Hotels Don't Need More AI Experts. They Need AI That Does the Work.

A hotel general manager decides it's time to introduce AI.

The business case seems clear. Guest inquiries arrive around the clock. Reservation teams spend hours answering repetitive questions. Employees switch between the PMS, messaging platforms, and operational systems to complete tasks that follow predictable patterns.

AI should help.

But then comes the implementation question: who is going to make it work?

Someone needs to evaluate vendors, configure workflows, connect systems, train employees, monitor results, and decide what the technology should be allowed to do.

For a hotel with 50 or 100 rooms, those responsibilities rarely belong to a dedicated AI department.

They land on the same people already running the property.

Hospitality's AI talent problem isn't simply a shortage of technical specialists. It's the growing expectation that existing hotel teams should become AI implementation experts on top of their regular jobs.

And that's a problem worth solving before buying another AI tool.

The AI Talent Gap Hospitality Doesn't Talk About Enough

Artificial intelligence promises to reduce routine work. Yet introducing it can create a different kind of workload.

An employee who previously answered guest inquiries may now need to configure automated responses, troubleshoot integrations, review AI-generated information, and decide when a conversation should be escalated.

The original work hasn't necessarily disappeared. It has changed shape.

This tension has been explored by hospitality industry researchers, including Michael J. Goldrich, founder of Vivander Advisors.

In February 2025, the HSMAI Foundation published The State of Generative AI in Talent Management for Hotel Sales, Marketing and Revenue Management, based on interviews and a survey of hospitality leaders.

The research examined how AI was changing commercial roles, professional development, recruitment, and organizational strategy.

One central implication was that successful AI adoption requires more than access to technology. Organizations need appropriate skills, leadership, and processes to use it effectively.

Read the HSMAI Foundation's research announcement.

That distinction matters for independent hotels, serviced apartments, and professional vacation rental operators.

A large hotel group may have dedicated technology and digital transformation teams.

A smaller operator may depend on a general manager, reservations supervisor, or operations director to manage the same decisions alongside their existing responsibilities.

The challenge isn't necessarily resistance to AI.

It's finding the capacity to implement it properly.

The Hidden Cost of Turning Hotel Employees Into AI Managers

Consider what happens when a hotel introduces an AI assistant to answer guest questions.

At first, the deployment appears straightforward. The assistant is trained on property information, connected to a communication channel, and made available to guests.

But operational questions quickly emerge.

What happens when the guest wants to modify a reservation rather than ask about the cancellation policy?

Who updates the PMS?

What if the guest requests an exception?

Who checks whether the response was accurate?

What happens when a message involves housekeeping, maintenance, or a payment-related issue?

If employees still need to complete every action manually, the hotel may have reduced the time spent answering questions without reducing the work behind them.

The AI is handling the conversation. The team is still operating the business process.

This is the hidden risk in hospitality automation: introducing technology that changes where employees spend their time without meaningfully reducing the work required.

For operators, the relevant measure isn't simply how many messages an AI assistant can answer.

It's how much of the underlying workflow the system can reliably complete.

Why AI Training Alone Won't Solve the Problem

AI literacy matters.

Hotel leaders should understand the capabilities, limitations, and risks of the technology their businesses use.

Employees need to know when to trust an automated process, when to intervene, and how to report a problem.

But training every hotel employee to become an AI specialist isn't a realistic operating strategy for most properties.

A front desk supervisor shouldn't need to understand complex AI architecture to manage a reservation workflow.

A housekeeping manager shouldn't need to become a software integration expert to receive a properly structured guest request.

And a general manager shouldn't have to spend hours writing prompts just to make routine automation function consistently.

The better question is how much technical complexity can be handled by the solution itself.

This doesn't eliminate the need for training.

It changes what employees need to learn.

Instead of teaching everyone how to build and maintain AI workflows, hotels can focus on understanding which processes are automated, what information the system uses, how exceptions are handled, and who remains accountable for the outcome.

That is a more practical approach to AI adoption for operators without dedicated technology departments.

From AI Tools to AI Digital Employees

Most hospitality software follows a familiar model.

A system stores information or provides functionality. An employee operates it.

The PMS manages reservations, but someone must create or modify a booking.

A messaging platform organizes guest conversations, but employees still need to determine what happens next.

