Allin AI is the AI assistant in alliants.ai. It replies to guests on web chat, text message, WhatsApp, email and social messaging apps, drawing on their booking and your own knowledge base. For your front office, guest services, reservations and concierge teams, it suggests replies, summarises conversations and helps with bookings, requests and guest profiles.

No. Allin handles routine questions within the limits you set and hands anything it should not handle to your team, with a summary. Staff can take over any conversation in one click and hand it back to Allin when they are done.

Web chat, text message, WhatsApp, email and social messaging apps such as Messenger and Instagram. Replies are formatted for each channel, kept short on text, and written in the guest's own language. On WhatsApp, Allin can offer buttons and pick lists, and on text message numbered options, so guests can answer with a tap or a number.

Yes. When Allin replies on its own, guests are told they are talking to AI, in line with Article 50 of the EU AI Act. Allin also keeps to your brand identity and cannot be talked into posing as another business.

Yes, once they are verified. General questions are answered straight away, but booking details are shared only after the guest confirms their booking reference and last name, or a one-time code. Allin can then look up the booking, past stays and the cancellation policy, and amend or cancel within your policies after the guest confirms.

Yes, where your brand switches it on. Verified guests can check room availability and rates and book a new stay in the conversation. Allin can also book restaurant tables and spa treatments, pass experience requests to your team, and answer questions on nearby places, weather and travel time. Each of these can be withheld for your brand.

You decide, channel by channel. For each messaging channel, brand-wide or per property, AI can be off, suggesting replies for staff to approve, or replying on its own. Replying on its own is switched on by Alliants for your brand. You also set the topics and situations that always go to a person.

Allin drafts a reply to each guest message, which staff send in one click or edit first. Staff can write in their own language and the reply is translated for the guest. Each conversation shows an AI summary, and a staff assistant finds bookings, guests and requests, drafts requests and suggests who should take them, and prepares a daily morning report.

It hands over. Anything Allin cannot finish becomes an open item in the conversation for staff to answer or close, and conversations it should not handle go to the right department with an AI summary. A strongly negative message opens a complaint item for your team.

No. Any action that changes data needs an explicit confirmation from the guest or a colleague before it runs, and higher-risk changes need staff approval. Staff see a confirmation card to approve or reject each proposed change, and every change Allin makes is recorded as an AI action.

Yes. Staff can switch Allin off on a guest's profile, so it stays out of every conversation with that guest and your team handles them directly. Availability depends on your property; ask your Alliants contact.

From what you tell it. Your team adds web pages, files and typed answers in one place, follows a checklist of what guests ask most, and can test a question to see what Allin would find. Answers can be set brand-wide or per property, and a property's own answer replaces the brand's for that hotel.

Your property management system (PMS), for booking lookups, new bookings, amendments and cancellations. SevenRooms for restaurant bookings, and Book4Time or Zenoti for spa treatments. Messages travel through Twilio and Meta, and a brand can connect its own knowledge base, which Allin checks first.

Allin uses OpenAI models by default, with Anthropic's Claude as a fallback. Every request goes through a dedicated Alliants AI service that holds model access, usage tracking and knowledge search, rather than going from each screen straight to a model provider.

We switch Allin AI on for your brand. Your admins then choose the AI mode for each channel and set up the brand persona with a guided set-up: the assistant's name, tone, formality, emoji use, sign-off and the topics it always hands over. You can copy one property's answers to set up the next.

是的。Alliants 已通过 ISO 27001 认证。安全措施已深度融入平台，涵盖从员工登录到访客数据存储及删除的各个环节。

On Amazon Web Services, in one regional deployment for your brand: Europe, the United States or Asia Pacific. Each brand has its own dedicated database, including its Allin knowledge base, and data is encrypted in transit.

Every reply is checked before it goes out for contact details, booking references and links that do not belong in it, and replies Allin sends on its own are checked against their sources. Guest messages and booking fields cannot redirect the AI, usage limits stop runaway exchanges, and changes are tested against reference and hostile conversations before release.

When a guest asks to be erased, the request clears their data across related records, including conversations and the AI's own records of each step, and keeps a record of what was cleared without holding any guest details. Consent is held per channel and checked before any message is sent.

Allin AI is part of the Signature, Messaging and All-In bundles. Pricing is agreed per contract, so talk to our team about the right bundle for your properties.

Talk to our team. We will show you Allin AI running in a hotel like yours and talk through how it would fit your properties and systems.