The hospitality algorithm: how AI is rewriting hospitality
According to research conducted by Iulm, adoption is not homogeneous and for now focuses on operations
Key points
Artificial intelligence is no longer a futuristic conference theme for the hospitality sector: technology has already entered the operational processes of most hospitality facilities, but has done so in an uneven, often tactical, rarely strategic manner. This is, in a nutshell, the picture taken by the research presented by Iulm University and the result of a survey conducted between April and July 2025 on a sample of 864 Italian companies in the sector.
Focus on generative AI for operations
The most evident datum is probably related to the diffusion of AI tools, used in at least one case by 83% of facilities; among those that do not yet use them, 40% plan to do so within the next three years. Adoption is therefore transversal but particularly present in realities such as agritourisms and B&s, confirming the trend whereby small-scale structures now have access to low-cost digital solutions that can be implemented immediately. But how is the capacity of algorithms and chatbots being exploited? Usage, as the study confirms, remains for now concentrated on operational applications and specifically involves generative AI tools for text production and translation, customer communication management, responses to online reviews and digital marketing content creation.
About half of the active structures declare a daily or weekly use of these tools, while the use of predictive systems for analysing demand, prices and competition is more limited and the use of customised recommendation solutions even less widespread. The intervention of technology is therefore concentrated on front office, sales & marketing, revenue management and management processes but does not replace the human decision-maker: if anything, it automates repetitive activities and supports him in analysing results. In essence, the report goes on to say, AI is an efficiency lever rather than a business model transformation engine. It is no coincidence that 63% of those interviewed judge it to have no or marginal perceived impact on performance, while the most relevant effects concern increased efficiency (24% rate it high or very high), improved customer satisfaction and online reputation (19%) and optimised pricing (15%).
Untapped potential and the skills challenge
The potential of AI, as the data collected by IULM clearly show, is high but still far from full maturity, partly because of a still weak 'governance' component. In fact, 68% of the structures do not have dedicated figures for AI and the use of the tools is left
to employee initiative, with a prevalent use (over 90%) of standard, very cheap (or even free) and ready-to-use solutions. Only 11% of companies claim to have a specific policy on AI and just 12% invest in dedicated training courses. The lack of skills and rules is counterbalanced by an average high level of satisfaction with this technology (especially for ad hoc solutions), but it is equally clear that there is no shortage of criticalities, from fears related to data processing to the reliability of the systems and the risk of compromising the relationship with the customer in the event of ungoverned use of chatbots and automated systems.


