AI Transparency & Human Oversight
Automation supports your team. Your business remains responsible for its customer communications.
1. How replies are produced
The free FAQ engine matches a customer question to stored business knowledge. It is deterministic matching, not a generative model. Optional Gemini generates replies from selected knowledge, product information and recent chat context. Provider limits, model availability and data-use terms can change.
2. Accuracy and boundaries
Prices and store details are based on the last successful sync. AI can misunderstand intent or produce incorrect information. This version does not autonomously charge customers, issue refunds, place orders or confirm appointments. Add accurate knowledge, test expected questions and review important outputs.
3. Human support
A workspace owner can pause automation for an individual conversation. Recognized requests for a human and some failed AI operations place the chat under human control. Your team must monitor the inbox and handle messages the automation cannot safely answer. Do not use the service as the sole basis for medical, legal, employment, credit or other high-impact decisions.
4. Provider choice and data
FAQ mode avoids an external AI call. Gemini requires an API key and acknowledgement of its data terms. A free provider tier may not suit confidential corporate conversations. Select a suitable paid or contractually protected service before processing such information.