Policy Before Prompt: Enforcing Machine-Readable Financial Rules on Autonomous AI Agents

Description and use

<p>An autonomous financial agent receives a simple instruction: pay approved suppliers before their invoices become overdue.</p> <p>It identifies an invoice, checks the beneficiary, and proposes a transfer. The reasoning looks sensible. But does the customer’s delegation permit the payment? Have the beneficiary’s bank details changed? Has another worker already consumed the remaining spending allowance?</p> <p>A prompt can describe these constraints. Trusted services must enforce them.</p> <h2> The Problem </h2> <p>“Policy before prompt” describes an order of authority. Financial rules remain

Problem

<p>An autonomous financial agent receives a simple instruction: pay approved suppliers before their invoices become overdue.</p> <p>It identifies an invoice, checks the beneficiary, and proposes a transfer. The reasoning looks sensible. But does the customer’s delegation permit the payment? Have the beneficiary’s bank details changed? Has another worker already consumed the remaining spending allowance?</p> <p>A prompt can describe these constraints. Trusted services must enforce them.</p> <h2> The Problem </h2> <p>“Policy before prompt” describes an order of authority. Financial rules remain

Outcome

<p>An autonomous financial agent receives a simple instruction: pay approved suppliers before their invoices become overdue.</p> <p>It identifies an invoice, checks the beneficiary, and proposes a transfer. The reasoning looks sensible. But does the customer’s delegation permit the payment? Have the beneficiary’s bank details changed? Has another worker already consumed the remaining spending allowance?</p> <p>A prompt can describe these constraints. Trusted services must enforce them.</p> <h2> The Problem </h2> <p>“Policy before prompt” describes an order of authority. Financial rules remain

Evidence

Policy Before Prompt: Enforcing Machine-Readable Financial Rules on Autonomous AI Agents

Source