A task management application tracks operational work, but someone must create, assign, and follow up on requests.

Each tool is useful. Yet the work of connecting them often remains manual.

An AI Digital Employee introduces a different operating model.

Instead of merely assisting employees with information, it can execute defined workflows within connected systems.

For example:

• Reservation inquiries: Retrieve relevant availability and rate information and support booking execution through a compatible PMS.
• Booking modifications: Process supported reservation changes according to configured property rules.
• Guest requests: Capture relevant details and create or route structured tasks for housekeeping or maintenance.
• After-hours communication: Handle routine inquiries when employees aren't immediately available.
• Human escalation: Recognize situations requiring staff judgment and pass along the necessary context.

Not every AI agent can perform these actions. The specific capabilities depend on its integrations, permissions, and operational design.

The distinction is important.

A conventional AI assistant may help an employee work faster. An AI Digital Employee is designed to take responsibility for completing supported portions of the work.

For hospitality operators, that can reduce the need to manually coordinate every routine interaction.

How Una Addresses the Operational Side of AI Adoption

Una by Polydom is a voice-first AI Digital Employee designed for hotels, serviced apartments, and professional vacation rental managers.

Rather than replacing a hotel's existing technology stack, Una is designed to work within it.

Through supported integrations and configured workflows, Una can communicate with guests, retrieve reservation information, perform authorized booking actions, coordinate operational requests, and escalate exceptions.

This matters because the value of automation doesn't come from the language model alone.

It comes from the combination of conversational understanding, access to operational systems, defined permissions, and the ability to execute the next step.

Consider a guest calling to extend their stay.

With a basic chatbot, the guest might receive an explanation of the hotel's extension policy.

With a properly configured AI Digital Employee, the request can become a reservation workflow: identify the booking, check relevant availability and conditions, process a supported modification, and communicate the outcome.

An employee becomes involved when an exception or additional judgment is required.

The result is not simply a faster answer.

It's potentially less routine work for the reservations team.

Learn how Una connects with property management systems and hotel operations.

Una still requires appropriate configuration, integration, testing, and oversight. No responsible AI deployment should be presented as entirely effortless or free of human supervision.

The goal is to reduce the ongoing operational burden, not pretend implementation requires no expertise.

What Happens When AI Becomes Part of Daily Hotel Operations?

The value of this approach becomes clearer when we examine a real deployment.

At Lee Gardens Extended Stay in Orlando, Florida, Una was introduced to support guest communication, direct reservations, and operational task coordination.

Over a ten-month reporting period, the property recorded:

• 4,104 guest conversations handled by Una, including 2,475 phone calls.
• 232 direct reservations made through Una, generating $80,968 in reported booking revenue.
• 172 operational tasks generated for hotel employees.
• 80 conversations requiring human interaction, approximately 1.95% of the total.

These figures describe one property's results. They do not prove specific labor savings, implementation costs, or a guaranteed return on investment.

But they illustrate something important about the allocation of work.

Thousands of conversations were handled without requiring an employee to participate in each interaction, while reservation and operational workflows continued to produce measurable activity.

The staff remained responsible for exceptions, decisions, and physical service delivery.

Una handled a defined portion of routine communication and execution.

Explore the Lee Gardens case study: $81K in Direct Bookings.

For hotel operators considering AI, this offers a more useful benchmark than counting how many tools their employees use.

The question is how much routine work can be completed reliably without repeatedly involving the team.

A Practical AI Adoption Playbook for Hotel Operators

Hotels don't need to begin with a complete transformation of their technology environment.

A more practical strategy is to identify a narrow operational problem, establish clear responsibilities, and measure what changes.

1. Find the Work That's Repeated Every Day

Start with processes employees handle constantly.

These may include responding to reservation inquiries, answering routine guest questions, processing supported booking modifications, or forwarding housekeeping and maintenance requests.

Look beyond the volume of conversations.

Identify how many manual actions follow each interaction and which of those steps could be automated.

The most promising opportunities are often the workflows that are predictable, repetitive, and governed by clear property rules.

2. Choose One Business Outcome

An AI deployment becomes difficult to evaluate when it promises to improve everything simultaneously.

Choose a primary objective.

For a property with limited overnight staffing, that may be consistent 24/7 guest communication.

