Artificial intelligence is beginning to change two fundamental parts of the hotel business: how travellers choose properties and how hotels deliver a stay. AI systems can influence hotel discovery, handle routine operational tasks and use connected guest data to support more personalised service. For hotel operators, the change goes beyond chatbots and standalone AI tools.
AI is becoming part of a wider technology environment that includes property management systems (PMS), customer relationship management (CRM) platforms, revenue systems and guest communication tools. When these systems can exchange reliable information, AI can support decisions and actions across more of the guest journey.
The challenge is that adopting AI is not simply a matter of adding another technology product. Hotels need accurate data, connected systems and clear rules about which decisions can be automated and which require human oversight. These foundations will determine whether AI improves efficiency and the guest experience or simply adds another layer of complexity.
How AI is changing hotel discovery
Hotel discovery is beginning to extend beyond traditional search engines and online travel agencies (OTAs). Travellers can increasingly use conversational AI to describe a trip in ordinary language and receive hotel recommendations based on criteria such as location, price, facilities and individual requirements.
This is also creating the foundations for more autonomous booking journeys. AI-assisted systems can help travellers search, compare and shortlist properties, while emerging agentic AI models are designed to take actions on a user's behalf.
The distinction matters. AI-assisted discovery is already possible, but fully autonomous hotel booking is not yet the standard way people buy accommodation. Technology platforms, hotel companies and intermediaries are still developing the technical and commercial models that could support more agent-led transactions.
In July 2026, Radisson Hotel Group and Accenture launched a hotel discovery application within ChatGPT. Travellers can use conversational queries to search more than 1,000 Radisson properties across more than 100 countries. Results can incorporate rates, inventory, amenities and location information before directing customers to the hotel group's website to complete a reservation.
For hotels, this creates another audience for digital information: AI systems that help people make decisions. Hotels have traditionally optimised digital information for travellers and search engines. As AI becomes part of the discovery process, property information also needs to be easy for machines to interpret accurately.
Traditional hotel SEO remains important, but AI-led discovery adds another consideration. A 2026 audit covering 12 AI models found that guest ratings and price had particularly strong effects on hotel selection. It also identified an influence from list position, showing that factors unrelated to a hotel's underlying quality can affect AI-generated recommendations.
Automation is changing routine hotel operations
The second major impact of AI in hotels is less visible to travellers but potentially just as important. Hotel automation can reduce repetitive administrative work and give employees more time for tasks that require judgement, problem-solving and personal service. Guest communication is an obvious example: questions about check-in times, breakfast, parking and Wi-Fi are predictable and frequently repeated.
Check-in offers another opportunity. Digital workflows can collect required pre-arrival information and send standard communications before a guest reaches reception. Similar processes can support check-out, housekeeping requests and other routine parts of a stay.
Not all of this automation requires AI. Conventional automation follows predefined rules, while AI can help interpret a request, identify patterns or generate an appropriate response within defined limits. The bigger opportunity for hotels comes when AI, automation and integrated operational systems work together.
AI can also support work behind the scenes. Hotel technology platforms can assist with demand and staffing forecasts, service requests, maintenance planning and revenue decisions. Connected systems can share inventory and availability data, reducing the need for employees to enter the same information repeatedly across separate platforms.
Integration is therefore more important than automation for its own sake. A guest-facing AI assistant that cannot access current reservation details, availability or hotel policies has limited operational value. When authorised systems can securely exchange relevant information, a request can move from conversation to action with less manual intervention.
Not every decision should be delegated to AI. Complex payment issues, unusual booking changes, complaints, safety concerns and exceptions to policy still require human judgement. Hotels need clear escalation rules so automated processes know when to hand control to an employee.
Operators should measure AI and automation against business outcomes rather than the volume of tasks handled by technology. Useful measures include response times, resolution and escalation rates, errors, staff workload, guest feedback and, where relevant, conversion or incremental revenue. A system that automates a large number of tasks but creates more exceptions to correct may simply move work rather than remove it.
Connected data is making personalisation more useful
Personalisation is one of the most frequently promised benefits of AI in hospitality, but its effectiveness depends heavily on the quality of the hotel's underlying data. Traditional hotel personalisation often relies on employees recognising returning guests or recording preferences in reservation notes. This can produce excellent service, but it is difficult to deliver consistently across shifts, departments and properties.
Connected hotel systems offer a more systematic approach. PMS, CRM and loyalty platforms can bring together appropriate information about previous stays, preferences and facility interactions. AI can then help determine which information is relevant at a particular point in the guest journey.
The aim is not simply to remember more about each guest. It is to use relevant information at the right moment. Context remains critical: previous choices do not automatically describe what a person wants on every visit. The same guest may travel for business on one occasion and leisure on another. Effective hotel personalisation needs to consider the circumstances of the current stay rather than simply repeat past behaviour.
Hotels need dependable data, agreed standards and appropriate connections between core platforms before they can expect AI to produce consistently useful recommendations or actions. Those operations teams responsible for building those connections will need skills in data architecture, system integration and AI oversight - training that can be accessed through programmes like AI for Operations courses.
There is also a governance requirement. Hotels hold personal and transactional information, and greater use of AI does not remove their responsibilities for how that information is collected, accessed and used. Operators need to consider data minimisation, access controls, retention policies, cybersecurity and human oversight alongside the commercial benefits of AI.
Why this matters for Operations professionals
The competitive advantage is unlikely to come from having the most AI tools. It will come from creating a technology environment in which machines can handle routine processes reliably while hotel teams remain focusing on moments that require human judgement, empathy and service. For operations professionals, the immediate priority is building the foundations: accurate data, connected core systems and clear rules about which decisions are delegated and which require human review.
Your membership also unlocks:
