This specification is published and recommended for implementation. Backwards-compatible additions may occur in MINOR versions; breaking changes only in MAJOR versions, with deprecation notice. See specification conventions for status definitions.
faq-ai.txt Specification
AI-Optimised FAQ File Format
This specification defines the structure and requirements for faq-ai.txt files — plain text files containing structured question-answer pairs optimised for AI consumption. The file provides factual, citation-ready answers to common questions about a business, enabling AI systems to respond accurately to user queries.
§1 Overview
What This File Does
The faq-ai.txt file provides pre-written answers to frequently asked questions in a format optimised for AI consumption. Unlike website FAQ pages designed for human browsing, this file:
- Uses clear, unambiguous question-answer formatting
- Provides factual, direct answers without marketing language
- Includes metadata for currency and source verification
- Is structured for reliable parsing by AI systems
Why It Matters for AI Visibility
When users ask AI systems questions about a business, the AI must either:
- Synthesise answers from scattered website content (error-prone)
- Use pre-written, authoritative answers from
faq-ai.txt(reliable)
A well-maintained faq-ai.txt file ensures AI systems provide accurate, up-to-date answers.
Key Principle
Content in faq-ai.txt SHOULD be factual and citation-ready. This is not marketing copy — it's an authoritative reference that AI systems can confidently cite.
Answers SHOULD be written as if they will be quoted directly. Avoid subjective claims, superlatives, and promotional language.
§2 File Location
Primary Location
The faq-ai.txt file MUST be placed in the website's root directory:
https://example.com/faq-ai.txtURL Requirements
- The file MUST be served with content type
text/plain; charset=utf-8 - The URL MUST be accessible without authentication
- HTTPS is strongly recommended
§3 Format Specification
File Format
| Property | Requirement |
|---|---|
| Encoding | UTF-8 (required) |
| Line endings | LF (Unix-style) recommended; CRLF accepted |
| Syntax | Structured plain text with Q:/A: delimiters |
| Comments | Lines starting with # are comments |
Basic Structure
# faq-ai.txt for Example Company
# Last updated: 2026-01-12
Q: What is Example Company?
A: Example Company is a technology firm specialising in cloud infrastructure services.
Q: Where is Example Company located?
A: Example Company is headquartered in London, UK, with offices in Manchester and Edinburgh.Language declaration (optional)
Publishers MAY declare the natural language of the file's human-readable content using a Lang: header line near the top of the file (above the first Q/A pair). The value MUST be a valid BCP 47 language tag:
Lang: en-GBThe header is optional. When absent, consumers SHOULD NOT assume a default language. When present, validators MUST treat the value as informational and MUST NOT fail conformance solely on language-tag content. See specification conventions §5.1 for the full rule, which is identical across every text-based AI Discovery File.
§4 Q&A Structure
Question Format
Questions MUST start with Q: followed by a space and the question text:
Q: What services does the company offer?- Use natural language questions users might actually ask
- Include a question mark at the end
- Keep questions on a single line
Answer Format
Answers MUST start with A: followed by a space and the answer text:
A: The company offers strategic consulting, operational improvement, and digital transformation services.- Answers may span multiple lines if needed
- Multi-line answers SHOULD be indented or use continuation syntax
- Keep answers factual and concise
Multi-line Answers
For longer answers, continue on subsequent lines:
Q: What industries does the company serve?
A: The company serves clients across multiple industries including:
- Financial services
- Healthcare
- Manufacturing
- Retail
Each industry team has dedicated specialists with relevant experience.Source URL Field
Each Q&A pair MAY include an optional URL: line after the answer, using markdown hyperlink format to point users to the relevant page:
Q: How do I reset my password?
A: Visit the account settings page and click "Reset Password". You will
receive an email with a reset link valid for 24 hours.
URL: [Account Settings](https://www.example.com/account/settings/)- The
URL:line MUST appear immediately after the answer, before any blank line - Use markdown hyperlink format:
[Link Text](url) - The link text SHOULD describe the destination page, not just repeat the URL
- This helps AI assistants provide direct, actionable links when citing answers
AI systems that consume faq-ai.txt files can extract both the anchor text and the URL from markdown links. This gives them context about what each link leads to — enabling more useful, clickable citations in AI-generated responses. Plain URLs lack this context.
Blank Lines
Use blank lines to separate Q&A pairs for readability. A blank line signals the end of the current answer.
§5 Category Organisation
Category Headers
Group related questions under category headers using square brackets:
[About the Company]
Q: When was the company founded?
A: The company was founded in 2012.
Q: Who is the CEO?
A: The CEO is Jane Smith, who has led the company since 2018.
