Scaling Hospitality AI: Standards, Systems & AIHA

An AI agent that works well at one hotel may face a very different test across an entire portfolio.

Different property management systems, operating rules, permissions, and escalation workflows can quickly turn a promising pilot into a complex integration challenge. What works in a controlled environment does not necessarily remain reliable across dozens of properties, each with its own technology stack and operational requirements.

For hotel operators, the real challenge is not simply adopting AI. It is making AI reliable, consistent, and scalable across everyday operations.

Two initiatives from the AI Hospitality Alliance (AIHA) address different sides of that challenge: developing shared technical foundations for hospitality AI and documenting real-world applications across the industry.

Polydom is contributing to this work as a Founding Partner of AIHA. Our Co-founder and CEO, Lana Udalov, serves on the Alliance's Advisory Board, while Polydom CTO Evgeny Chernyshov participates in two technical committees focused on the architecture required to make hospitality AI work at scale.

Why Hospitality AI Needs Shared Technical Foundations

Under the leadership of AIHA Founder Ira Vouk, the Alliance has brought together hospitality operators, technology providers, researchers, consultants, and industry organizations to address challenges that no single company can solve independently.

In July 2026, AIHA announced eight technical workstream committees as part of its 12-month industry roadmap.

At the time of Polydom's original report, committee participation had grown to 84 seats, 59 individual contributors, and 37 companies. These figures reflect the participation reported at that stage of the initiative, rather than its current membership.

The workstreams were established to develop practical blueprints, standards, guidelines, and measurement frameworks covering:

Agentic protocols adapted to hospitality
Middleware orchestration
Scalable AI platform architecture
AI-ready data and shared terminology
AI visibility criteria and performance indicators
Organizational AI readiness
AI governance and operating guidelines

The purpose is not to promote a single technology provider or property management system. It is to help the hospitality industry establish common principles for deploying AI across different systems and operating environments.

The AIHA Standards library now provides open access to published frameworks, implementation guides, and architectural resources developed through this collaborative work.

Polydom CTO Evgeny Chernyshov Joins Two Technical Committees

Polydom CTO Evgeny Chernyshov joined two workstreams directly relevant to the challenges encountered when deploying AI digital employees in operational hospitality environments.

Middleware Orchestration Layer Blueprint

This committee focuses on how AI systems should coordinate actions across hotel platforms and operational applications.

An AI agent may need to retrieve live availability, create or modify a reservation, update a PMS record, initiate an operational task, or escalate a situation to a human employee.

Each action may involve different systems, permissions, data structures, and business rules.

Connecting these processes reliably is an orchestration challenge, not simply a conversational AI challenge.

Scalable AI Platform Architecture Blueprint

The second committee focuses on reference architecture for hospitality AI platforms that can remain reliable and extensible as deployments expand.

A solution that functions successfully at one hotel may encounter new challenges when introduced across a portfolio with different PMS configurations, integration requirements, and operational policies.

As Evgeny explained in his announcement about joining the committees:

“If we get the middleware and architecture layers right, everything built on top of them gets easier for everyone—vendors and hoteliers alike.”

For hotel groups and professional property managers, the commercial implications are significant.

Every additional property and integration can introduce maintenance requirements, operational inconsistencies, and potential points of failure. Shared architectural principles could reduce unnecessary duplication and make future AI deployments easier to maintain and extend.

That does not mean standardization will eliminate integration complexity. It means the industry has an opportunity to address recurring problems collectively rather than requiring each operator and vendor to solve them independently.

Why Cross-Platform Collaboration Matters

One of the most valuable aspects of the AIHA initiative is participation from across the hospitality technology ecosystem.

Senior leaders and technical contributors associated with major hospitality platforms have been represented within the broader Advisory Board and committee structure, including:

Mews: Founder Richard Valtr, serving on the Advisory Board
Cloudbeds: Founder and CEO Adam Harris, alongside additional middleware committee participation
Oracle Hospitality: Laura Calin, SVP of Oracle Consumer Industries
Apaleo: Founder Philip von Ditfurth, with technical and product leaders participating in multiple committees
RMS Cloud: CEO Adam Seskis, alongside product, engineering, and technical contributors

Their involvement creates an opportunity to consider different architectures, integration models, and operating environments when developing shared industry guidance.

Participation should not be interpreted as an endorsement of any particular vendor, integration, or commercial solution.

For hotel operators, the value lies in the possibility of more consistent approaches to connecting AI systems with the platforms they already use.

A growing hotel group should not have to rebuild its entire AI operating model every time it adds another property or introduces a different technology provider.

What Polydom Brings to the Discussion Through Una

The architectural questions being addressed by AIHA are closely connected to Polydom's practical experience developing and deploying Una by Polydom.

Una is a voice-first AI Digital Employee designed to support hospitality operations through guest communication, eligible reservation actions, workflow coordination, and human handoffs.

Instead of requiring staff to manage every routine exchange manually, Una can carry out supported actions within an operator's existing systems, according to configured permissions, business rules, and available integrations.

Consider a guest requesting a reservation change.

Answering the question is only one part of the work. Completing the request may require checking the reservation, reviewing applicable policies, accessing availability, making an authorized system change, communicating the outcome, and involving a team member if an exception requires human judgment.

The business value comes from coordinating those steps reliably, not simply generating a convincing response.

Through live implementations across hotel and vacation-rental operations, Polydom encounters the practical challenges that emerge when AI moves beyond demonstrations and into everyday hospitality workflows.

These include integration reliability, data accuracy, escalation logic, differences between properties, and maintaining operational consistency as deployments expand.

Una's specific actions depend on each property's integrations, permissions, and configuration. Human teams retain responsibility for decisions and exceptions that require their authority or judgment.

