Four places a language model does work inside the engine. In every one it proposes, a person confirms, and the ledger never moves on the model's word.
Every direct debit and direct credit is classified by rules first: is this an established series, has the payee's bank account just changed, how large is it against its history. The model then writes a plain-English explanation of each finding for the approver. The approver decides. The rules decide without the model if it is unavailable.
Approver reads, approver decides
Standard formats, ISO 20022 camt.053 and SWIFT MT940, are parsed outright. For a bank's own CSV or spreadsheet export, the model reads a sample of rows and drafts a format profile: which column is the date, the amount, the reference, the sign. A person reviews the profile once; every later statement from that bank imports on it.
Drafted once, reviewed once, reused
A statement line that no matching rule recognises goes to the matching assistant, which proposes the account it belongs to with a confidence. Confirm is what moves the money, and confirming teaches the matcher, so the next line from the same payer is proposed the same way.
Proposed by the model, confirmed by a person
Describe a new financial instrument in plain language. The designer reads the configuration model, drafts the account type, prototypes the capture screens in a sandbox and forecasts it to maturity. When the numbers are right it goes live at runtime, as configuration.
Drafted by the model, proven by the forecaster
No posting, payment, match or configuration reaches the ledger on the model's output alone. The confirm, approve or deploy step is a person's, and it is on the event.
Where a rule can answer, it does. The model explains, drafts or proposes where a rule cannot, and a rules-only fallback takes over when the model is unavailable.
Each use is a capability your organisation turns on or leaves off. Off means every proposal comes from a rule or from a person.
The model sees what the task needs: an instruction's amount, payee and rule findings; a sample of statement rows; a description of an instrument. Anthropic's Claude today, behind one client.
What this means for you: the tedious half of the back office, reading statements, explaining exceptions, matching money, gets faster, while the accountable half stays with your people and on the audit trail.