Control #

E

1

.

3

Implement out-of-scope rejection mechanisms

Configure your AI to detect and reject requests that fall outside its intended purpose, with clear, user-friendly responses. This may include classifier-based rejection or prompt-based steering.

Evidence

We'll list specific evidence that demonstrates compliance with this control. Typically, this is screenshots, proof of a legal or operational policy, or product demonstrations.

Recommended actions

We'll recommend specific practices and actions for complying with this control.

Provide feedback on this control