Ai Preferences

What an API provider says about AI: whether its content and API output may be used to train AI or generative AI, used for AI inference or search, and whether an AI agent may sign up, act or buy on a person's behalf, plus the machine signals it publishes (an AIPREF header, Content Signals, ai.txt, llms.txt).

Also known as: AI Policy, AI Terms, AIPREF, Content Signals

Example

ai-preferences.yaml 
- name: AI Preferences
  type: AIPreferences
  description: The ai preferences of the Example Weather API, harvested from its own page.
  url: ai-preferences/example-weather-api-ai-preferences.yml
  source_date: '2026-11-09'
  data:
    ai_preferences: 0.1.0
    provider: Example Weather API
    aid: example-weather-api
    url: https://weather.example.com/legal/api-terms
    source_date: '2026-11-09'
    x-apis-io-derivation:
      synthetic: false
      method: harvested
      read: '2026-11-09'
      tool: engine/business/harvest-business.py
      model: claude-haiku-4-5
    train_ai:
      stance: prohibited
      text: You may not use API output to train machine learning or artificial intelligence models.
    train_genai:
      stance: prohibited
      text: You may not use API output to train machine learning or artificial intelligence models.
    ai_use:
      stance: allowed
      text: You may use API output as input to AI assistants that answer end-user questions.
    agent_onboarding:
      stance: unstated
    agent_actions:
      stance: conditional
      text: Automated agents may call the API only with a key issued to the account holder.
    agent_purchases:
      stance: unstated
    signals:
      aipref_header: false
      content_signals: true
      ai_txt_url: null
      llms_txt_url: https://weather.example.com/llms.txt

Standards

Tools

Related Properties

Tags


All Common Properties