Infrastructure is credit judgement, applied at scale

There is a persistent myth that technology and credit judgement compete — that systematising a lending business means removing the human from decisions that require experience. Our experience is the opposite.

Judgement does not fail in the abstract. It fails operationally. It fails when the valuation being relied on is not the latest one. It fails when a condition agreed on a phone call never reaches the settlement checklist. It fails when the person assessing serviceability is working from a different version of the financials than the person structuring the facility. None of these are failures of intelligence. They are failures of infrastructure.

GS Engine is the group’s answer to that problem. It is not a decisioning engine, and it was never intended to be one. It does not approve loans, score applicants or replace a credit committee. Its job is narrower and more valuable: to guarantee that every human decision in the credit chain is made with complete information, full provenance and a continuous view of the portfolio.

In practice, that means one transaction record from the first conversation to final repayment. The broker’s original submission, every valuation and title search, each version of the financial model, the credit paper, the approval conditions, the facility documents, the settlement checklist, the post-settlement covenant calendar — one file, one version of the truth, visible to everyone who is entitled to see it.

Provenance is the underrated half of this. It is not enough to have a number in a system; you need to know where it came from and when. GS Engine keeps data points attached to their sources and their dates, so a credit decision can always answer the question an auditor, a funding partner or a future workout will ask: what did we know, and when did we know it?

The portfolio view completes the loop. Individual facilities are supervised continuously — arrears signals, covenant positions, expiry horizons, exit progress — and the patterns across facilities feed back into origination appetite. When the system shows that a particular asset class or corridor is exiting slower than underwritten, that intelligence reaches the next credit decision automatically, not through anecdote.

The result is not faster credit for its own sake. Speed is a by-product. The product is consistency: the same standards applied to the hundredth transaction as to the first, at whatever pace the transaction genuinely requires. That is what it means to say infrastructure is credit judgement, applied at scale.