If new rails were an existential threat to the card networks, the networks' shipping record would look defensive. It looks acquisitive instead, and reading what they have actually built is the fastest way to understand where this market is going.
What the networks have shipped
The record below is verified against the networks' own releases as of August 2026; product names are quoted as published.
Visa has run stablecoin settlement in production long enough to disclose a run rate: roughly 7 billion dollars annualised across settlement as of its March 2026 disclosures, roughly double the November 2025 figure, alongside more than 160 stablecoin-linked card programmes live or in development. "Visa Tokenized Asset Platform" (VTAP) launched in October 2024 with BBVA as first announced bank client. The "Visa Stablecoin Platform" launched in beta on 16 July 2026, an enterprise environment for minting, moving, and managing stablecoins with Open USD (OUSD), the 140-plus-organisation Open Standard consortium coin, as its first supported asset; on the Q3 2026 earnings call Visa said the platform will be integrated with Pismo, its cloud-native issuer-processing subsidiary, that Pismo already supports tokenised deposits, and that Visa is open to acquisitions across the stablecoin stack (Ledger Insights).
Mastercard has run its "Multi-Token Network" (MTN) since 2023, and the first live transaction on it is an Asia story: Standard Chartered Bank (Hong Kong), Mox, and Libeara completing a tokenised-deposit-against-tokenised-carbon-credit swap in the HKMA's Fintech Supervisory Sandbox. "Mastercard Crypto Credential", the verified-identity layer, went live for peer-to-peer transfers in May 2024 and extended toward stablecoin flows through 2025 and 2026, while the network enabled USDG, PYUSD, USDC, and FIUSD across its rails and folded stablecoin spending into "Mastercard One Credential". Both networks, note, joined the Open Standard consortium behind OUSD.
Around them, the adjacent incumbents are making the same shape of move. Swift's shared ledger with 17 pilot banks is correspondent banking absorbing the tokenised-deposit rail rather than fighting it. Circle's "Circle Payments Network", whose Managed Payments tier launched on 8 April 2026 so that banks and payment firms can settle over USDC without touching crypto operationally, is the attacker building scheme-like orchestration, which rather concedes the point about where the durable business sits.
The margin-migration read
Put the shipping record against the bundle economics of Part 2 and a consistent picture forms. The networks are behaving as if settlement, the thing blockchains commoditise, was never the business. The business is everything wrapped around settlement: acceptance, identity, fraud and dispute infrastructure, issuer connectivity, and the rulebook that makes strangers transact. Visa putting stablecoin mint-and-redeem inside Pismo, and Mastercard putting verified identity in front of stablecoin transfers, are both bids to own the orchestration layer above whichever settlement asset wins, collecting fees that look suspiciously like scheme fees on rails they no longer need to operate.
On this reading, the fight most commentary frames as displacement is better read as margin migration. Value moves from moving money, which is being competed toward zero on every rail at once, to trusting money: knowing the counterparty, reversing the error, insuring the fraud, and connecting the endpoints. Incumbents hold most of the assets that matter for the second job. What they lose is exclusivity over the first, and with it some pricing power, which is a real loss but a survivable one. The parties with the most to lose from this reframing are not the networks; they are the correspondent banks whose economics were mostly float and opacity, and the remittance operators priced against a 6.36% world-average benchmark that both stablecoins and scheme links undercut.
What would falsify the map
The verdicts in Part 1 are readings of current evidence, and 5 developments would force revisions.
- Nexus going live at scale on its reported 2027 timetable, which would close the stablecoin remittance window across the founding corridors faster than assumed.
- GENIUS Act licences arriving from January 2027 with bank-grade issuers attached, which would accelerate the stablecoin side of B2B in dollar corridors.
- HKDAP and HSBC's Hong Kong dollar coin finding real distribution, which would test whether a licensed bank-adjacent stablecoin can beat both cards and A2A links in a single market.
- A major jurisdiction extending interchange regulation, which would thin the card bundle from the inside and re-open point of sale.
- Swift's ledger pilots converting to production volume, which would mark the incumbents fully absorbing the rail and would largely end the displacement framing.
Watching those 5, rather than the weekly announcement flow, is the efficient way to keep this map current.