AI use cases in travel and hospitality: 12 systems to build in 2026

The AI use cases travel and hospitality companies build first are demand forecasting for pricing, multilingual guest messaging, disruption rebooking with automatic refund handling, room and content mapping across suppliers, and chargeback defense. Most run on data already sitting in the PMS, the booking engine, the GDS or the passenger service system, and the rules that matter most are about how prices and refunds are presented to travelers.

The twelve systems below span hotels and resorts, airlines, online travel agencies and travel management companies, organized into revenue and distribution, guest and traveler service, and operations and payments. For each we note the feeds it needs, what it is wired to, what to watch out for and an illustrative crew and duration. Ryz serves enterprises across industries, including travel and hospitality. Our AI pod teams build in your cloud and repos, working on your team with your revenue, digital and operations staff.

About crews and budgets. Each crew and duration is an illustrative planning assumption rather than a quote. We calculate costs from Ryz's public rates only: mid-level $7,000 to $10,000, senior $10,000 to $15,000 and tech lead $15,000 and up per engineer per month, quoted per team, with the lead held at $15,000.

Consumer rules that shape travel AI

• Airline refunds. The Department of Transportation's refund rule, codified in 14 CFR Part 260 and published in April 2024 (Federal Register notice), requires automatic refunds when a carrier cancels or significantly changes a flight and the passenger does not accept alternatives. A significant change includes departure or arrival moving by more than three hours domestically or six hours internationally. Refunds are due within seven business days for credit card purchases and 20 calendar days for other payments.
• Total price display. The FTC's Rule on Unfair or Deceptive Fees (16 CFR Part 464), in effect since May 12, 2025, requires short-term lodging businesses and the platforms that sell them to show the total price, including mandatory fees, whenever a price is advertised.
• EU passenger rights. Regulation (EC) No 261/2004 sets compensation, care and rerouting duties for flights covered in the EU, which any rebooking or claims agent must apply correctly.
• Payments and privacy. Card data falls under PCI DSS, and guest profiles under GDPR and state privacy laws. Keep card numbers out of prompts, transcripts and logs entirely.

Revenue and distribution

1. Demand forecasting for hotel pricing

Models forecast demand by date, room type and segment from booking pace, cancellations, events and competitor rates, and recommend rate and restriction changes that revenue managers accept or adjust.

• Feeds: reservation history and pace, cancellation and no-show data, competitor rate shops, local event calendars.
• Wired to: PMS (Oracle OPERA Cloud, Mews, Cloudbeds), revenue management systems (IDeaS, Duetto), channel manager (SiteMinder).
• Watch out for: one-off events that distort history, and revenue managers who stop trusting a model that cannot explain itself. Many groups build on top of an existing RMS rather than replacing it. See machine learning development.
• Illustrative crew and duration: tech lead plus two senior engineers and one mid-level engineer, five to six months: $42,000 to $55,000 a month, $210,000 to $330,000 in total.

2. Total-price and fee display checks

The system crawls the brand's own booking flows and partner listings, reads every displayed price, and flags any place where resort, destination or cleaning fees are missing from the first price shown, so the issue is fixed before a regulator or class action finds it.

• Feeds: rendered booking pages, rate and fee configuration, partner listing data.
• Wired to: booking engine, CMS, channel manager, ticketing for fixes.
• Watch out for: prices that change by session, currency and device, so the crawler must cover all of them. The fee rule applies to lodging; airfare display falls under separate DOT rules.
• Illustrative crew and duration: tech lead plus one senior engineer, two to three months: $25,000 to $30,000 a month, $50,000 to $90,000 in total.

3. Ancillary and upgrade offers

Models pick which offer to show each traveler and when: a room upgrade at check-in, a seat or bag bundle, late checkout, a spa slot. Offers only show inventory that actually exists, at prices set by the revenue team.

• Feeds: booking and stay history, inventory, offer acceptance data, loyalty profiles.
• Wired to: PMS or passenger service system, merchandising engine, mobile app and email platforms, airline NDC APIs.
• Watch out for: overselling upgrades the property cannot deliver, and personalization that relies on data guests did not consent to share.
• Illustrative crew and duration: tech lead plus two senior engineers, four to five months: $35,000 to $45,000 a month, $140,000 to $225,000 in total.

4. Group and event RFP responses

Hotel sales teams receive hundreds of group and meeting RFPs through Cvent and email. The system extracts dates, room blocks, meeting space and catering needs, checks availability and displacement against forecast transient demand, and drafts the proposal for the sales manager.

