Questions answered from the source
Service, reservations and property teams get the answer from your own fare rules, conditions and procedures, with the clause or the step cited.
Travel and hospitality · AI agents inside the company
Travel and hospitality agents. A person decides. Data stays.
Katonic runs AI agents inside your own data centre or private cloud. Agents read your policies, your fare and rate rules and your booking and operations records, prepare the work, and stop for a named person.
No agent confirms, changes or cancels a booking, issues a refund or compensation, sets a rate or a fare, or answers a guest or a passenger unseen. It prepares, and a person decides.
Guest, passenger and booking data stay on infrastructure you control.
Guest, passenger and booking data stay on your infrastructure
A model you serve
Each person's permissions kept
No agent speaks for the company unseen
Enquiries, booking changes, refund requests, compensation claims and commission statements run on reading, checking and drafting. That is where an agent helps most, and where a wrong answer to a guest or a passenger costs most.
The hotel templates we opened send replies and compensation offers to guests with approval only an optional extra.
None of the vendor pages we read offers on-premises or customer-cloud hosting. And running there is not enough: what matters is which tool waits for a person, whose identity the agent uses, and what is recorded.
A company is held to what its agent tells a customer. A tribunal has held an airline to what its chatbot told a passenger.
Tools must be chosen one by one: of 102 public interface descriptions from three publishers the importer read, none was refused; they carry 3,942 operations, 2,286 sent as changes.
Guest and passenger records are personal data, and rates, contracts and commission terms are the company's own. Both have to stay inside the company, and card numbers stay in the payment system: no agent reads them.
What the rules ask
The company stands behind what a guest or a passenger is told. Passengers are told their rights when a flight is delayed or cancelled. A price set for the person says so. The company knows where the data is.
Agents read and draft. People keep every decision about the booking, the refund, the price and the reply.
Questions answered from the source
Service, reservations and property teams get the answer from your own fare rules, conditions and procedures, with the clause or the step cited.
Work prepared for a decision
Booking changes, refund requests, compensation claims, group proposals and commission statements are checked and set out for the person who decides.
Patterns you can act on
Feedback themes, rate differences, turnaround delays and audit exceptions are listed and counted, with the records behind each.
A record of every step
Who asked, which tool ran and what was decided.
From guest and passenger service to reservations, disruption, sales, revenue, hotel and airport operations, loyalty and finance.
Service and bookings
Enquiries, policy questions, booking changes, refund requests
Disruption and claims
Notices, compensation claims, rebooking options, baggage claims
Sales and revenue
Proposals, partner contracts, rate parity, pickup and pace
Operations and finance
Handovers, maintenance requests, commission statements, night audit
Each is a use case an airline, a hotel group or an airport can build in Katonic Studio. A name with a link has its own page.
Every agent follows one rule: it prepares, and a named person decides.
Go deeper
Guides and answers
An airline or a hotel group that wants AI agents on its own infrastructure has more than one kind of product to choose from. The products sold for this fall into four types: agent platforms you run yourself, infrastructure platforms, automation platforms with agent features, and developer frameworks. Our comparison describes fifteen of them with the same fields, read on each vendor’s own pages: where each runs, who operates it, governance, price and limits. For travel, connectors to reservation and operations systems and approval before anything reaches a guest or a passenger matter most.
On-premises AI agent platforms compared: 15 options by type →An agent reaches your documents and systems in one of four ways, all through the same tool gateway and the same approval rules.
Connection · 01
A guest's letter, a group enquiry, a supplier contract, a banquet order, a commission statement.
Connection · 02
Fare rules, conditions of carriage, cancellation policies, brand standards, operating procedures and partner contracts, indexed with each reader's permissions kept.
Connection · 03
↗Give Katonic the description of an interface you already run, and choose which operations become tools. This is the route to reservations, guest profiles, room status, flights and work orders.
Connection · 04
↗Servers from the catalogue, deployed inside your own cluster, for the data warehouse, reporting databases and web search.