For a reservations team, it may be increasing the number of direct inquiries that progress to completed bookings.

For a multi-property operator, it may be reducing repetitive coordination between guest communication channels and operational teams.

The outcome should determine which workflow is automated first.

3. Define What AI Can and Cannot Do

Before deployment, operators should establish clear boundaries.

Which reservation changes can be completed automatically?

When should a guest request be escalated?

Who can approve exceptions?

Which operational tasks can be created or routed through connected systems?

And who is responsible for reviewing problems?

These questions are more important than choosing a system based solely on how natural its conversations sound.

4. Measure Completed Work, Not Just AI Activity

A high number of automated responses doesn't necessarily indicate business value.

Operators should also evaluate:

• The number of supported bookings or modifications completed.
• The share of conversations requiring human escalation.
• The volume and quality of operational tasks generated.
• Whether routine requests are handled outside staffed hours.
• The frequency of errors, exceptions, and manual corrections.
• The workload that remains with employees after implementation.

These measures help distinguish actual operational improvement from automation that merely shifts work elsewhere.

5. Train Employees for Oversight and Judgment

AI adoption shouldn't mean removing human expertise from operations.

Employees need to understand the processes Una is configured to handle, where its authority ends, and when intervention is necessary.

Supervisors should be able to review performance and identify recurring issues.

The hotel's team remains accountable for service standards and decisions requiring discretion.

The goal is not to turn every employee into an AI engineer.

It's to make AI part of a manageable operating model.

The Human Skills Hotels Should Protect

There is a danger in treating AI adoption purely as a staffing reduction exercise.

Hospitality businesses depend on people who understand guests, properties, teams, and situations that don't follow standard procedures.

A reservation system can store a guest's preferences. It cannot independently determine how a hotel should respond to every sensitive situation.

An AI agent can route a maintenance request. Someone still needs to assess and resolve the physical problem.

Automation can help enforce established procedures. Human leaders remain responsible for deciding when those procedures require exceptions.

This is why hospitality's future workforce strategy shouldn't simply be about hiring fewer people or expecting existing employees to manage increasingly complex technology.

It should be about allocating responsibilities more intelligently.

The HSMAI Foundation's research on AI and talent management emphasizes the importance of developing human capabilities alongside technology adoption.

Explore HSMAI's hospitality AI talent research.

For hotel operators, that means protecting the judgment, service expertise, and relationships that distinguish good hospitality while reducing repetitive administrative work.

The best outcome isn't a hotel that needs no people. It's a hotel where people have more capacity to do the work that matters most.

Do Hotels Need AI Specialists to Adopt AI Agents?

Not necessarily.

A hotel does not need to build its own AI engineering department to use a commercially available AI agent.

However, operators still need an appropriate implementation process, reliable integrations, property-specific configuration, staff training, and ongoing oversight.

The requirements depend on the complexity of the deployment.

An independent hotel introducing after-hours guest communication may face a different implementation challenge from a hotel group coordinating reservations and operational workflows across multiple locations.

The key is to select technology that fits the property's existing systems and operating model.

Operators should ask vendors what implementation involves, which workflows are supported, how changes are managed, and what responsibilities remain with the hotel team.

A solution that reduces routine execution but introduces significant ongoing supervision may offer less value than expected.

The Real AI Talent Advantage Is Operational

Hospitality's AI talent challenge won't be solved by expecting every general manager, reservations supervisor, and front desk employee to become a technology specialist.

Nor will it be solved by replacing hospitality professionals with systems that still require constant human intervention.

The opportunity is to rethink how work is divided.

AI can handle supported, repeatable execution within connected systems.

Employees can retain responsibility for exceptions, service quality, guest relationships, and decisions requiring judgment.

Hotel leaders can focus on measurable business outcomes instead of adopting technology for its own sake.

That is the operating model Una by Polydom was created to support.

Hotels don't necessarily need more people to operate AI. They need AI that can reliably handle more of the work their people already do.

Explore How Una Works in Your Property

Curious how Una works in real hospitality operations?

Talk to our human sales team, explore existing deployments, and identify which routine workflows could be supported within your property's technology stack.

Book a quick call here to discuss guest communication, direct booking automation, and operational coordination with Una.