• Feeds: RFPs, space and room inventory, past group pickup and wash, transient forecasts.
• Wired to: sales and catering systems (Delphi, Salesforce), Cvent, PMS.
• Watch out for: displacement math, which should come from the revenue system, not the language model.
• Illustrative crew and duration: tech lead plus two senior engineers, three to four months: $35,000 to $45,000 a month, $105,000 to $180,000 in total.

Guest and traveler service

5. Multilingual guest messaging and reservations agent

An agent answers guest messages by text, WhatsApp, chat and phone in the guest's language: modify a booking, request early check-in, ask about parking, report a problem in the room. It writes to the PMS and opens service tickets for staff. A Ryz AI pod built an AI driver-support agent that covers about 218,000 driver calls a year in three languages, with roughly a 95% reduction in cost per call; multilingual support follows the same pattern.

• Feeds: property knowledge, reservation data, message and call history, service request outcomes.
• Wired to: PMS, guest messaging platforms, telephony, housekeeping and maintenance apps.
• Watch out for: property-level details that differ across a portfolio, and complaints that need a manager quickly. See AI voice agent development.
• Illustrative crew and duration: tech lead plus four senior engineers, five to seven months: $55,000 to $75,000 a month, $275,000 to $525,000 in total.

6. Disruption rebooking and refund handling

When weather or operations cancel flights, the system offers each affected passenger rebooking options by their priorities, processes accepted changes, and issues refunds automatically where the DOT rule requires them, with EU261 compensation checks on covered flights.

• Feeds: schedule changes, inventory, passenger records, fare rules, the carrier's refund and compensation policies.
• Wired to: passenger service system (Amadeus Altea, Sabre, Navitaire), refund and payments systems, messaging channels.
• Watch out for: the refund deadline and the definition of a significant change are legal requirements, so they live in deterministic rules. The model handles conversation and option ranking, never eligibility.
• Illustrative crew and duration: tech lead plus four senior engineers, six to eight months: $55,000 to $75,000 a month, $330,000 to $600,000 in total.

7. Review and survey insight

The system reads reviews from booking sites and post-stay surveys across every property, classifies issues by department and root cause (noisy HVAC in a room range, slow breakfast at a specific hotel), and sends each general manager a weekly list with the quotes behind it.

• Feeds: reviews, survey responses, stay data to link feedback to rooms and dates.
• Wired to: reputation and survey tools (Medallia, Qualtrics, TrustYou), BI dashboards.
• Watch out for: sarcasm, mixed-language reviews and duplicates across sites. See NLP development.
• Illustrative crew and duration: tech lead plus one senior and one mid-level engineer, two to three months: $32,000 to $40,000 a month, $64,000 to $120,000 in total.

8. Corporate travel policy booking assistant

For travel management companies and corporate travel teams, an assistant books trips within each company's policy, explains why an option is out of policy, requests approval when needed and tracks travelers during disruptions for duty of care.

• Feeds: travel policies per client, traveler profiles, GDS and NDC content, approval workflows.
• Wired to: GDS (Sabre, Amadeus, Travelport), online booking tools, expense systems (SAP Concur), messaging.
• Watch out for: fares that change between search and booking, and policies with many exceptions. The assistant must confirm the final price before ticketing. See AI agent development.
• Illustrative crew and duration: tech lead plus three senior engineers, five to seven months: $45,000 to $60,000 a month, $225,000 to $420,000 in total.

Operations and payments

9. Housekeeping and labor forecasting

Forecasts of arrivals, departures, stayovers and outlet covers drive housekeeping, front desk and restaurant schedules a week ahead, updated daily as bookings change.

• Feeds: reservation forecasts, historical labor standards, outlet sales, event schedules.
• Wired to: PMS, workforce management, point-of-sale.
• Watch out for: union rules and scheduling laws that limit last-minute changes, so the forecast must land early enough to be used.
• Illustrative crew and duration: tech lead plus two senior engineers, three to four months: $35,000 to $45,000 a month, $105,000 to $180,000 in total.

10. Engineering and maintenance ticket triage

Guest complaints and staff reports become maintenance tickets classified by system, urgency and room, with repeat issues flagged so chief engineers can fix root causes and take rooms out of order before more guests are affected.

• Feeds: service requests, work order history, room and equipment data.
• Wired to: hotel maintenance and service platforms, PMS room status.
• Watch out for: free-text tickets with no room number, and safety issues such as gas smells that must page a person immediately.
• Illustrative crew and duration: tech lead plus one senior engineer, two to three months: $25,000 to $30,000 a month, $50,000 to $90,000 in total.

