AI BriefingNew AI Traffic Options for Cloudflare's Customers
Key point
Cloudflare has introduced a new management framework that classifies and controls traffic based on the purpose of AI bots.
Details
Cloudflare has announced a new classification system that goes beyond simple 'AI bot blocking,' allowing traffic to be managed in a more granular way based on how bots behave and what they are for.
The new classification system is based on three core Use Cases:
- Search: Behavior that indexes content to build a database for answering questions later. Website owners can expect referral traffic or fair compensation from this process.
- Agent: Automated behavior that performs tasks in real time on behalf of a user (e.g., ChatGPT-User, Gemini/Claude's browser-use agents). It visits web applications to complete a user's request.
- Training: Crawling behavior that permanently absorbs data into a model's architecture in order to improve the model's performance.
Cloudflare strongly recommends that bot operators separate their crawlers by purpose into Search, Agent, and Training, and plans to use this to help website owners exercise more fine-grained control over access permissions for each purpose.
This summary was generated automatically by AI. Check the original for the author's claims and context. Copyright belongs to the original author.
Our guide explains how the AI works. Report summary errors, attribution issues, or removal requests via Contact.