Playbook for content publishers

Your main question is usually: "How do I allow useful discovery without exposing the site naively?"

Read first

• AI Usage Policy
• Glossary
• Source precedence
• Patterns library

Your risk model

You care most about:

• AI answer-generation usage;
• model-training posture;
• archive replay;
• scraper pressure;
• editorial reuse without clear boundaries.

Do

• separate answer-generation from training in your reasoning;
• decide what kind of machine access is acceptable;
• choose a preset that reflects editorial goals, not just technical habit;
• document the difference between openness and naivety.

Don’t

• collapse all bots into one category;
• assume robots.txt alone solves hostile scraping;
• use vague language like "block AI" when the real policy is more granular.