ON-DEMAND
Why AI Agents Haven't Worked in Finance, and What Finally Changed
Making AI useful for finance requires three things: human-defined rules and guardrails to shape the work, consistent deterministic logic to produce repeatable outputs, and controlled reasoning—through orchestration, tools, and memory—to make agents reliable, transparent, and useful.
In this pre-recorded session, Glenn Hopper will show how to harmonize those forces through the practical example of a 13-week cash flow forecast. The example demonstrates both the insight finance must provide—judgment, assumptions, risk awareness, and decision context—and the strength of a well-built agent harness to deliver the control, transparency, and repeatability finance needs. Attendees will leave with a clearer view of how AI can accelerate finance workflows while preserving the human judgment required to move from inputs to action.
Learning Objectives:
- Define an “agent harness” and how it makes probabilistic AI agents reliable enough for finance use.
- Evaluate a working agent-driven 13-week cash flow forecast, including where deterministic controls, probabilistic methods, and human-review gates each belong in the workflow.
Presented by: Glenn Hopper, Managing Director, Head of AI, VAi Consulting
Join Ramp for their September Mini-Series
This on-demand session is pre-recorded. To earn FPAC and/or CTP/CCM credits, complete the accompanying quiz and session evaluation at the conclusion of the recording. On completion you will receive your credits confirmation.
Exclusively sponsored by
