AI-Driven Digital Twins Could Reshape Hospitality, but Skills and Infrastructure Hold the Key

Artificial intelligence is entering hotel operations at a moment when the industry is under pressure to control costs, improve energy efficiency and deliver more personalized services. AI-enabled Digital Twins offer a powerful route toward those goals. By combining real-time data, sensors, predictive analytics and virtual simulations, they can help hotels anticipate equipment failures, manage energy use, allocate resources and make faster operational decisions.

However, the technology's value does not guarantee its adoption. Hotels must have the infrastructure, skills, financial capacity and institutional support needed to integrate Digital Twins into everyday operations. A study titled "Organizational adoption of AI-driven digital twins: evidence from the hospitality industry," published in Frontiers in Artificial Intelligence, examines these conditions through the responses of 492 hotel managers in Serbia and Hungary. The research was conducted by Tamara Gajić, Dragan Vukolić, Marko D. Petrović, Anastasiya A. Golubeva, Dunja Demirović Bajrami, Marija Bojić, Sergey V. Pashkov, Nesrin Atasoy, Tolga Kara and Emin Atasoy.

The research offers a practical warning for the global AI transition. Digital Twins may become powerful tools for more efficient and sustainable hospitality, but their success will depend on the institutions and organizations surrounding them. Hotels need more than access to advanced technology. They need the capabilities, confidence and policy environment required to make it work.

Digital Twins Need an Ecosystem, Not Just a Business Case

DTs create virtual representations of physical operations by combining real-time data, sensors, cloud systems, artificial intelligence and predictive analytics. In a hotel, such systems can connect information on occupancy, energy consumption, equipment performance, guest preferences and service delivery.

The potential gains are significant. Managers could identify operational inefficiencies before they become costly, predict equipment failures, optimize heating and cooling, improve resource allocation and support more precise revenue decisions. Digital Twins could also help hotels reduce energy consumption and strengthen sustainability efforts.

Implementation, however, is more demanding than purchasing new software. Hotels must integrate data from different systems, maintain reliable sensor networks, train staff and adjust established workflows. In many cases, they must also address cybersecurity, data governance and interoperability concerns.

The study treats adoption as an organizational transformation process rather than a simple technology-acceptance decision. Its analytical framework combines established technology-acceptance factors with two additional dimensions: digital readiness and institutional support.

Digital readiness refers to an organization's technical infrastructure, employee skills and ability to integrate advanced digital systems. Institutional support includes regulatory frameworks, public programmes, industry guidance and wider conditions that reduce uncertainty around technology adoption.

Serbia and Hungary Reveal Two Different Adoption Patterns

The researchers collected data between January and June 2025 from managers working in three-, four- and five-star hotels in major urban tourism destinations. The sample included 234 respondents from Serbia and 258 from Hungary.

The study found that six factors significantly influenced managers' intention to adopt AI-enabled Digital Twins in both countries:

• expected performance benefits
• ease of use
• social influence
• facilitating conditions
• digital readiness
• institutional support

The relative strength of these factors differed substantially. In Serbia, social influence was the strongest driver, followed by facilitating conditions and ease of use. The coefficient for social influence was 0.415, compared with 0.300 for facilitating conditions and 0.285 for effort expectancy.

This pattern suggests that Serbian managers pay close attention to competitors, customers, professional associations and broader digital transformation trends. They also place considerable weight on whether the necessary infrastructure, technical assistance and financial resources are available.

Hungarian managers displayed a different hierarchy. Ease of use was the strongest driver, with a coefficient of 0.343, followed by digital readiness at 0.320. Social influence, facilitating conditions and expected performance benefits followed.

Managers in Hungary appeared more focused on whether Digital Twin systems could be integrated efficiently into existing operations and whether their organizations possessed the technical and human capabilities required for implementation.

The cross-country differences were statistically significant for ease of use, digital readiness, social influence and facilitating conditions. Digital readiness had a stronger influence in Hungary, while social influence and facilitating conditions had stronger effects in Serbia.

