Mastercard has begun a test to introduce stablecoins into the settlement process carried out after card payments.

Cointelegraph, a blockchain media outlet, reported on April 20 that Mastercard is working with U.S. financial services company SoFi Technologies to pursue a plan to settle card transactions using a regulated digital dollar, SoFiUSD.

The key to the experiment is not changing the consumer payment experience. The focus is on applying blockchain-based assets to the settlement stage in which funds move between banks after a transaction is approved. Card users would pay the same way as before, while only the settlement process handled afterward by the issuing bank and the acquiring bank would be replaced with a stablecoin.

SoFi Bank plans to settle Mastercard credit and debit card transactions in SoFiUSD. SoFi's payments infrastructure platform, Galileo Financial Technologies, is expected to support other banks and fintech issuers connected to its network so they can choose the same approach.

SoFiUSD is a dollar-pegged stablecoin issued by a U.S. nationally chartered bank. It is reported to be structured to maintain cash reserves on a one-to-one basis, and has been presented as closer to a bank-issued digital currency than tokens based on typical cryptocurrency projects.

Mastercard aims to implement this structure through its Multi-Token Network, or MTN. The network aims to support various forms of tokenised money, including stablecoins, tokenised deposits and digital representations of fiat currency. Mastercard has set out a direction of building a settlement ecosystem in which regulated digital assets operate alongside existing financial infrastructure.

A change in the settlement structure also changes the benefits banks expect. Because stablecoins operate on a blockchain, transactions could be available 24 hours a day regardless of traditional bank business hours. The idea is that this could reduce delays in cross-border payments and make liquidity management more efficient for financial institutions.

Mastercard has chosen an approach that upgrades back-end systems rather than discarding the existing card model. It would keep the flow through payment authorisation and merchant verification, while handling only final settlement with a digital dollar. The company said it will support multiple forms of tokenised money, and SoFi also presented a position that participants in the Galileo network would be able to choose stablecoin settlement.

The move also aligns with a broader trend of stablecoins expanding beyond cryptocurrency trading tools into payment infrastructure. DefiLlama data put the stablecoin market value at about $314.0 billion in March 2026, and monthly transaction volume in 2025 rose at one point to as high as $969.9 billion.

Competition among global card networks is also accelerating. Visa has already tested cross-border settlement using stablecoins such as USD Coin, or USDC, and has reviewed options that would allow companies to send payments directly to stablecoin wallets. This has made clearer a trend in which card networks seek to position themselves as connectors between traditional finance and digital assets rather than competing with blockchain.

Conditions for wider adoption still remain. Financial institutions can fully participate in stablecoin settlement only if standards are clear on issues such as reserve backing structures, redemption guarantees, anti-money laundering compliance and operational capacity. On this point, regulated stablecoins issued by licensed institutions are more likely to be adopted first.

Challenges also remain. Key issues include the complexity of system integration between banks and payment processors, regulatory differences by country, liquidity management between fiat currency and digital assets, and interoperability between blockchains and financial networks. Changes felt by consumers may be limited, but the test is drawing attention because the core competitiveness of card networks could shift from the front-end payment experience to back-end settlement efficiency.