Alerts
Alerts notify you when a conversation has been flagged for potentially concerning content. This is where you'll spend most of your time in Sensible.
How alerts work
- Your child has a conversation with an AI assistant
- Sensible's backend analyzes the conversation for safety concerns
- If something is flagged, an alert is created with a severity level and reason
- The alert appears on your Alerts page
Alert states
| State | Meaning |
|---|---|
| Unread | New alert you haven't looked at yet |
| Read | You've viewed the alert |
| Dismissed | You've reviewed it and decided no action is needed |
Filtering alerts
You can filter alerts by:
- All — Show every alert
- Unread — Only alerts you haven't viewed
- Severity — Filter by low, medium, high, or critical
What to do with alerts
When you see an alert:
- Read it — Understand what was flagged and why
- Check the conversation — Read the full prompt and response for context
- Decide — Is this something that needs a conversation with your child?
- Dismiss or follow up — Mark it as dismissed if no action is needed, or have a conversation with your child about it
Alert severity levels
See Severity Levels for a detailed breakdown of what each level means and examples.
Not every alert needs action
Low-severity alerts might be perfectly normal curiosity. Use your judgment as a parent — you know your child best.