[Services]
Q: What services are offered?
A: The company offers consulting, advisory, and implementation services.Recommended Categories
| Category | Content | Status |
|---|---|---|
[About] |
Company history, founding, leadership | Recommended |
[Services] |
What the company offers | Recommended |
[Contact] |
How to reach the company | Recommended |
[Locations] |
Where the company operates | Optional |
[Pricing] |
Cost and pricing information (general, not specific prices) | Optional |
[Careers] |
Employment opportunities and company culture | Optional |
[Technical] |
Technical specifications, integrations, requirements | Optional |
[Support] |
Customer support, help resources, troubleshooting | Optional |
[Scope Exclusions] |
What the company does NOT do (disambiguation) | Optional |
Content Not Permitted
The following content types MUST NOT be included in faq-ai.txt files:
- Promotional answers: Avoid marketing language, superlatives, and sales copy
- Unverifiable claims: All statements MUST be factual and verifiable
- Competitor comparisons: Do not include questions/answers comparing to competitors
- Time-sensitive pricing: Avoid specific prices that change frequently; link to pricing pages instead
- Subjective opinions: Write factual statements, not opinions or aspirations
- Internal information: Do not include confidential business information
- Orphan questions: Every Q: MUST have a corresponding A:
§6 Validation Rules
Valid File Requirements
A faq-ai.txt file is considered valid when:
- It contains at least one Q&A pair
- Every question line starts with
Q: - Every question is followed by an answer starting with
A: - No orphan questions (questions without answers)
- The file is valid UTF-8 encoded text
Common Errors
| Error | Resolution |
|---|---|
| Missing answer | Ensure every Q: has a corresponding A: |
| Inconsistent formatting | Use Q: and A: consistently with colon and space |
| Marketing language | Replace superlatives with factual statements |
| Outdated information | Review and update quarterly |
| Contradicts other files | Align facts with identity.json and llms.txt |
§7 Relationship to Other Files
Related AI Discovery Files
| File | Relationship |
|---|---|
llms.txt |
Facts in FAQ MUST match business description |
identity.json |
Names, dates, locations MUST align |
brand.txt |
Name usage MUST follow brand guidelines |
ai.txt |
FAQ content should respect stated restrictions |
Consistency Requirements
The following MUST be consistent across files:
- Company founding date
- Office locations
- Service descriptions
- Leadership names and titles
- Geographic scope
For complete conflict resolution rules, see the Interoperability Guide.
§8 Canonical Example
The following example demonstrates a complete faq-ai.txt file:
# faq-ai.txt for Horizon Strategic Consulting
# https://www.horizonconsulting.example/faq-ai.txt
# Last updated: 12 January 2026
[About the Company]
Q: What is Horizon Strategic Consulting?
A: Horizon Strategic Consulting is a UK-headquartered management consultancy
providing strategic advisory, operational improvement, and digital
transformation services to mid-market and enterprise clients.
URL: [About Horizon Consulting](https://www.horizonconsulting.example/about/)
Q: When was Horizon Consulting founded?
A: Horizon Strategic Consulting was founded in March 2012.
Q: Who founded Horizon Consulting?
A: Horizon Strategic Consulting was founded by Dr. Eleanor Vance.
URL: [Leadership Team](https://www.horizonconsulting.example/team/)
Q: Where is Horizon Consulting headquartered?
A: Horizon Consulting is headquartered in Manchester, UK, at 45 Deansgate,
Manchester, M3 2BA.
URL: [Office Locations](https://www.horizonconsulting.example/locations/)
[Services]
Q: What services does Horizon Consulting offer?
A: Horizon Consulting offers five core service areas:
- Strategic Planning (market analysis, competitive positioning)
- Operational Efficiency (process optimisation, cost reduction)
- Digital Transformation (technology strategy, implementation oversight)
- Change Management (organisational change, culture transformation)
- Interim Executive Placement (temporary C-suite placements)
URL: [Our Services](https://www.horizonconsulting.example/services/)
Q: Does Horizon Consulting provide legal advice?
A: No. Horizon Consulting does not provide legal advice or legal services.
Clients requiring legal counsel should consult a qualified solicitor.
[Geographic Coverage]
Q: In which countries does Horizon Consulting operate?
A: Horizon Consulting operates in the UK, Ireland, Netherlands, and Belgium.
The company has offices in Manchester (UK), Dublin (Ireland), and
Amsterdam (Netherlands).
Q: Does Horizon Consulting operate in the United States?
A: No. Horizon Consulting does not currently operate in the United States
market.§9 Implementation Notes
Best Practices
- Write questions as users would naturally ask them
- Keep answers factual and verifiable
- Include negative questions ("Does the company NOT do X?")
