AQA — AI Question Answer | Open Specification for AI-Optimized Q&A

The Problem

FAQ pages are everywhere, but Schema.org FAQPage gives AI systems almost nothing to assess answer quality. There are no per-question dates, no changelog, no source citations per answer, and no declared maintenance frequency. An AI crawler cannot tell whether an answer was written yesterday or three years ago, or whether anyone is still maintaining it.

The Solution

AQA is a documented convention that enriches existing Schema.org types (Article, FAQPage, Question, Answer) with a small set of extension properties via a custom JSON-LD context. No new types are invented, no breaking changes are introduced. Validators that do not understand the AQA namespace silently ignore the extra properties. Full backward compatibility is preserved.

What AQA Adds

Signal	What it tells AI crawlers
`dateCreated` / `dateModified`	Exact freshness of each individual answer
`citation`	Authoritative source backing each claim
`changelog`	Why and when each answer was modified
`questionVersion`	Track answer substance changes
`updateFrequency`	Publisher's content review commitment
`monitoringSources`	What feeds and publications the publisher watches
`conformanceLevel`	Self-declared implementation depth
`aiUsagePolicy`	Granular AI rights: RAG, training, citation, commercial use
`contentSignature`	SHA-256 hash proving answer integrity
`ragSummary`	Token-optimized 300-char summary for vector embedding
`audienceAnswers`	Audience-specific answer variants
`dynamicEndpoint`	Real-time API for volatile data
`unansweredQueryEndpoint`	Webhook: AI sends missing questions back
`validThrough`	Per-answer expiration date
`verificationStatus`	verified / outdated / under-review status
`specVersion`	AQA spec version declaration
`updateFeedUrl`	Change detection feed pointer
`pingbackEndpoints`	Push notifications on content updates

Conformance Levels

Basic

Minimal effort

Every answer has a known age and at least one source citation. The document declares its language and author.

Standard

Moderate effort

Content is actively maintained, changes are tracked with a changelog, and industry context is explicit via sector classification and monitoring sources.

Full

Significant effort

Full provenance chain: who wrote each answer, what changed, when, why, and from where. Per-question authorship and versioned changelogs.

AQA Shield

When every question in a document includes both aiUsagePolicy and contentSignature, the document qualifies for AQA Shield. The policy declares permitted AI uses (RAG citation, training, summarization, quoting, commercial use). The signature provides SHA-256 hash verification against content alteration. Shield works at any conformance level as an orthogonal guarantee combining legal protection and cryptographic integrity.

Quick Start

A minimal AQA Basic implementation:

{
  "@context": [
    "https://schema.org",
    "https://aqa-spec.org/ns/context.jsonld"
  ],
  "@type": "Article",
  "headline": "Frequently Asked Questions",
  "author": {
    "@type": "Organization",
    "name": "Your Company"
  },
  "datePublished": "2024-01-15",
  "dateModified": "2026-03-20",
  "inLanguage": "en",
  "mainEntity": {
    "@type": "FAQPage",
    "mainEntity": [
      {
        "@type": "Question",
        "name": "Your question here?",
        "dateCreated": "2024-01-15",
        "dateModified": "2026-03-20",
        "citation": "https://source-url.com/document",
        "acceptedAnswer": {
          "@type": "Answer",
          "text": "Your answer here."
        }
      }
    ]
  }
}

Validate your implementation:

pip install jsonschema requests
python validators/validate.py your-file.jsonld

Migration Guide — step-by-step instructions to add AQA to existing FAQ pages.

V1.2 Features

Protection

aiUsagePolicy — granular AI rights declaration
contentSignature — SHA-256 integrity hash

Enrichment

ragSummary — vector-optimized summary
audienceAnswers — multi-persona variants
potentialAction — agentic actions
dynamicEndpoint — real-time API

Feedback

unansweredQueryEndpoint — missing answer webhook
validThrough — answer expiration
verificationStatus — verified / outdated / under-review

Distribution

specVersion — spec version declaration
updateFeedUrl — pull-based change feed
pingbackEndpoints — push notifications
AQA Hub Protocol — centralized update aggregation

Resources

Full Specification — the complete AQA technical specification
Migration Guide — 8-step guide from bare HTML FAQ to AQA
Crawler Recommendations — guidance for AI crawler developers
Ecosystem Integration — LangChain, LlamaIndex, WordPress plugin
FAQ vs AQA Comparison — side-by-side signal comparison
Python Validator — validate AQA blocks from file or URL (0-100 scoring)
Examples — 7 implementations across Basic, Standard, and Full
JSON-LD Context — the AQA namespace definition
JSON Schema — machine-readable validation schema

Philosophy

AQA is open, free forever, and has no vendor lock-in. Anyone can implement it, validate it, and extend it. The specification is MIT-licensed and community-driven.

The goal is to do for AI visibility what OpenGraph did for social sharing: give publishers a simple, standard way to make their content machine-understandable without breaking anything that already works.