The Value of AI Is Recognized, but Readiness Determines Action

Managers in both countries broadly recognized the potential organizational value of Digital Twins. Performance expectancy was statistically significant in each sample, with coefficients of 0.198 in Serbia and 0.244 in Hungary. The difference was not statistically significant, indicating that hotel managers in both settings generally associated the technology with better operational performance, cost control, resource allocation and managerial decision-making.

The gap emerged when managers considered how adoption could actually happen. Serbian respondents appeared more dependent on peer examples, external expectations and organizational support. Hungarian respondents placed greater emphasis on internal capabilities and implementation efficiency.

Institutional support had a positive but comparatively weaker influence in both countries. Its coefficient was 0.180 in Serbia and 0.150 in Hungary. The difference between the two countries was not statistically significant.

This finding suggests that policy support can create an enabling environment, but it is unlikely to compensate for weak internal capabilities. Regulations, incentives and public programmes may encourage adoption, yet hotels still need trained employees, reliable infrastructure and systems that fit everyday operations.

The model explained 48.2% of the variation in adoption intention in Serbia and 53.7% in Hungary. The results provide substantial evidence that technology perceptions, organizational capacity and institutional conditions operate together, while also leaving room for other factors not captured by the model.

Policy Must Close the Capability Gap

Connectivity and technology subsidies alone will not be enough if hotels lack the skills, financing and technical support needed to integrate advanced systems. Countries with lower levels of digital maturity may need coordinated programmes combining infrastructure investment, workforce training, advisory services, access to finance and stronger collaboration between governments, hotel associations, universities and technology providers.

Demonstration projects could help reduce uncertainty by showing how Digital Twins perform in actual hotel environments. Shared digital platforms and technical support centres could also lower the costs faced by smaller establishments that cannot independently build complex data systems.

More digitally mature markets may require a different policy emphasis. Interoperability standards, innovation partnerships, cybersecurity guidance and the diffusion of proven applications could help hotels integrate AI more efficiently into existing operations.

Hotel managers should assess organizational readiness before committing to large-scale investments. Employee training, data-quality checks and workflow redesign need to accompany the technology itself. A Digital Twin that cannot communicate with existing hotel-management systems, or that staff find difficult to use, may create additional costs rather than operational gains.

Technology providers also face a clear responsibility. Platforms should be designed for practical integration, transparent decision-making and gradual adoption. User-friendly interfaces and flexible architecture could be particularly important in markets where technical expertise remains uneven.

The implications extend beyond hospitality. Similar patterns are likely to emerge in logistics, manufacturing, healthcare, transport and public infrastructure, where AI systems depend on physical assets, real-time data and organizational coordination.

Strong Evidence, but Limited Reach

The study benefits from a clear comparative design and a sample of managers directly involved in technology and investment decisions. The researchers pretested the questionnaire, consulted academics and information-technology professionals, and conducted statistical checks to assess reliability, validity and comparability between the two national samples.

That said, several limitations require caution. The research uses a cross-sectional design, so it cannot establish whether the identified factors cause adoption or show how attitudes change over time. The findings are also based on self-reported managerial assessments rather than verified implementation records or measured operational outcomes.

Purposive sampling helped the researchers reach relevant hotel managers but limits the statistical generalizability of the results. Serbia and Hungary offer a useful comparison, but their experience cannot automatically represent hotels in other European markets or in developing countries across Asia, Africa and Latin America.

The study measures intention to adopt rather than completed deployment. Managers may express interest in DTs without securing investment, changing workflows or sustaining the technology over time. Future research should therefore track adoption from initial consideration through implementation and evaluate whether promised benefits appear in energy use, costs, service quality and employee productivity.

Additional studies could examine cybersecurity readiness, organizational culture, leadership support, data governance and the role of small and medium-sized hotels. Qualitative interviews and detailed case studies would also reveal how managers negotiate technical risks, employee concerns and investment decisions in practice.