# Detailed feedback loop for AI Knowledge base

**URL:** <https://community.hubspot.com/t/detailed-feedback-loop-for-ai-knowledge-base/158874>\
**Category:** HubSpot Ideas\
**Tags:** help-desk\
**Created:** [September 14, 2026, 3:15pm UTC](https://community.hubspot.com/t/detailed-feedback-loop-for-ai-knowledge-base/158874 "2026-09-14T15:15:05Z")\
**Posts on this page:** 1\
**Page:** 1

<div class="post-metadata">

**Author:** ![SyN-Thmn](https://avatars.discourse-cdn.com/v4/letter/s/bbe5ce/32.png) [@SyN-Thmn](https://community.hubspot.com/u/SyN-Thmn)\
**Post date:** [September 14, 2026, 3:15pm UTC](https://community.hubspot.com/t/detailed-feedback-loop-for-ai-knowledge-base/158874/1 "2026-09-14T15:15:06Z")

</div>

The Knowledge Base agent has strong potential for turning support history into useful documentation but it has a lot of limitations that lead to inaccurate and not relevant articles.

How can I provide feedback or reason why the articles is rejected, wrong, confusing, or low quality.´? How can AI learn our preferences or change subsequent suggestions other than providing a basic signal with the Dismiss or Thumbs down / Not helpful / “How did Breeze do?  
A detailed feedback loop so the Knowledge Base agent can learn why a suggestion was rejected.

I would our agents to be able to provide a detailed feedback loop so the Knowledge Base agent can learn why a suggestion was rejected.