What Is AI-SEO?

AI-SEO is the technical discipline of making business information machine-readable for AI systems through structured files, schema markup, crawler permissions, and edge delivery infrastructure. It is the technical layer that enables AEO and GEO — the implementation substrate on which AI visibility is built.

Where traditional SEO focuses on HTML content, backlinks, and page authority, AI-SEO focuses on machine-readable structured data — files that AI crawlers read directly, schema markup that AI systems parse without rendering the page, and edge infrastructure that serves these signals reliably at sub-10ms latency globally.

Traditional SEO optimises the page for the human who clicks. AI-SEO optimises the signal layer for the AI system that never visits the page — it reads the structured data directly and constructs its understanding without human-style browsing.

The Core AI-SEO Files Every B2B Business Needs

AI-SEO begins with deploying a set of machine-readable files at the domain root. These are the files AI crawlers look for first when they try to understand a business.

📄 /ai.json

Primary Entity Definition

The most important AI-SEO file. Defines the business entity — name, category, services, location, canonical URL, legal identifiers. Read by all major AI systems on first crawl.

📄 /llms.txt

LLM Instruction File

Tells large language models what the business is, what it sells, and what it permits AI to do with its content. The AI equivalent of robots.txt. See full guide: llms.txt →

📄 /entities.json

Entity Relationship Graph

Complete structured graph of all business entities — services, team, locations, certifications — with typed relationships. Input for the AI knowledge graph.

📄 /intents.json

Buyer Intent Registry

Maps every service to the buyer queries that should trigger a recommendation. Includes negative intents — queries the business should NOT appear for.

📄 /robots.txt

Updated AI Crawler Permissions

Must explicitly permit all major AI crawlers. Many businesses accidentally block them with legacy rules. See below for the full list of crawlers to permit.

📄 /ai-proof.json

Proof Registry

SHA-256 hashes and OpenTimestamps anchors for all core authority documents. Makes claims independently verifiable by any AI system or auditor.

AI Crawlers — Who to Permit in robots.txt

Most B2B websites have legacy robots.txt files that were written before AI crawlers existed. A common misconfiguration is a wildcard Disallow that blocks all crawlers including AI systems. The following crawlers must be explicitly permitted:

GPTBotChatGPT / OpenAI
Google-ExtendedGemini / Google
anthropic-aiClaude / Anthropic
PerplexityBotPerplexity
BingbotCopilot / Microsoft
CCBotCommon Crawl
FacebookBotMeta AI
OmgilibotOmgili / AI training
# AI-SEO optimised robots.txt
User-agent: *
Allow: /
 
# Explicitly permit AI crawlers
User-agent: GPTBot
Allow: /
User-agent: anthropic-ai
Allow: /
User-agent: Google-Extended
Allow: /
User-agent: PerplexityBot
Allow: /
 
# AI discovery files
Sitemap: https://yourdomain.com/sitemap.xml

Schema.org Markup — The 25 Most Important Types

Schema.org JSON-LD is the universal vocabulary for structured data. Every B2B website needs these types deployed correctly. For the complete reference, see: Schema.org for AI — The 25 Most Important Types →

  • Organization — business entity definition. Required on homepage.
  • Service — every service the business offers. Required on service pages.
  • FAQPage + Question + Answer — FAQ structured data. Enables direct AI citation.
  • BreadcrumbList — navigation context. Required on every page.
  • WebSite + WebPage — site and page identity.
  • Article + TechArticle — content type identification.
  • Person — team member profiles. Links to authority signals.
  • LocalBusiness — for businesses with physical locations.
  • HowTo + HowToStep — process and methodology content.
  • DefinedTerm + DefinedTermSet — glossary and terminology pages.

The ai.json Structure

The ai.json file is the primary entity definition that AI systems retrieve first. It must include at minimum:

{
  "@context": "https://schema.org",
  "@type": "Organization",
  "name": "Your Business Name",
  "legalName": "Your Legal Entity Name S.R.L.",
  "url": "https://yourdomain.com",
  "foundingDate": "2024",
  "areaServed": "EU",
  "services": [...],
  "sameAs": ["LinkedIn URL", "Crunchbase URL"]
}

Edge Delivery — Why It Matters for AI-SEO

Traditional websites serve content from an origin server — one location, subject to downtime, cache issues, and platform dependency. For AI crawlers that need reliable, always-current signal files, origin delivery creates risk.

EDGE INJECTOR™ deploys all AI signal files to Cloudflare Workers KV — served from 300+ edge nodes globally in under 10ms. AI crawlers always receive the current, correctly structured signals regardless of what the origin website is doing. The signal layer and the website are completely decoupled.

The Canonical AI-SEO Rule

Site = CONTENT. Edge = SIGNALS. These must be fully decoupled. The website serves human visitors. The edge serves AI signals. Changes to one never affect the other. This is the architectural principle behind the ADI™ platform. See: The AI Era Website →

AI-SEO vs Traditional SEO — What Stays, What Changes

What stays the same: Quality content still matters. Accurate business information still matters. Consistent NAP (name, address, phone) still matters. Canonical URLs still matter.

What changes: The primary audience for structured data is now AI systems, not just search engines. The format requirements are stricter — JSON-LD entity graphs, not just meta tags. The delivery infrastructure must be edge-first. The proof layer — SHA-256 and OTS anchoring — has no equivalent in traditional SEO.

Frequently Asked Questions About AI-SEO

What is AI-SEO?

AI-SEO is the technical discipline of making business information machine-readable for AI systems through structured files, schema markup, crawler permissions, and edge delivery infrastructure. It is the technical layer that enables AEO and GEO.

Does traditional SEO still matter?

Traditional SEO still matters for human search engine traffic. AI-SEO is an additional layer — not a replacement. A business needs both: SEO for search engine ranking and AI-SEO for AI system discovery and recommendation.

What is llms.txt and why does it matter?

llms.txt is a plain-text file deployed at a domain root that provides instructions and context to large language models about the business. It tells AI systems what the business is, what it sells, and what it permits AI to do with its content.

What is ai.json?

ai.json is the primary machine-readable entity definition file for AI systems. It contains structured business identity data — name, category, services, location, legal identifiers, and canonical URL — in a format AI systems can reliably parse and cross-reference.

How does edge delivery improve AI-SEO?

Edge delivery via Cloudflare Workers KV serves AI signal files from the nearest network edge node in under 10ms globally. This ensures AI crawlers always receive current, correctly structured signals regardless of the origin server's state.