11. Room and property content mapping

OTAs and wholesalers receive the same hotel from many suppliers with different names, addresses and room descriptions. The system matches properties and room types across suppliers so travelers see one hotel with comparable rooms, and flags mismatches for content teams.

• Feeds: supplier property and room content, geocodes, images, manually confirmed matches.
• Wired to: content and inventory platforms, supplier APIs, search.
• Watch out for: room types that sound alike but differ in bed type or view, which cause complaints at check-in. See embeddings.
• Illustrative crew and duration: tech lead plus two senior engineers and one mid-level engineer, four to six months: $42,000 to $55,000 a month, $168,000 to $330,000 in total.

12. Booking fraud and chargeback defense

Models score bookings for fraud (stolen cards, last-minute high-value bookings, mismatched traveler details) and, when a chargeback arrives, assemble the evidence package: booking record, confirmation, check-in and folio. A Ryz AI pod built fraud detection for a global fleet company that scored 244K+ invoices in under 30 seconds each and surfaced $5.94M in fraud confirmed by the client's fraud team (case study).

• Feeds: booking and payment data, device signals, chargeback outcomes, stay records.
• Wired to: payment gateway and fraud tools, PMS or reservation system, dispute portals.
• Watch out for: blocking real last-minute travelers, and PCI scope. The model sees tokens, never card numbers.
• Illustrative crew and duration: tech lead plus three senior engineers, four to six months: $45,000 to $60,000 a month, $180,000 to $360,000 in total.

How to pick your first use case

• Start where contact volume peaks. Guest messaging and disruption handling deliver visible results the first busy weekend or storm.
• Put legal rules in code, not prompts. Refund triggers, fee display and compensation eligibility should be deterministic and tested; the model handles language.
• Pilot at a handful of properties or one hub. Property and airport differences break portfolio-wide launches. Prove it locally, then template it.
• Clean the content you depend on. Property facts, room types and policies feed every guest-facing system, so fixing them early pays several times.
• Score readiness. The AI readiness scorecard takes about ten minutes.

What it costs

Cost is the number of engineers multiplied by months and by monthly rate. Ryz rates are $7,000 to $10,000 for mid-level, $10,000 to $15,000 for senior and $15,000 and up for tech leads, quoted per team.

• Fee display or maintenance pilot (tech lead plus one senior engineer) for three months: $25,000 to $30,000 a month, $75,000 to $90,000 in total.
• Guest messaging pod (tech lead plus four senior engineers) for six months: $55,000 to $75,000 a month, $330,000 to $450,000.
• Full AI pod of about seven senior engineers including a tech lead and an ML engineer: $75,000 to $105,000 a month, typical for airline disruption programs.

Messaging, telephony, cloud and model usage are paid in your own accounts and are not included. AI development cost covers the full budget. You meet the named engineers first, and a team can usually start in under two weeks once you say go. If you need an out-of-the-box guest messaging app for a single property, buy one; Ryz pods build across your PMS, booking and service systems at portfolio scale.

Related

• Travel and hospitality engineering teams
• Hire AI agent developers
• AI integration services
• AI agents for customer service
• AI use cases in telecommunications
• AI use cases in retail and ecommerce

FAQ

What are the top AI use cases in hospitality?

Demand forecasting for pricing, guest messaging agents, upgrade and ancillary offers, group RFP drafting, review analysis, labor forecasting, maintenance triage and chargeback defense are the hotel systems that most often reach production.

How do airlines use AI during flight disruptions?

Airlines use AI to rank rebooking options for each passenger, handle the conversation across channels and process accepted changes. Refund and compensation eligibility stay in rules that follow the DOT refund rule and EU261.

Does the FTC fee rule affect hotel booking flows?

Yes. Since May 12, 2025, short-term lodging businesses and the platforms that sell them must show the total price, including mandatory fees, whenever they show a price. Automated checks across booking pages and partner listings help catch gaps.

Can an AI agent change a reservation in our PMS?

Yes, when it calls the PMS through its API with the same permissions as a staff member and confirms the change and price with the guest first. Changes involving payment disputes or exceptions go to staff.

Can a Ryz pod integrate with OPERA, Sabre or Amadeus?

Yes. Pods integrate with property management, GDS, passenger service and booking systems through their APIs and partner programs, in your cloud and repos. You control access throughout and own all code and IP.

Questions we didn't answer? Email info@ryzlabs.com.