.

OFAC has not issued specific Agora-related guidance as of 2026-05, but Treasury speeches and public statements indicate:

Tokenized assets settled on shared infrastructure are subject to existing sanctions law.
Atomic settlement does not exempt counterparties from screening.
US-supervised participating banks must apply OFAC screening at the wallet/address level.
A central bank participating in Agora does not enjoy sovereign-immunity insulation for participating commercial banks’ compliance obligations.

The chilling effect: some emerging-market central banks may decline to join Agora because participation imposes US sanctions screening obligations on their commercial bank participants. This pushes them toward mBridge instead, which by virtue of having no US central bank doesn’t impose US sanctions screening at the platform level. See The jurisdiction list as a tool of monetary protectionism for the broader pattern.

CBDC vs tokenized-deposit architecture trade-off

The Agora vs mBridge split is increasingly understood as not just a geopolitical split but an architectural one:

Architecture	mBridge	Agora
Reserve money	Wholesale CBDC	Wholesale CBDC OR tokenized central bank reserves
Commercial money	(no commercial bank money layer)	Tokenized deposits issued by commercial banks
Counterparty model	Central bank ↔ central bank with sub-participants	Central bank ↔ commercial bank ↔ end customer atomic
Disintermediation risk for commercial banks	High (central banks could go direct to corporates)	Low (commercial bank deposit issuance preserved)
Privacy from central bank	Low (CBDC inherently visible to issuing CB)	Higher (deposits stay with commercial banks)
Settlement finality	CBDC = sovereign finality	Tokenized deposit = bank-credit risk, central bank provides reserve-leg finality
Cross-border friction	CBDC ↔ CBDC swap natively	Tokenized deposit ↔ tokenized deposit + reserve-leg swap

The mBridge model is CBDC-pure: simpler, more sovereign-direct, but it implicitly disintermediates commercial banks. This is acceptable in jurisdictions where commercial banking is state-owned or state-aligned (China, GCC sovereign-wealth-backed). It is unacceptable in jurisdictions where the commercial banking lobby is politically powerful (US, EU, Japan).

The Agora model is tokenized-deposit + CBDC hybrid: preserves commercial bank intermediation, more architecturally complex, but politically viable in commercial-bank-dominated jurisdictions. The 2026 expansion of Agora to include BlackRock, Citi, Morgan Stanley is precisely the commercial financial sector’s coordinated push to ensure that any future cross-border tokenized money infrastructure preserves the deposit-token + asset-manager layer (compare Institutional Market Stablecoins = Only Bank-Issued Deposit Tokens Are Structurally Viable).

Three-pole map — final 2026 picture

The CBDC + tokenized-money landscape is now cleanly trifurcated:

Pole	Anchor central banks	Anchor commercial / asset participants	Architecture	Reach
Agora	Fed, BoE, BoJ, SNB, Banque de France, BoK, Banco de México	JPM, Citi, HSBC, BNP, Santander, Deutsche, UBS, MUFG, BlackRock, Morgan Stanley, Mastercard, Visa	Tokenized correspondent banking + tokenized asset issuance	G7 + KR + MX + open recruitment
Ensemble + Guardian	HKMA + MAS	HSBC, Standard Chartered, DBS, UBS, JPM, Goldman, BlackRock, Apollo, Fidelity	Asset tokenization-led; hybrid wholesale	HK + SG financial-center axis; coordinated with Agora but separately governed
mBridge	PBoC, HKMA, BoT, CBUAE, SAMA, [BCB pending], [BI pending]	(limited commercial-bank layer; CBDC-direct architecture)	Pure multi-CBDC settlement	China + GCC + ASEAN-pivot + BRICS+ recruitment

The HK pole appears in both Ensemble (Asia hub) and mBridge (legacy founding) — HKMA’s strategic position is uniquely bilateral, see Ensemble vs mBridge dual-track.

The Singapore Guardian pole vs Agora is the Asian financial center triangulation documented in Asia Tokenisation Infrastructure: Three Poles.

The cross-border net is that all three poles will need some interop, eventually. mBridge ↔ Agora interop pilots have been discussed in BIS reports but not formally launched. The political path to formal interop runs through HKMA and MAS — both poles’ bridges to mBridge.

Sources

BIS Project Agora press releases (bis.org, 2024-2026)
BIS mBridge project disclosures (2021-2025)
PBoC Digital Currency Research Institute statements
HKMA cross-border CBDC initiative updates
Bank of Thailand, CBUAE, SAMA mBridge participation statements
US Treasury OFAC sanctions framework (2024-2026 updates)
BlackRock CBDC and tokenized-cash public commentary
Citi Treasury Services tokenized deposit product announcements (2025-2026)
Morgan Stanley institutional tokenization initiatives
BIS Annual Report 2025 + 2026 interim statements
Hong Kong Project Ensemble updates (HKMA)
Singapore MAS Project Guardian updates
BCB DREX-mBridge interop reports (public-presented)