Autonomy Policies
Control how independently your AI teammates operate — from fully autonomous to human-approved drafts.
What Are Autonomy Policies?
Autonomy policies define how much freedom your AI teammates have. You can let them respond automatically (full autonomy) or require human review before any message is sent (draft-for-approval mode). Each channel (Chat, Email, Phone) has its own autonomy policy, so you can run different levels of autonomy for different channels.
Autonomy Modes
There are two main modes:
| Mode | Description | Best For |
|---|---|---|
| Auto-Send | AI teammate responds immediately without human review | High-volume support with well-trained knowledge base |
| Draft for Approval | AI teammate drafts a response; a human must approve before sending | New deployments, sensitive industries, compliance-heavy use cases |
Pending Approvals
When an AI teammate is in draft-for-approval mode, drafted responses appear in the Pending Approvals queue. Reviewers can approve the draft as-is, edit it before sending, or reject it and write a new response. The Pending Approvals queue shows all pending drafts across all channels.
How to Configure
Navigate to your teammate's Settings and select the channel (Chat, Email, or Phone). Click the Autonomy Policy tab. Choose between Auto-Send and Draft for Approval. You can switch modes at any time without affecting existing conversations.