<!-- mobian-agent-page publisher="dailydev" canonical="https://daily.dev/posts/the-agent-can-pay-but-who-gave-it-permission--tmux5nn9g" -->

---
title: The Agent Can Pay, But Who Gave It Permission? | daily.dev
description: As AI agents gain the ability to make purchases autonomously, the real challenge isn't execution — it's authorization. Drawing on Google's Agent Payments...
canonical: https://daily.dev/posts/the-agent-can-pay-but-who-gave-it-permission--tmux5nn9g
twitter:card: summary_large_image
twitter:site: @dailydotdev
og:type: website
og:site_name: daily.dev
og:title: The Agent Can Pay, But Who Gave It Permission? | daily.dev
og:description: As AI agents gain the ability to make purchases autonomously, the real challenge isn't execution — it's authorization. Drawing on Google's Agent Payments...
og:url: https://daily.dev/posts/the-agent-can-pay-but-who-gave-it-permission--tmux5nn9g
og:image: https://api.daily.dev/og/posts/TMuX5nn9G.png
og:image:alt: The Agent Can Pay, But Who Gave It Permission?
og:image:width: 1200
og:image:height: 630
og:locale: en
---

> ## Documentation Index
> Fetch the complete documentation index at: https://daily.dev/llms.txt
> Use this file to discover all available pages before exploring further.

# The Agent Can Pay, But Who Gave It Permission?

**[Medium](https://daily.dev/sources/medium_js)** · 7 min read · 0 upvotes · 0 comments

## Summary

As AI agents gain the ability to make purchases autonomously, the real challenge isn't execution — it's authorization. Drawing on Google's Agent Payments Protocol (AP2), Visa Intelligent Commerce, and Mastercard Agent Pay, this piece breaks down the four key constraints that make agent mandates safe: scope (what the agent can buy), threshold (spending limits), policy (standing rules), and escalation (when to hand back to the human). It argues that revocation must be a first-class product feature since protocols don't solve it, that audit logs must be explainable to non-technical users, and that the mandate itself should be treated as a durable, lifecycle-managed product object — not a forgotten checkbox — visible and revocable from a single screen.

## Full article

daily.dev links to this article rather than hosting it. Read it at the original source: <https://medium.com/@saturniadesign/agentic-payments-mandate-design-fb8a4ae61015>

## Questions this post answers

### What is a mandate in Google's Agent Payments Protocol (AP2)?

A mandate is a signed statement of what an AI agent may spend, on what, within which limits, and for how long. AP2 chains three signed credentials: an Intent Mandate capturing what the user authorized, a Cart Mandate binding the exact items and price, and a Payment Mandate that the network sees and clears.

_Teams evaluating agentic checkout flows follow protocol breakdowns like this one on daily.dev._

### Does AP2 handle revoking a payment mandate once it's been signed?

No, AP2's own documentation notes that revocation is not solved inside the protocol. A signed mandate is intentionally non-repudiable so no party can deny what was agreed, meaning the signature proves the grant but does not retract it, leaving revocation as a feature product teams must build themselves.

_Anyone architecting agent payment consent flows can track gaps like this on daily.dev._

### What is prompt playback in agentic payment systems?

Prompt playback is a field in AP2 that provides a natural-language restatement of what the agent understood the user to want when it made a purchase. It serves as a starting point for explainability in audit logs, though products still need to turn it into something a non-technical user can read at a glance and challenge if incorrect.

_Developers designing explainable audit trails for agent purchases can find related design discussions on daily.dev._

## Similar posts on daily.dev

- [AI agents are learning to spend money. Who will handle the payments?](https://daily.dev/posts/ai-agents-are-learning-to-spend-money-who-will-handle-the-payments--oahxtfqem) · The Next Web · 0 upvotes · 0 comments
- [A Guide to 5 Agent Payment Protocols](https://daily.dev/posts/a-guide-to-5-agent-payment-protocols-fidninthl) · Nordic APIs · 1 upvotes · 0 comments
- [AI-to-AI Payments: Why Autonomous Agents May Need Their Own Economy](https://daily.dev/posts/ai-to-ai-payments-why-autonomous-agents-may-need-their-own-economy-bwhrntein) · DEV · 3 upvotes · 0 comments
- [How Agentic Commerce Will Change E-Commerce Forever](https://daily.dev/posts/how-agentic-commerce-will-change-e-commerce-forever-nklwisix1) · Codemotion · 1 upvotes · 0 comments

---

Tags: [#ai-agents](https://daily.dev/tags/ai-agents), [#authorization](https://daily.dev/tags/authorization)

