Every missed call is a booking that goes to a competitor who picked up.

Chat today. Voice needs a telephony gateway that is not chosen or built yet.

The AI Receptionist answers web chat: bookings, FAQs and routing, with calendar booking that prevents double-booking and a handoff to a person for anything outside its script. It has run only against simulated callers and a simulated calendar, never against a live phone line, voice gateway or calendar.

Phone calls are not built: there is no telephony account and no speech engine. Until there are, the Receptionist does not pick up a phone. A voice, when there is one, will be plainly synthetic, will say it is an AI and will never be a copy of a real person's voice.

The Outreach Engine is text only. Nothing here makes cold calls.

Yes. The lead, reception, outreach and support systems each hand over to a person: the Lead Response Engine the moment a lead asks something it cannot answer, the AI Receptionist for anything outside its script, the Outreach Engine the moment a lead replies, and the E-commerce Support Assistant the instant something is not a confident, rule-based answer.

It uses the official WhatsApp Business API, and only that. A reply inside the 24-hour window after a customer writes needs no template. A message outside the window needs an approved template and an opted-in contact. Messaging strangers and cold sends are not done, because WhatsApp's policy allows contact only with people who opted in. Broadcasts go only to opted-in contacts.

The API is reached through a provider. Whether your current number can be used, and what your setup needs, a person will check with you before you sign.

It depends on the product, and the build names what it reads:

• AI Receptionist and Lead Response Engine: your knowledge base and business hours.
• Dormant List Revival: your customer list, with consent records for who may be contacted.
• E-commerce Support Assistant: your order data, your return policy and your product data.
• Connected Operations Layer: you connect each account through its own sign-in, and can revoke it.

The exact checklist for your system is agreed when scope is signed. The full list is not on file, so a person will give it to you.

Yes, the systems are AI, and they say so. The first message a system sends on its own, on chat, email or WhatsApp, states that it is an AI. That is built into the sending step. There is no setting to switch it off.

Voice is not built yet. A voice, when there is one, will be plainly synthetic, will say it is an AI and will never be a copy of a real person's voice.

This answer box is not AI. It searches a fixed list of written answers.

We cannot promise it never gets something wrong. No system can, and we will not say so.

What it is built to do instead:

• Answer from the data and policy you give it. The support and reception systems say so when a question is not covered, and pass it to a person.
• Stop and tell you on a complaint, a legal threat, a request for a human, a STOP, or a data-erasure request. It sends at most one neutral acknowledgment.
• Say it is an AI on the first message, so nobody is misled about who they are talking to.
• Keep money questions inside limits you set. Above them, it sends a holding line and asks you.
• Write every action to a tamper-evident log you can check afterwards.