Configure AI allowance limits and fallback
Before Visitor AI can go live, an admin must confirm what happens when the workspace uses its included verified AI outcomes. Open Settings → Automation → AI limits & fallback.
Choose allowance behavior
- Stop new public AI engagements: the default. New engagements use the saved fallback after the included allowance is reached.
- Continue at €0.49 per verified outcome: enable pay-as-you-go and set a monthly euro cap.
An already-active conversation may finish against reserved capacity. New engagements follow the current allowance policy.
Choose the visitor fallback
- Route to the team or contact form.
- Show the saved unavailable message.
Optional operational guardrails
Request and token limits are separate from verified-outcome billing. Use them as additional safety controls, then test that the Agent stops or hands off as intended when a configured cap is reached.
Was this article helpful?