Last week Brale and Visa announced a new project running on Canton. This is somewhat epic for a few reasons.
Visa is one of the most important payment networks on the planet. Canton is a privacy-preserving protocol built for traditional financial workflows. Brale is a stablecoin issuer that does a lot on Canton.
Before unpacking why that matters, I think it's helpful to level set on something simpler: how card settlement works today.
Most people experience card payments as instant. You tap a card, swipe a card, or enter a card online and the transaction is approved in a few seconds. From the cardholder's perspective, the money moved. From the merchant's perspective, the sale happened. From the network's perspective, however, a much larger settlement workflow still has to complete.
At the highest level, the card experience looks something like this:
sequenceDiagram
participant Cardholder as Card User
participant Merchant as Merchant
participant Acquirer as Merchant Bank / Acquirer
participant Network as Card Network
participant Issuer as Card Issuer
Cardholder-->>Merchant: [DATA] Present card for purchase
Merchant-->>Acquirer: [DATA] Authorization request
Acquirer-->>Network: [DATA] Authorization request
Network-->>Issuer: [DATA] Authorization request
Issuer-->>Network: [DATA] Approve / Decline
Network-->>Acquirer: [DATA] Authorization response
Acquirer-->>Merchant: [DATA] Authorization response
Merchant-->>Cardholder: [DATA] Purchase complete
We all use this flow every day. It is one of the mechanisms that makes the global economy work, and it is one of the reasons companies like Visa are so critical to commerce.
Under the hood, authorization is not the same thing as settlement. Authorization is the message. Settlement is the money.
A simplified fiat settlement flow looks more like this:
sequenceDiagram
participant Issuer as Card Issuer / Issuing Bank
participant Network as Card Network
participant Acquirer as Merchant Bank / Acquirer
participant Merchant as Merchant
Issuer-->>Network: [DATA] Net settlement obligation
Issuer->>Network: Transfer USD Bank Rail Net Amount
Network-->>Acquirer: [DATA] Settlement instruction
Network->>Acquirer: Transfer USD Bank Rail Net Amount
Acquirer-->>Merchant: [DATA] Merchant payout instruction
Acquirer->>Merchant: Transfer USD Bank Rail Merchant Payout
This works. It works at enormous scale. It works across countries, currencies, banks, merchants, and cardholders. It is easy to underestimate how incredible that is.
It is also not instant settlement.
There are windows. There are batches. There is operational risk. There is liquidity trapped between authorization, clearing, settlement, and merchant payout. Over time, companies like Square and others improved the end-user experience by using balance sheet and capital to speed up the appearance of money movement. I wrote about this idea years ago in Solving for Liquidity. You can make money feel faster by putting more money in the middle.
Stablecoin settlement gives us another option.
Instead of only speeding up old settlement with capital, you can rewire part of the settlement layer itself. In reverse, that means the fiat settlement flow can become a stablecoin settlement flow:
sequenceDiagram
participant Issuer as Card Issuer / Issuing Bank
participant Visa as Visa / Card Network
participant Brale as Brale (Issuance, Signing & Settlement)
participant Canton as Canton Network
participant Acquirer as Merchant Bank / Acquirer
participant Merchant as Merchant
Issuer-->>Visa: [DATA] Net settlement obligation
rect rgba(200, 120, 50, 0.08)
Visa-->>Brale: [DATA] Settlement instruction
Brale->>Canton: Transfer SBC Canton Net Settlement Amount
Canton->>Acquirer: Confirm Receipt
end
Acquirer-->>Merchant: [DATA] Merchant payout instruction
Acquirer->>Merchant: Transfer USD Bank Rail Merchant Payout
Or, in a more fully stablecoin-native version:
sequenceDiagram
participant Issuer as Card Issuer / Issuing Bank
participant Visa as Visa / Card Network
participant Brale as Brale (Issuance, Signing & Settlement)
participant Canton as Canton Network
participant Acquirer as Merchant Bank / Acquirer
participant Merchant as Merchant
Issuer-->>Visa: [DATA] Net settlement obligation
rect rgba(200, 120, 50, 0.08)
Visa-->>Brale: [DATA] Settlement instruction
Brale->>Canton: Transfer SBC Canton Net Settlement Amount
Canton->>Acquirer: Confirm Receipt
end
rect rgba(200, 120, 50, 0.08)
Acquirer->>Merchant: Transfer SBC Canton Merchant Payout
end
Merchant-->>Acquirer: [DATA] Receipt acknowledged
The consumer experience does not need to change.
The cardholder still taps, swipes, or checks out online. The merchant still gets paid. The issuer still funds its obligation. The network still coordinates the flow.
The difference is what happens underneath.
A settlement transfer on Canton can complete in seconds and runs 24/7. In our experience at Brale, Canton transfers complete in roughly 13 seconds. That means settlement can move outside the constraints of banking-hour windows and batch-based systems.
The important part about Canton is not only speed. Public blockchains are very good at making activity visible. That is useful in many contexts, but it is not how institutional finance works. Card settlement is not supposed to broadcast sensitive transaction and counterparty data to the world.
Canton is interesting because it is privacy preserving. The parties who need to see a workflow can see it. The parties who do not need to see it do not. That makes the experience feel much closer to how banks already work with core providers, payment processors, and third-party infrastructure today.
From the outside, the flow is still simple:
sequenceDiagram
participant Cardholder as Card User
participant Issuer as Card Issuer
participant Network as Card Network
participant Bank as Merchant Bank
participant Merchant as Merchant
Cardholder-->>Issuer: [DATA] Card transaction
Issuer-->>Network: [DATA] Authorization and settlement obligation
Network-->>Bank: [DATA] Settlement and payout instruction
Bank-->>Merchant: [DATA] Merchant paid
That is the point.
The abstraction does not have to break for the end user. The rails underneath can get better while the cardholder and merchant experience stays familiar.
This is why the Visa, Brale, and Canton work matters. It is not a science project about crypto replacing cards. It is a practical settlement upgrade for institutions that already understand cards, settlement, risk, reconciliation, and balance sheets.
Stablecoins are useful here because they can do two things businesses care about:
- Increase efficiency.
- Create new revenue.
That's it. That is the whole game.
If a bank, credit union, processor, or network can settle faster, hold less trapped liquidity, reduce operational risk, and introduce programmable settlement workflows, that is good business. If they can do it with a compliant stablecoin issuer, with privacy-preserving infrastructure, and without changing the end-user card experience, that starts to look less like a crypto experiment and more like a better settlement system.
The regulatory environment matters too. With GENIUS now creating a federal framework for payment stablecoins, and broader market-structure work continuing in DC, banks and credit unions have a much clearer path to evaluate stablecoins issued by regulated, high-quality issuers. Reserve design matters. Compliance matters. Balance sheet treatment matters. Offshoring and bad reserve structures do not belong in this part of the market.
That is one of the reasons Brale exists.
Brale handles the issuance infrastructure, key management, signing, Canton settlement, and network configuration so participants can onboard and integrate through APIs. The goal is not to make every financial institution become a blockchain company. The goal is to let financial institutions use stablecoin settlement where it makes business sense.
Rewiring the system is now a choice.
And in this case, it is a choice that can make settlement faster, more private, more programmable, and more useful for the institutions that already move the world's money.
It is fun to be building that with the Visa team.