We’ve improved how the Decision Engine handles situations where a customer explicitly asks to speak with a human agent. This update gives you clearer control and prevents issues seen in some configurations.

What’s new

• You can now decide per block whether a process should be interrupted (“escaped”) when a human agent is requested.

• This setting is available in both Collect Data (formerly UIG) and Option blocks.

• The update does not change your current behavior — it simply gives you the ability to manage it directly in the UI.

This update ensures more predictable behavior and gives you full control over when your AI Agent should hand off to a human.