This operational perspective is what Polydom brings to the AIHA technical committees: experience with the conditions under which AI must actually perform, rather than assumptions about what it might eventually accomplish.

AIHA and HEDNA: Mapping Real-World Hospitality AI Applications

Shared technical architecture is only one part of the challenge.

The other is understanding where AI is already being applied, which operational problems it addresses, and what evidence exists to support adoption decisions.

In July 2026, AIHA and HEDNA launched a crowdsourced Hospitality AI Use Case Knowledge Base.

The initiative invited hoteliers, technology providers, consultants, educators, and other hospitality professionals to contribute real-world and pilot applications of AI.

As Ira Vouk explained when the initiative was announced:

“The hospitality industry does not need more abstract conversations about AI. It needs practical examples of where AI is being tested, adopted, and measured.”

The project organizes hospitality AI applications across five major functional areas:

Guest Experience
Revenue and Business Intelligence
Operations
Distribution and Commerce
Marketing and Sales

Contributors were invited to identify the operational problem, describe how AI was applied, explain the outcomes or observed benefits, and document the role of technology systems and employees.

The objective was to create a shared resource that would help hospitality professionals move from general enthusiasm about AI to more informed implementation decisions.

From an Open Call to a Published Industry Resource

The initiative has since progressed beyond its initial crowdsourcing stage.

According to the AIHA research publication, the study published in September 2026 collected 198 industry submissions, consolidated into 109 unique AI use cases across 39 hotel systems.

The publicly available catalogue subsequently expanded to 112 use cases as additional submissions were incorporated.

These figures describe submitted and categorized applications. They should not be interpreted as evidence that every use case has demonstrated commercial ROI or successful deployment at scale.

Nevertheless, the catalogue offers a more concrete starting point for operators evaluating AI opportunities.

Rather than asking only whether AI can automate a particular activity, hotel leaders can examine more useful questions:

Which operational problem is the application intended to solve?
What hotel systems and data does it require?
Does the AI provide information, recommend an action, or execute an authorized workflow?
What remains the responsibility of hotel employees?
What results have been documented, and how strong is the supporting evidence?
Could the same approach work across multiple properties?

The distinction between a documented use case and a proven business outcome is important.

A catalogue can help identify possibilities, compare approaches, and expose recurring operational challenges. Each individual implementation still requires evaluation against its own costs, constraints, risks, and business objectives.

Why Hospitality AI Standards Matter to Hotel Operators

Hospitality AI standards are not merely an engineering concern.

For a hotel owner, general manager, or technology leader, their potential value can be understood through three practical questions.

Can the AI work with our existing systems?

Most operators already depend on several technology platforms for reservations, distribution, guest communication, payments, and day-to-day operations.

An AI solution must operate within that environment rather than assuming the hotel will replace its existing systems.

Shared approaches to integration and interoperability could make deployments easier to evaluate and reduce unnecessary technical fragmentation.

Can we maintain consistent operations as we grow?

A multi-property operator may have different workflows, teams, policies, and technology configurations across its portfolio.

Scaling AI requires a way to preserve appropriate operating rules while accommodating those differences.

The goal is not to make every property identical. It is to support consistent execution without losing the flexibility needed for local operations.

Can we trust the actions the AI performs?

An AI system that answers a question carries a different operational responsibility from one that changes a reservation or initiates a workflow.

Operators need to understand which actions are authorized, how they are validated, when human approval is necessary, and how exceptions are handled.

Architecture and governance become especially important when AI moves from providing information to taking action.

These are precisely the kinds of implementation questions that make the AIHA initiatives relevant beyond the technology community.

From Shared Knowledge to Shared Infrastructure

Since the original announcement, AIHA has begun publishing the practical outputs of its technical workstreams.

The AIHA Standards library includes resources addressing scalable AI platform architecture, orchestration, AI-ready hospitality data, organizational readiness, governance, and other implementation priorities.

These publications provide voluntary industry guidance, not a universal guarantee that different hotel systems or AI products will work together automatically.

Their significance is that hospitality organizations now have more concrete material to reference when assessing architecture, vendor approaches, operational controls, and potential deployment risks.

Together, the two initiatives address complementary needs.

The technical workstreams provide frameworks for how hospitality AI can be built and governed. The AI Use Case Catalogue helps the industry understand where AI is being applied and what operational questions deserve attention.

For operators, both are necessary.

Without reliable implementation foundations, promising AI applications may become difficult to maintain and scale.

Without practical use cases and meaningful evidence, architectural discussions risk becoming disconnected from the actual needs of hotel teams and their guests.

Building the Next Stage of Hospitality AI Together

Polydom is proud to contribute to this industry-wide effort through Lana Udalov's participation on the AIHA Advisory Board, Evgeny Chernyshov's technical committee work, and our experience developing Una for real hospitality operations.

We also recognize the leadership of Ira Vouk in bringing together organizations that might otherwise approach these challenges independently, and HEDNA's contribution to building an open knowledge base for the industry.

For hotel operators, the larger lesson is straightforward.

The future of hospitality AI will not be determined by how impressive an individual demonstration looks. It will depend on whether AI can work reliably within existing systems, execute appropriate workflows, support human teams, and maintain consistency as operations grow.

Better architecture alone will not deliver these outcomes. But without it, scaling AI becomes much harder than it needs to be.

Explore the AIHA Standards and Frameworks and the Hospitality AI Use Case Catalogue to understand how the industry is developing shared foundations for practical AI adoption.

And if you are evaluating how an AI Digital Employee could support workflows across your own hotel or property portfolio, explore Una by Polydom and the role it can play within your existing operating environment.