Ripple Treasury adds governed AI agents for enterprise finance

Ripple's GSmart expansion puts policy-cited AI agents into treasury workflows, targeting the governance gap slowing enterprise adoption.

By Disrupts Editorial Team Thu, 24 September 2026 AI level 2 of 5

AI level 2 of 5: drafted by our AI editorial assistant from the company's announcement; fact-checked, edited and signed off by Mark Walker, Editorial Director. What the levels mean

A bright, modern office features rows of empty workstations with light wood desks, ergonomic chairs, and multiple computer monitors, lit by natural light from large windows.

Ripple Treasury has expanded GSmart, its treasury-native AI platform, adding orchestrated agents across forecasting, liquidity, risk, reconciliation and reporting. The move positions Ripple squarely at a fault line that is widening across corporate finance: AI capability is outpacing the governance frameworks enterprises need before they will let it near critical financial operations.

The distinction Ripple is drawing is architectural. GSmart separates deterministic financial calculation from AI interpretation. Calculation engines handle the numbers; AI layers interpret policy, surface anomalies and explain recommendations. No action executes without a named treasury policy clause being cited and a human approver signing off. The company frames this as "treasury-native AI" rather than a general-purpose agent dropped into a finance context.

The governance gap that GSmart targets

The governance problem is not specific to Ripple or to treasury. Gartner projects that the average Fortune 500 company could have more than 150,000 AI agents in operation by 2028, yet only 13 per cent of organisations believe they currently have the governance infrastructure to manage that scale. For treasury teams, where a miscalculation or an unauthorised transaction can trigger regulatory or counterparty consequences, that gap carries acute operational risk.

Ripple Treasury says early uptake within its existing customer base is meaningful. Sixty per cent of eligible customers have activated Risk Insights, which flags exposure anomalies and policy breaches, and 44 per cent are using Forecast Insights, which compares forecast against actual cash flows to identify emerging liquidity shortfalls. Those are not pilot numbers. They suggest that when governance controls are baked in from the start, enterprise treasury teams will move faster than the conventional wisdom around AI hesitancy would predict.

"Every CFO is under pressure to embrace AI, but they're equally responsible for ensuring every financial decision is explainable, governed and compliant," said Renaat Ver Eecke, SVP of Ripple Treasury. "Rather than asking customers to blindly trust an AI system, GSmart works within each organisation's own treasury policies to surface recommendations transparently, while ensuring humans remain in control of every decision."

Convergence angle: agentic finance meets digital asset infrastructure

The broader significance for cross-sector investors and capital allocators sits in what GSmart is being built alongside. Ripple Treasury launched native digital asset capabilities earlier in 2026, meaning GSmart's intelligence layer now sits above a platform that handles both traditional cash positions and digital asset balances in a single view. For CFOs managing stablecoin or tokenised-asset exposure alongside conventional treasury operations, that is a consequential convergence: agentic AI governance tooling fused with blockchain-native settlement infrastructure.

This positions Ripple in a competitive space that established treasury management vendors, core banking platforms and a growing cohort of fintech challengers are all eyeing. The pressure is structural. As stablecoins move closer to regulatory clarity in both the US and the EU, corporate treasury teams face a genuine operational question: which platform can manage fiat liquidity, digital asset custody and AI-driven forecasting under a single governance layer? Ripple is betting the answer requires all three to be native to the same system, not bolted together via integration.

For macro investors tracking enterprise software spend, the agentic-AI-in-finance category is attracting capital across the stack, from pure-play treasury platforms to the AI infrastructure providers whose compute underpins them. The governance angle is becoming a differentiator. Regulated industries, including financial services, healthcare and defence procurement, are where human-in-the-loop architectures are likely to win market share against fully autonomous alternatives, at least until liability frameworks catch up with capability. Ripple's explicit policy-gating design is as much a regulatory strategy as a product one.