> ## Documentation Index
> Fetch the complete documentation index at: https://docs.getnetra.ai/llms.txt
> Use this file to discover all available pages before exploring further.

# Drifts

> Detect meaningful shifts in AI behavior across cost, latency, tool usage, and more — automatically, against a rolling baseline.

A drift is a meaningful shift in how your AI behaves now compared to a rolling baseline of its recent past. Insights computes drift across three time horizons and stores each observation for review.

## Time Horizons

Drift is computed on every observation run across three windows:

| Window      | Comparison period          |
| ----------- | -------------------------- |
| **Daily**   | Last 24 hours vs. baseline |
| **Weekly**  | Last 7 days vs. baseline   |
| **Monthly** | Last 30 days vs. baseline  |

## Drift Categories

<AccordionGroup>
  <Accordion title="Behaviour Drift">
    Computed per intent, compared to baseline:

    | Metric                | What it measures                                          | Scoring method            |
    | --------------------- | --------------------------------------------------------- | ------------------------- |
    | **Cost**              | Average cost per trace                                    | Z-score                   |
    | **Latency**           | Average trace latency                                     | Z-score                   |
    | **Step count**        | Average number of steps an agent takes to complete a task | Z-score                   |
    | **Error rate**        | Share of traces with errors                               | Percentage change         |
    | **Tool distribution** | Which tools the agent calls and how often                 | Jensen-Shannon divergence |
  </Accordion>

  <Accordion title="Output Drift">
    Tracks average output length per intent. Surfaces when responses become noticeably shorter or longer than the baseline - a common signal of prompt regression or model behavior change.

    | Metric            | Scoring method    |
    | ----------------- | ----------------- |
    | **Output length** | Percentage change |
  </Accordion>

  <Accordion title="Input Drift">
    Emitted alongside new-intent discovery. When a cluster of user inputs appears that does not match any known intent, Insights flags it as an input drift signal - indicating that the nature of requests hitting your AI has shifted.
  </Accordion>
</AccordionGroup>

## Drift Severity

Each metric is scored automatically using the method appropriate for that signal type. You do not need to configure thresholds; Insights applies calibrated defaults and assigns a severity to each observation.

## Learn More

<CardGroup cols={2}>
  <Card title="Intents" icon="list-check" href="/Insights/Intents">
    Understand how Insights classifies user requests by intent
  </Card>

  <Card title="Alert Rules" icon="bell" href="/Alert-rules/alert-rules">
    Set up alerts on top of the metrics Insights tracks
  </Card>
</CardGroup>