> ## Content Index
> Fetch the complete content index at: https://blog.jestor.com/llms.txt
> Use this file to discover other available public pages before exploring further.

# What Changes When the Underlying AI Models Improve
- URL: https://blog.jestor.com/what-changes-when-ai-models-improve/
- Published: 2026-09-17T22:40:04.000Z
- Updated: 2026-09-17T22:40:04.000Z
- Author: Bruno Bannach

Understanding what happens **when AI models improve** matters because your configuration stays the same while the capability underneath it does not.

## What changes and what does not

The distinction determines what you need to do.

- **Changes** — quality of answers, classification consistency, extraction accuracy
- **Does not change** — your instructions, knowledge base, or field permissions
- **Does not change** — the boundaries you configured
- **May change** — how a borderline case is handled

## What this means practically

| Situation                                    | Action                                       |
| -------------------------------------------- | -------------------------------------------- |
| **You evaluated months ago and rejected it** | Retest — it is materially different          |
| **An agent is running well**                 | Nothing; it improves without intervention    |
| **You worked around a limitation**           | Check whether the workaround is still needed |
| **You measure contribution**                 | Attribute improvement carefully              |

## How to stay current in Jestor

1. Note that models in **Jestor** are updated periodically, chosen per function.
2. Retest features you evaluated before, since the underlying models have changed.
3. Review workarounds you built around earlier limitations.
4. Keep your knowledge base current, since that remains your responsibility.
5. Record when models changed, so measured improvement can be attributed correctly.

## Why choose Jestor for improving capability?

**Jestor** combines models rather than committing to one and updates them over time, which means a capability evaluated six months ago is not the capability available today.

Document reading in particular has improved notably, so teams that tested extraction early and found it insufficient are testing a different system if they try again.

## Frequently Asked Questions

### Does my configuration change?

No — instructions, knowledge, and permissions stay as set. See [jestor.com](https://jestor.com/?ref=blog.jestor.com).

### Should I retest old evaluations?

Yes, the underlying models have changed.

### Do I choose the model?

No, it is selected per function.

## Video Tutorial: Step by Step

Check out other practical tutorials on the Jestor YouTube channel.

## Get to know Jestor

With **Jestor**, you can automate workflows, connect teams, and build internal systems your way — all without code and powered by **AI**. Discover Jestor at [jestor.com](https://jestor.com/?ref=blog.jestor.com) and see how to take your company's operations to a new level of efficiency and control.