Every year the financial system gets faster on the inside. Instant payment schemes, 24/7 clearing, tokenized deposits, public blockchains that settle in seconds. And yet a payment that has to cross from one of these systems into another still behaves much as it did twenty years ago.
That border is what we call the seam: the point where value leaves one closed system and enters another. It is where reconciliation happens by hand, where correspondent chains add days, and where a new bilateral bridge has to be built for every new pair of counterparties.
The evidence
The problem is not hypothetical, and it is not getting solved on its own.
- In October 2025 the Financial Stability Board concluded that "it is unlikely that satisfactory improvements at the global level will be achieved in line with the 2027 Roadmap timetable" for cross-border payments. (FSB)
- CLS estimates that most of the same-day FX settlement market — at least USD 500 billion — settles bilaterally and without payment-versus-payment protection. (CLS)
- In May 2026, the first cross-border, cross-bank redemption of a tokenized US Treasury fund took four organisations, each handing the transaction to the next: the asset leg on a public ledger, the fund's redemption processing, a payment-network instruction, and a bank's correspondent settlement. (PR Newswire)
That last example is a genuine milestone. It also shows the pattern: even the most advanced participants still stitch the seam together one partnership at a time.
Why bridges have not fixed it
Most attempts to connect fiat systems and digital-asset systems take one of three shortcuts. They take custody of the value in transit. They hide the process inside a proprietary platform. Or they merge liabilities, so that no one can say with confidence who was responsible for which step.
Each shortcut works technically and fails institutionally. A regulator cannot supervise what it cannot see. A bank cannot hand client funds to an unlicensed intermediary. A treasury cannot sign off a process whose responsibility map is blurred.
A different approach: a contract, not a platform
Aros Financial Core starts from the opposite direction. AFC is not a platform that institutions join; it is the process of executing one contract between independent parties:
- the initiating institution, which completes KYC, AML and sanctions checks before anything enters;
- licensed anchors, which move money or digital assets under their own licences;
- AST, which executes the process and records every significant state on an append-only record.
Each party keeps its licence, its systems and its regulator. Each sees only the data its role requires. Funds in transit stay with licensed institutions — never with us. When the process ends, in exactly one terminal state, the AFC instance ends with it.
AFC does not ask anyone to become someone else. Banks stay banks. Supervisors keep supervising. The contract connects them.
Where this matters most
AFC is not meant to replace the large networks that already work well inside their own perimeter. Its value is at their external boundaries: where a tokenized fund meets a bank that is not in its consortium, where an FX trade falls outside a settlement window, or where a public programme needs a verifiable record of every movement from allocation to final use.
That is the seam. It is where we build.