- Microsoft and Cloudflare want websites to behave as conversational applications
- The content of the Cloudflare Author Indices automatically, promising updates without problems for the website
- NLWEB presents structured MCP final points for AI agents seeking reliable access
Microsoft and Cloudflare have announced a new collaboration that seeks to facilitate websites for both people and automated systems.
The initiative merges the Microsoft “NLWEB” standard with the “Authorag” infrastructure of Cloudflare, providing a model for conversational search.
Instead of navigation based on keywords, the system allows the interaction of natural language, presenting direct responses instead of link lists.
From search engines to answer engines
The objective of the new launch is to make any site work as an AI application, where human visitors and AI agents can ask questions and receive structured answers.
“Together, NLWEB and Author allow editors to go beyond the search tables, making conversational interfaces for websites easy to create and implement,” said RV Guha, creator of NLWEB, CVP and technical member in Microsoft. “This integration will allow each website to be easily prepared for people and for trust agents.”
The traditional search is based on keywords, leaving users to examine multiple links to achieve relevant information, but the defenders of this new approach argue that the model no longer coincides with the expectations formed by tools such as Chatgpt, Copilot and Claude.
These systems provide immediate responses, and people expect more and more than websites do the same.
However, although that frame is convincing, it assumes that all users prefer conversational responses about navigation.
The association also emphasizes the role of AI agents as a new class of “visitors.”
These automated systems generally scrape the pages or follow keyword consultations, but NLWEB presents a structured “MCP” end point that gives them controlled access to the site data.
This could reduce the inefficiency of scraping while allowing website owners to define the terms of access.
However, questions remain about whether the site operators will see tangible benefits or mainly additional technical loads.
The Cloudflare Author component handles the tracking, indexation and integration of the site content into a database of administered vectors.
The system promises to maintain fresh information with continuous updates and provide observability through the cloudflare ai gateway.
While this sounds efficient, indeed indexing and access pipes in the hands of a third -party infrastructure provider.
That raises concerns about the cost, dependence and if the owners of the site will yield too much control over how their data is handled.
By frameting websites such as first -class data sources for AI, Microsoft and Cloudflare tools are being positioned in direct competition with traditional search engines, particularly Google.
If AI Writer Models and LLM systems are increasingly relying on structured access instead of scraping, search patterns could change significantly.
To enable the search for conversation on a website through Authorag in Cloudflare, users log in to the Cloudflare board, create a new instance of Authorag using the quick implementation option of the NLWEB website, select the desired domain and begin to indexation.
Once the process is completed, the content of the site can be searched through natural language consultations, and the owners can obtain a preview or embed the conversation interface to prove how visitors will appear.