GSmart: What Treasury-Native AI Actually Means

It can summarise a cash position, draft a variance note and answer demo questions well. Then it meets funding cascades, concentration limits, covenant triggers and multi-entity approvals, and runs into a wall it was never built to see.

Ask Treasurers piloting AI right now what worries them and the same four answers come back: where the model actually processes your data, accuracy on mission-critical numbers, pricing that climbs the more your team uses the tool and governance that would not survive an audit. A 2026 Avalara survey of more than 1,500 finance leaders found only 7% of organisations prioritise governance over speed of deployment.

Treasury-native AI is built to close that gap. The piece sets out what it means in practice: AI built from inside treasury operations, where every agent proposes a move, cites the exact policy clause behind it and waits for a human decision.