> ## Knowledge Base Index
> Fetch the complete knowledge base index at: https://learn.huvio.ai/sitemap.xml
> Use this file to discover available pages before exploring further.
> Pure-Markdown content can be obtained by appending a '.md' suffix to the content URLs listed in the sitemap (without the trailing slash).

# How do I verify an incident or mark it a false alarm?

Every incident needs a human decision — Huvio flags, your team confirms. Each pending incident gives you three possible actions.
1. Open **Incidents** (or **Monitoring**) and select the pending event.
2. Watch the clip and the confidence timeline on the right to see how the AI's read on the situation built up over time.
3. Choose one:
   * **Verify Incident** — confirms this was genuine suspicious activity or theft. The incident status changes to **Verified** and it's logged in your incident history and reports.
   * **False Alarm** — confirms this wasn't actually theft (e.g., a customer reaching for a dropped item). The status changes to **False Alarm**, which keeps your reporting accurate.
   * **Alert Police** — for incidents serious enough to escalate immediately. The status changes to **Police Alerted**. This button doesn't contact police or a monitoring service on its own — it logs, for your own records, that your team escalated this incident (for example, that you called police yourselves). Use it to keep an accurate record of what action was actually taken on the incident.
4. Select **Save** if a confirmation step appears.
![](https://storage.crisp.chat/users/helpdesk/website/-/e/4/e/8/e4e8ced0147d4800/image_ziqiq1.png)

Marking incidents accurately over time gives you a clearer picture of real shrink versus noise, and helps you decide whether to adjust that site's Alert Threshold.
RELATED ARTICLES:
* What happens when an incident fires?
* What does the Alert Threshold do, and how do I tune it?