[View this post on daily.dev](https://daily.dev/posts/the-agent-can-pay-but-who-gave-it-permission--tmux5nn9g)

```json
{"@context":"https://schema.org","@graph":[{"@type":"Organization","@id":"https://daily.dev/#organization","name":"daily.dev","url":"https://daily.dev","logo":{"@type":"ImageObject","url":"https://daily.dev/apple-touch-icon.png","width":180,"height":180},"sameAs":["https://twitter.com/dailydotdev","https://github.com/dailydotdev","https://www.linkedin.com/company/daily-dev-ltd"]},{"@type":"WebSite","@id":"https://daily.dev/#website","url":"https://daily.dev","name":"daily.dev","publisher":{"@id":"https://daily.dev/#organization"},"potentialAction":{"@type":"SearchAction","target":{"@type":"EntryPoint","urlTemplate":"https://daily.dev/search?q={search_term_string}"},"query-input":"required name=search_term_string"}}]}
{"@context":"https://schema.org","@type":"TechArticle","headline":"The Agent Can Pay, But Who Gave It Permission?","url":"https://daily.dev/posts/the-agent-can-pay-but-who-gave-it-permission--tmux5nn9g","mainEntityOfPage":{"@type":"WebPage","@id":"https://daily.dev/posts/the-agent-can-pay-but-who-gave-it-permission--tmux5nn9g"},"datePublished":"2026-07-21T07:32:18.087Z","dateModified":"2026-09-14T08:32:33.991Z","description":"As AI agents gain the ability to make purchases autonomously, the real challenge isn't execution — it's authorization. Drawing on Google's Agent Payments...","image":"https://media.daily.dev/image/upload/f_auto,q_auto/v1/posts/dcfa436d38d8d0f9b754cefdc57387e5?_a=AQAEuop","thumbnailUrl":"https://media.daily.dev/image/upload/f_auto,q_auto/v1/posts/dcfa436d38d8d0f9b754cefdc57387e5?_a=AQAEuop","isAccessibleForFree":true,"articleSection":"Medium","inLanguage":"en","publisher":{"@type":"Organization","name":"daily.dev","url":"https://daily.dev","logo":{"@type":"ImageObject","url":"https://daily.dev/apple-touch-icon.png","width":180,"height":180}},"author":{"@type":"Organization","name":"Medium","logo":"https://media.daily.dev/image/upload/t_logo,f_auto/v1/logos/medium","url":"https://daily.dev/sources/medium_js"},"commentCount":0,"discussionUrl":"https://daily.dev/posts/the-agent-can-pay-but-who-gave-it-permission--tmux5nn9g","interactionStatistic":[{"@type":"InteractionCounter","interactionType":{"@type":"LikeAction"},"userInteractionCount":0},{"@type":"InteractionCounter","interactionType":{"@type":"CommentAction"},"userInteractionCount":0}],"keywords":"ai-agents,authorization","timeRequired":"PT7M"}
{"@context":"https://schema.org","@type":"BreadcrumbList","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https://daily.dev"},{"@type":"ListItem","position":2,"name":"Medium","item":"https://daily.dev/sources/medium_js"},{"@type":"ListItem","position":3,"name":"The Agent Can Pay, But Who Gave It Permission?"}]}
{"@context":"https://schema.org","@type":"FAQPage","@id":"https://daily.dev/posts/the-agent-can-pay-but-who-gave-it-permission--tmux5nn9g#faq","mainEntity":[{"@type":"Question","name":"What is a mandate in Google's Agent Payments Protocol (AP2)?","acceptedAnswer":{"@type":"Answer","text":"A mandate is a signed statement of what an AI agent may spend, on what, within which limits, and for how long. AP2 chains three signed credentials: an Intent Mandate capturing what the user authorized, a Cart Mandate binding the exact items and price, and a Payment Mandate that the network sees and clears. Teams evaluating agentic checkout flows follow protocol breakdowns like this one on daily.dev."}},{"@type":"Question","name":"Does AP2 handle revoking a payment mandate once it's been signed?","acceptedAnswer":{"@type":"Answer","text":"No, AP2's own documentation notes that revocation is not solved inside the protocol. A signed mandate is intentionally non-repudiable so no party can deny what was agreed, meaning the signature proves the grant but does not retract it, leaving revocation as a feature product teams must build themselves. Anyone architecting agent payment consent flows can track gaps like this on daily.dev."}},{"@type":"Question","name":"What is prompt playback in agentic payment systems?","acceptedAnswer":{"@type":"Answer","text":"Prompt playback is a field in AP2 that provides a natural-language restatement of what the agent understood the user to want when it made a purchase. It serves as a starting point for explainability in audit logs, though products still need to turn it into something a non-technical user can read at a glance and challenge if incorrect. Developers designing explainable audit trails for agent purchases can find related design discussions on daily.dev."}}]}
```

