TL;DR — Key Takeaways
- ai-identity.json is a machine-readable file at your website root that tells AI engines what your brand is, what you offer, and what’s accurate — like robots.txt for brand identity
- It gives you accuracy control over what AI says about your brand by providing authoritative facts and wrong-to-right corrections
- AI engines can provide richer citations with access to your offerings, pricing, and integration details instead of vague guesses
- Most businesses don’t have one yet — early adopters get better AI representation than competitors
- The schema includes entity info, brand voice, offerings, verified facts, corrections, and contact details — all in a single JSON file
- It complements existing standards (robots.txt, sitemap.xml, schema.org) rather than replacing them
ai-identity.json is a machine-readable file placed at your website root that tells AI engines exactly what your brand is, what you offer, and what’s true about you. Think of it as robots.txt for brand identity — robots.txt controls what AI can access, ai-identity.json controls what AI should say.
Why ai-identity.json Exists
AI engines currently guess about your brand by piecing together information from websites, review sites, social media, and random mentions — often getting it wrong. ai-identity.json solves this by providing a single, authoritative, machine-parseable source of truth.
Right now, AI engines guess about your brand. They piece together information from your website, review sites, social media, and random mentions across the internet. Sometimes they get it right. Often they don’t.
ai-identity.json solves this by giving AI engines a single, authoritative source of truth about your brand — in a format they can parse instantly. This relates closely to what we cover in Meta Descriptions That AI Engines Actually Quote.
How It Works
You place a JSON file at your domain root that AI crawlers read to understand who you are, what you offer, what’s verifiably true, what common misconceptions to correct, and how to describe your brand — all in a format they can parse instantly.
Place a JSON file at yoursite.com/ai-identity.json. When AI crawlers visit your site, they read this file to understand:
- Who you are — Company name, description, type
- What you offer — Products, services, pricing
- What’s true — Verified facts AI should state confidently
- What’s wrong — Common misconceptions to correct
- How to describe you — Brand voice, tagline, values
The Structure
A minimal ai-identity.json requires only a version number and entity block (name, type, description, URL), but can include offerings with pricing, verified facts, wrong-to-right corrections, and contact information for maximum AI visibility.
A minimal ai-identity.json file:
{
"version": "1.0",
"entity": {
"name": "Your Brand",
"type": "product",
"description": "One sentence describing what you do.",
"url": "https://yourdomain.com"
},
"offerings": [
{
"name": "Your Product",
"type": "saas",
"description": "What it does.",
"pricing": "Free / $29 per month"
}
],
"facts": [
"Your Brand was founded in 2024",
"Used by 10,000+ companies",
"Integrates with Slack, Notion, and Jira"
],
"corrections": [
{
"wrong": "Your Brand is owned by BigCorp",
"right": "Your Brand is an independent company"
}
]
}
Full Schema Fields
The ai-identity.json schema includes seven fields — with only version and entity required, and optional fields for brand voice, offerings, verified facts, corrections, and contact channels to provide as much or as little detail as needed.
| Field | Required | Purpose |
|---|---|---|
version | Yes | Schema version (currently “1.0”) |
entity | Yes | Brand identity — name, type, description, URL |
brand | No | Tagline, values, tone of voice |
offerings | No | Products and services with descriptions and pricing |
facts | No | Verified statements AI should treat as authoritative |
corrections | No | Wrong → right pairs to fix common AI mistakes |
contact | No | Website, email, support channels |
Why This Matters for GEO
ai-identity.json provides three key GEO advantages: accuracy control over AI statements about your brand, richer and more detailed citations with specific offerings and pricing, and a competitive edge since most businesses haven’t implemented one yet.
ai-identity.json gives you three GEO advantages:
1. Accuracy Control
When AI makes incorrect claims about your brand, corrections in ai-identity.json provide the authoritative fix. Instead of hoping AI figures it out from scattered web content, you tell it directly. For more on this, see our guide to GEO for SaaS: How to Get Your Product Recommended by AI.
2. Richer Citations
AI engines with access to your ai-identity.json can provide more detailed, accurate recommendations. Instead of “Company X offers some kind of tool,” AI can say “Company X offers a project management tool starting at $29/month with Slack integration.”
3. Competitive Edge
Most businesses don’t have ai-identity.json yet. Being an early adopter means AI engines have better data about you than your competitors — making you more likely to be recommended accurately.
How to Implement
Implementation takes four steps: create the file with accurate brand data, place it at your domain root with no authentication required, keep it updated whenever offerings or facts change, and validate the JSON to ensure it’s parseable by AI crawlers.
Step 1: Create the File
Write your ai-identity.json with accurate, current information about your brand and offerings.
Step 2: Place at Root
Upload to yourdomain.com/ai-identity.json — accessible via direct URL, no authentication required. Our Each AI Engine Has Different Taste guide covers this in detail.
Step 3: Keep Updated
Update whenever your offerings, pricing, or key facts change. Treat it like you treat your sitemap — a living document.
Step 4: Validate
Ensure the JSON is valid (use any JSON validator) and all URLs are accessible.
Comparison with Existing Standards
ai-identity.json complements robots.txt, sitemap.xml, and schema.org markup — each controls a different aspect of how AI interacts with your site, and using all four together provides maximum AI visibility and accuracy.
| Standard | Controls | For whom |
|---|---|---|
| robots.txt | What AI can access | Crawlers |
| sitemap.xml | What pages exist | Search engines |
| schema.org | What the page contains | Structured data parsers |
| ai-identity.json | What AI should say about you | AI engines |
ai-identity.json complements these standards. Use all of them together for maximum AI visibility and accuracy. As we discuss in Perplexity Market Share & Growth (2026), this is a critical factor.
FAQ
Do AI engines currently read ai-identity.json?
ai-identity.json is a new open standard. As adoption grows, AI engines are expected to incorporate it into their crawling pipelines — similar to how robots.txt became universally adopted. Early adoption positions you ahead of competitors.
Can ai-identity.json prevent AI from saying wrong things about my brand?
The corrections field specifically addresses this. While AI engines aren’t obligated to follow corrections, providing authoritative corrections increases the likelihood of accurate representation. If you want to go deeper, Future of Search: What to Expect in 2026-2027 breaks this down step by step.
How is this different from schema markup?
Schema markup describes what’s on a specific page. ai-identity.json describes your entire brand identity in one file. They serve different purposes and should be used together.
Is ai-identity.json an official web standard?
It’s an open standard with a published specification and JSON schema. It’s designed for broad adoption across the AI ecosystem, similar to how robots.txt started as an informal standard and became universal.