System names are left out on purpose. You supply the address and the sign-in. Nothing here is a tested integration with a named product.
It runs on the company's own infrastructure or private cloud, with a model the company serves. Guest, passenger and booking data stay there.
Every tool call passes the company's policy. A policy can allow a tool, record its use, hold it for a person's approval, or block it, for the whole company or for one property or station.
A tool that changes something waits for approval unless the company has decided otherwise for that tool. That is how no booking, refund, reply or rate changes unapproved.
No agent speaks for the company unseen. What reaches a guest, a passenger or a partner has been read and sent by a person.
The agent acts as the person using it. Where the company's API checks the caller, Katonic passes the signed-in user's identity, and the agent cannot claim another.
Every call is recorded, with who asked, which tool, and what was decided, and an agent, a tool or a whole server can be switched off at once.
From visual workflows to full code - pick the path that fits your team and project.
Visual Builder
Follow guided steps to build agents. Configure each stage through the UI. No code required.
Best for
Service, revenue and finance analysts
Conversational Builder
Staff describe how they check a refund request against the fare rules or a commission statement against the bookings. AI generates the agent. Developers refine together.
Best for
Staff + Developers together
Full IDEs
JupyterLab, VS Code, RStudio with GPU access, pre-installed frameworks, and direct deployment.
Best for
Data Scientists & ML Engineers
The rules on passengers, consumers, workers and AI differ by country and point the same way. Each is a matter of how the system is built.
The company stands behind the answer
A company is held to what its agent tells a customer: a tribunal has held an airline to what its chatbot told a passenger. Here an agent drafts what goes to a guest or a passenger, and a person reads and sends it.
Passengers are told their rights
Under EU air passenger rights rules, passengers must be told their rights when a flight is delayed or cancelled. Here an agent drafts the notice from your approved wording, with the rights it must state, and a person checks and sends every notice.
A price set for the person says so
Under EU consumer law, a price personalised by automated decision-making must say so when a consumer books a stay online. Here no agent sets a rate or a fare, and none prices by the person. Revenue agents prepare reports and lists for the revenue manager.
Workers are not scored, and an AI says it is an AI
In the European Union, AI that allocates tasks to workers or evaluates them is high-risk from 2 December 2027, and people must be told when they deal with an AI. Here no agent allocates shifts or scores a colleague, and what reaches a guest is sent by a person.
Katonic AI is certified to ISO 27001. This is a summary of published rules, not legal advice.
In an airline's contact centre, a service agent asks what the fare rules say about a name change, and an agent answers from the rules with the clause cited; an exception goes to a supervisor. In the claims team, an agent sets a compensation claim against the passenger rights rules and the flight record, says which condition it meets or fails, and drafts the decision note for the claims handler, who accepts, rejects or pays. In a hotel group's finance office, an agent sets a travel agency's commission statement against the bookings and the contract rate and lists each difference, and accounts payable approves or disputes it. Each one reads what the company already has, changes nothing by itself, and leaves a record.
Seven questions to ask any supplier before an agent touches a booking or a refund, with our answers.
On infrastructure you control: your data centre or your private cloud. The model is one you serve there.
What the person using it can read. Knowledge sources keep each reader's permissions, and where your API checks the caller, Katonic passes the signed-in user's identity. No agent reads card numbers.
Only what you allow. A tool that changes a booking, a reply or a refund waits for approval unless you have decided otherwise for that tool.
A named person. No agent confirms, changes or cancels a booking, issues a refund or compensation, sets a rate or a fare, or denies boarding.
No. An agent drafts, and a person reads and sends what goes to a guest, a passenger or a partner. When the policies do not answer, the agent says so.
Who asked, which tool ran and what was decided, for every call.
At once. An agent, a tool or a whole server can be switched off by your administrators.
A policy Q&A agent can run on the first day, because no system of record is touched. Talk to our team about building it on your infrastructure.