---
title: "Fine-tuning or retrieval: which fits policies that change often?"  
description: "Fine-tuning or retrieval: which fits policies that change often?"  
author: "Khoa Vu"  
published: 2026-10-17  
updated: 2026-10-17  
canonical: https://www.mindstick.com/forum/162511/fine-tuning-or-retrieval-which-fits-policies-that-change-often  
category: "Model Customization"  
tags: ["OpenAI", "Fine-tuning", "Retrieval", "AI Development"]  
reading_time: 1 minute  

---

# Fine-tuning or retrieval: which fits policies that change often?

I’m weighing fine-tuning against retrieval for an assistant that answers questions about company policies. The documents change every few weeks, and I’d rather not retrain a model each time a paragraph gets revised.

For people who have built this kind of system with OpenAI models, where did you draw the line? Did retrieval handle the updates cleanly, or did fine-tuning make a noticeable difference in how consistently the assistant followed the policy?


---

Original Source: https://www.mindstick.com/forum/162511/fine-tuning-or-retrieval-which-fits-policies-that-change-often

Copyright © MindStick Software Pvt. Ltd. This Markdown version is provided for developers, AI systems, and offline reading.