Twelve questions to ask before you hire an AI employee (and our straight answers)

Hosted by Mr. NetShow.

You would not hire a person without asking hard questions. Do the same with an AI employee. Below are the twelve we would ask, with our own answers. Each one is true today, and each links to the page that shows it. Where something is still being built, we say so.

The twelve questions

1. What can it do without asking me?

It greets each visitor by voice first, with a face, and says it is an AI. It answers from your own documents, pages and questions you approve, and it books the next step: a call, a visit or a callback request (the Front Desk). That is talking, answering and booking. Anything more waits for you.

2. What may it never do?

It never claims to be a person (it says it is an AI). It never spends money or sends anything in your name until a person approves it (approvals). You also tell it what needs approval and what should never happen (your rules).

3. Where does my data live, and who sees it?

We process prompts, files and generated outputs to answer and to keep context, and we do not sell your data. Our privacy page says what we collect and how it is handled. On the Enterprise AI Workforce package, your AI employees run on your own server or cloud (Enterprise).

4. Which AI models does it use?

Leading families, chosen per task by the Harness router: Claude, GPT, Gemini, Llama, Mistral and Qwen (the router). We name the family, never a version number, because versions change every few months.

5. What does it cost?

Each package has a starting price, and your exact price is set in writing after the build call (the ladder):

• AI Front Desk: from $699 a month, with setup quoted.
• AI Specialist: from $2,999 a month, with setup quoted.
• AI Team: from $4,999 a month, with setup quoted.
• Enterprise AI Workforce: from $9,999 a month, with setup quoted.
• Custom: priced after the build call, in writing.

Nothing is charged online (no online charge).

6. Does the face cost me by the minute?

No. The face is drawn by your visitor's own browser, not streamed from a server, so an hour on screen costs what a minute costs: $0 a minute (why it is $0).

7. How does it speak?

Voice first. The NetShow voice is included. GPT Live is the premium voice, with a daily cap. Your brand never speaks in a robotic browser voice; the browser's built-in voice is only a labelled backup voice. The microphone opens only when a visitor presses Talk (the voices).

8. What happens when it is wrong?

Anything going out waits for approval, so a wrong draft stops with you (approvals). Every run leaves a receipt, so you can see what happened (receipts). And each month we look at what it did with you and improve it (after you book).

9. Who approves posts and emails?

You do. On the packages that write for you, every post, email or code change waits for your approval (approvals).

10. How do I see what it did?

You see calls, transcripts and visitor questions in one place. That is live today (what you see). Each run's receipt shows what was asked, which model and tool did it, and what it cost (run history). A weekly summary is in build.

11. How do I stop it?

Say no. Nothing that spends or goes out in your name runs until a person approves it (approvals). Every priced lane stops for the day at its daily ceiling, and a closed lane stays closed (ceilings). Want it stopped for good? Tell us: a person replies within one business day (reach us).

12. What are the contract terms?

They are named on the build call. We have no published terms to quote here, so we will not invent any. Before anything is built, you get in writing the role, what the employee will and will not do, the price and the setup plan (what you get in writing).

FAQ

Is an AI employee a chatbot?

No. It greets visitors by voice, with a face, and it can book the next step (what it does).

Do I need to be technical?

No. Our team sets it up with you as part of the build (done for you).

What if my question is not on this list?

Bring it to the build call. We answer it there, and in writing.

Ask us all twelve

A good hire starts with good questions. We would rather you ask them before anything is built than after. On the build call, bring this list and your own. We will tell you what your AI employee can do today, what is still in build, what it will never do, where your data goes and what it will cost. Then you get it in writing before any work starts. Tell us the job in your own words; there is nothing to prepare.