This article has been contributed to AI Journal by Oleksandr Potapenko, Co-Founder & CEO, and invited expert. With over 20 years of experience in corporate finance, banking, and fintech, Oleksandr brings deep expertise in building secure, compliant payment infrastructure, leading investment groups, and developing scalable, auditable systems for digital financial services. In this article, he shares his perspective on how event-sourced ledgers are reshaping financial systems, enabling transparency, regulatory compliance, and the foundation for AI-driven financial intelligence.
***
In today’s regulatory environment, proving how you arrived at a financial figure is just as important as the figure itself. While our last article discussed how AI can police financial systems, that intelligence relies on a fundamental shift in digital bookkeeping. This is a primer on that shift: the move away from ledgers that forget, to ledgers that remember everything.
The Old Way: The Erasable Whiteboard
Think of a traditional bank ledger like a single line on a whiteboard showing your account balance.
Account A Balance: £1,000
When you spend £100, the bank erases the old balance and writes a new one.
Account A Balance: £900
In database terms, this is an UPDATE command. The old value of £1,000 is gone, overwritten. To figure out what happened, auditors must dig through separate, often incomplete, log files to find the record of the £100 transaction. This is slow, messy, and makes it difficult to prove a complete history. Reconciling the entire bank’s books requires massive, overnight “batch jobs” to compare all these fragile snapshots.
The New Way: The Never-Ending Captain’s Log
Event sourcing treats a ledger like a ship captain’s log or a checkbook register. You never erase anything; you just add a new line for every event that occurs.
Aug 7, 09:21 – Deposited £1,000 into Account A Aug 7, 10:15 – Debited £100 from Account A for groceries Aug 7, 17:30 – Debited £25 from Account A for transit
Your current balance isn’t a single, stored number that gets overwritten. It’s the result you get by replaying all the events in the log from the beginning. This calculation is incredibly fast—taking just milliseconds—and it has a huge advantage: you have a perfect, unchangeable record of every single thing that has ever happened. The audit trail is the ledger.
Why This Matters for Modern Regulation
This shift from mutable rows to an immutable log isn’t just a technical preference; it’s a direct answer to the demands of today’s toughest financial regulations.
MiCA & Crypto Traceability
The EU’s Markets in Crypto-Assets (MiCA) regulation requires firms to maintain a clear chain of custody for digital assets. Every on-chain transfer (like sending Bitcoin or Ethereum) is itself an “event” with a unique transaction hash.
With an event-sourced ledger, this becomes simple. Finray’s Bitkonto, for example, captures every on-chain transfer and records it as an immutable fact in its own ledger. This creates a perfect, one-to-one mapping between the on-chain world and the firm’s off-chain books, providing the seamless traceability that MiCA demands.
Basel III G Audit Trails
The Basel III framework requires banks to prove their capital adequacy at any given point in time, especially during supervisory reviews. With the old “whiteboard” method, this meant restoring old, cumbersome backups to see what the system looked like on a past date.
With event sourcing, as used in Finray’s Corebanq, a supervisor can simply ask: “Show me the exact capital position on December 31st of last year.” The system replays the event log up to that precise moment and presents a verifiable, non-repudiable answer in seconds.
Real-Time Reconciliation: The End of the Wait
The true power of this model is realized when you combine different event streams. Imagine a single “watcher” process observing the event log from the traditional bank (Corebanq) and the crypto-asset ledger (Bitkonto) simultaneously.
This watcher constantly recalculates the total position by adding up all the events from both logs in real time. If the totals ever diverge – even by a single cent or Satoshi – an alert is raised instantly. This is real-time reconciliation. It replaces the slow, error-prone nightly batch jobs and manual checking of the past with a system that guarantees parity between different ledgers, second by second.
This immutable, event-sourced foundation is the bedrock of modern finance. It provides the data integrity and historical fidelity required to satisfy regulators and is the essential source of truth upon which advanced AI compliance tools can be confidently built.