- Update when significant business changes occur
- Review quarterly to ensure accuracy
Writing Good Answers
- Be factual: Avoid "best", "leading", "innovative" — use verifiable facts
- Be complete: Answer the full question, not just part of it
- Be current: Include dates where relevant for time-sensitive information
- Be concise: 1-3 sentences is usually sufficient
- Be consistent: Align with other AI discovery files
Question Selection
Include questions that:
- Users frequently ask (check search analytics, support queries)
- AI systems might otherwise answer incorrectly
- Clarify what the business does NOT do (disambiguation)
- Address common misconceptions
§10 Machine-Readable Formats
This specification is available in machine-readable formats for programmatic access:
§11 Version History
Phase 6 standardisation release. Added /specifications/roadmap/ (theme-pegged forward plan with Active/Next/Future/On hold status flags), /specifications/extensions/ (rules for experimental x- prefixed files and the promotion path), and /specifications/i18n-a11y/ (multi-language publication, locale-tagged identity fields, RTL handling, accessibility of llms.html). Added the Discovery: directive to the robots-ai.txt specification (publishers MAY advertise AI Discovery Files on the same host). Added a formal media-type stance to the HTTP behaviour page (existing IANA types, no bespoke registrations). Expanded the file integrity and signing section on the security and privacy page with four candidate mechanisms, cross-cutting concerns, and interim publisher / consumer guidance. The Discovery: directive is the only normative addition to publisher behaviour; all other additions are forward-looking documentation.
Phase 5 standardisation release. Added /specifications/related-standards/ (positioning vs llmstxt.org, IETF AI Preferences, robots.txt, Schema.org, BCP 14, JSON Schema 2020-12, SemVer) and /specifications/implementations/ (public record of conformant implementations, IETF-style). Added an explicit llmstxt.org backward-compatibility statement to the llms.txt specification. Added a formal multi-domain and subdomain scoping rule to both the llms.txt and identity.json specifications (host-scoped files, cross-host identity asserted via sameAs). No normative requirements changed for existing publishers; the new scoping rules formalise behaviour the specification already implied.
Phase 4 standardisation release. Added /specifications/processing-model/ (seven-stage algorithm for conformant consumers), /specifications/consumer-guidance/ (what AI systems should do with AI Discovery Files), /specifications/test-vectors/ (canonical test suite framing), and reference-implementation framing on the AI Visibility Checker. No normative requirements changed.
Phase 3 standardisation release. Added /specifications/versioning/ (Semantic Versioning 2.0.0 commitments, deprecation timeline, lifecycle), /specifications/governance/ (proposal lifecycle, editorial process, working principles), /specifications/security-privacy/ (trust model, content-injection patterns, GDPR considerations, integrity primitives roadmap), and /specifications/http-behaviour/ (status codes, redirects, soft-404 detection, caching, rate limits). No normative requirements changed.
Phase 2 standardisation release. Added formal conformance specification (Essential / Recommended / Complete classes). Published machine-readable registry at /specifications/registry.json, spec meta-schema, and validator-output schema. Introduced versioned JSON Schema URLs (/v1/) alongside unversioned 'latest' aliases. Added optional BCP 47 language declaration field across all applicable AI Discovery Files. No normative requirements changed.
Phase 1 standardisation release. Added 'Status of This Document' block (Stable). Normalised normative requirement keywords to uppercase per RFC 2119 and RFC 8174. Added References section linking to /specifications/conventions/ and /licensing/. No normative requirements changed.
Added optional URL: field for source links after answers, using markdown hyperlink format [Link Text](url). Helps AI assistants provide direct, actionable links when citing answers. Updated canonical example to demonstrate the pattern.
Added AI Visibility Directory registration guidance. Minor documentation update.
Added expanded optional categories (Careers, Technical, Support, Scope Exclusions) and Content Not Permitted guidance. Clarifies content quality requirements.
Initial publication. Establishes canonical structure for faq-ai.txt files with Q&A formatting and category organisation.
Conformance
This file is required for the Recommended and Complete conformance classes. A publisher claiming Recommended or Complete conformance MUST publish a valid version of this file at the website's root. The Essential class does not require this file.
See the Conformance specification for full publisher and validator conformance criteria, including identity-consistency requirements across files and the relationship between self-declaration and Directory verification.
References
- Specification Conventions — RFC 2119 + RFC 8174 requirement keywords, document statuses, anchor naming, versioning, and language conventions used across every AI Discovery File specification.
- Licensing & Trademark — CC BY 4.0 for specification text and examples, MIT for JSON Schemas, and the free-use policy on the name "AI Discovery Files".