# Kill Switch

> A kill switch is the ability to pause or stop any AI agent instantly, from one place, regardless of which vendor or platform the agent runs on. It is the last-resort control that makes every other AI risk acceptable: whatever goes wrong, it can be stopped now.

- Part of: [The Sentry AI Transformation Glossary](https://sentrysolutions.ai/glossary)
- Canonical: https://sentrysolutions.ai/glossary/ai-agent-kill-switch

The hard part is not the button, it is the reach. Stopping an agent you built is easy. Stopping the embedded agent inside a SaaS product, or the one a departing employee wired up, requires that the agent was registered and connected to a control plane in the first place.

This is why the kill switch is an architecture rather than a feature: register everything, connect everything at whatever depth each agent allows, and the stop control follows. Every agent is killable, or the estate is not governed.

## Related terms

- [Human-Gated Approvals](https://sentrysolutions.ai/glossary/human-gated-approvals)
- [Agent Register](https://sentrysolutions.ai/glossary/agent-register)
- [Control Plane](https://sentrysolutions.ai/glossary/control-plane)