# Verifiable Intent (VI) vs Agent Payments Protocol (AP2)

The map lists them as complementary: sites and agents commonly implement both. They answer different questions: Verifiable Intent is a step 9 (Account) standard, Agent Payments Protocol step 7 (Pay), so they are not substitutes.

[9 Account](/category/account/) [Verifiable Intent](/protocols/vi/) · [7 Pay](/category/pay/) [Agent Payments Protocol](/protocols/ap2/)

|  | [Verifiable Intent](/protocols/vi/) VI | [Agent Payments Protocol](/protocols/ap2/) AP2 |
| --- | --- | --- |
| Lifecycle step | 9 Account: Prove what happened; be judged | 7 Pay: Settle the bill |
| Answers | After an agent completes a purchase on a user's behalf, can the merchant, issuer and consumer all point to the same cryptographic record of what the user actually authorized? | Did the user actually authorize what this agent is about to buy and pay for, and can the merchant prove it later? |
| Governed by | Mastercard; contributed to the FIDO Alliance Payments TWG (2026-04-28); built on FIDO Alliance, EMVCo, IETF and W3C primitives | Google; core specification donated to the FIDO Alliance (Payments TWG and Agentic Authentication TWG), 2026-04-28; the google-agentic-commerce/AP2 repository now hosts only samples and an SDK, not the normative spec text |
| Status (issuer's words) | Draft (v0.1); open-sourced specification and reference implementation | Security/authorization protocol for agent-performed payments; core spec now stewarded by FIDO Alliance |
| Version | 0.1-draft (2026-02-18) | v0.2.0 (2026-04-28) |
| Form / transport | document format | endpoint |
| Found at | none | none |
| Implementation | prototype | multi-implementation — Google (launch blog post, 60+ launch partners) |
| Evidence | 4 verified | 4 verified, 2 reported |
| Primary use case | Let agents buy on behalf of a person | Let agents buy on behalf of a person |
| Last verified | 2026-10-23 (current) | 2026-10-23 (current) |
| Specification | [https://verifiableintent.dev/spec/](https://verifiableintent.dev/spec/) | [https://raw.githubusercontent.com/google-agentic-commerce/AP2/main/docs/ap2/specification.md](https://raw.githubusercontent.com/google-agentic-commerce/AP2/main/docs/ap2/specification.md) |

## When to use which

- Choose Verifiable Intent when your question is “After an agent completes a purchase on a user's behalf, can the merchant, issuer and consumer all point to the same cryptographic record of what the user actually authorized?” and you accept its maturity: status "Draft (v0.1); open-sourced specification and reference implementation", implementation prototype.
- Choose Agent Payments Protocol when your question is “Did the user actually authorize what this agent is about to buy and pay for, and can the merchant prove it later?” and you accept its maturity: status "Security/authorization protocol for agent-performed payments; core spec now stewarded by FIDO Alliance", implementation multi-implementation (Google (launch blog post, 60+ launch partners)).

Guidance is derived from the two records (question, status, implementation, prerequisites, pairings), not from opinion. Each record's page carries its numbered references.

## Read the full articles

- [Verifiable Intent](/protocols/vi/): After an agent completes a purchase on a user's behalf, can the merchant, issuer and consumer all point to the same cryptographic record of what the user actually authorized?
- [Agent Payments Protocol](/protocols/ap2/): Did the user actually authorize what this agent is about to buy and pay for, and can the merchant prove it later?

Ask the MCP server to compare any two: the compare-protocols prompt, or get on both ids. [Connect](/connect/).

[vi.json](/data/protocols/vi.json) · [ap2.json](/data/protocols/ap2.json) · [Markdown](/compare/vi-vs-ap2/index.md)