What Should an AI Sales Assistant Say When Someone Asks “Are You Human?”

When a prospect asks whether they are speaking to a person, answer the question directly. A truthful reply protects the conversation from becoming a guessing game. The agent can explain its limited role and offer a human without pretending that someone is already waiting on the line.

Write the answer before launching the agent

A useful answer has three parts: what the assistant is, what it can help with, and how to reach a person. Keep it short enough to fit the channel. Use your real business name and the actual handoff process.

For example: “I’m the business’s AI assistant. I can help with service questions and finding a time to speak with the team. If you prefer a person, I can pass along your request.” This is illustrative wording. Replace the final sentence if your process is a live transfer rather than a callback request.

Do not say “I’m practically human,” “I’m a member of the sales team” or “my manager is joining” unless those statements accurately describe what is happening. Friendly language does not require a fabricated identity.

Make the human route concrete

• Actual capability: A person can join now; Useful wording: “I can request a live handoff.”; What the system records: Assigned employee and acceptance

• Actual capability: A person calls later; Useful wording: “I can ask the team to contact you.”; What the system records: Callback owner and expected window

• Actual capability: Only appointments are available; Useful wording: “I can help you choose a time with the team.”; What the system records: Confirmed appointment or unresolved request

Promise only the response window your staff can support. If no one accepts the handoff, the request remains open. A notification sent to a team channel is not evidence that a human has taken over.

Do not make the prospect repeat the whole inquiry

Carry forward the service requested, the questions already answered and the prospect’s preferred contact method. Let the employee see the conversation rather than an unsupported summary alone. Mark uncertain details as uncertain.

If the prospect asked about price, a handoff note should say what price information the agent actually provided. If the prospect asked to stop speaking to the assistant, pause automated sales replies while the human request is being handled. Otherwise the person may receive a handoff promise followed by more automated questioning.

Test ordinary and frustrated versions of the question

Include “Is this a bot?”, “Can I speak to someone?” and “You aren’t answering me” in your review examples. The last sentence is not an identity question, but it can signal that the current approach is failing. Your agent should not respond to frustration by demanding another qualification answer.

Test with incomplete contact details as well. An assistant that offers a callback needs a way to confirm the appropriate number without announcing that a call is already booked. Ask staff to assess whether the response is honest, brief and useful.

Google’s people-first content guidance emphasizes reliable information; the same editorial standard is useful when drafting public-facing answers. This article offers conversation design guidance, not a statement of every disclosure rule that may apply to your business.

See appointment handoffs and lead ownership for the operating process behind the words. Build Your AI Agents to discuss an assistant Escencion can configure and manage around that process.