Why many hotels still struggle to make AI work

The real challenge is not artificial intelligence itself, but the outdated hotel systems sitting underneath it

Artificial intelligence is becoming one of the hospitality industry’s favourite buzzwords, with promises of smarter pricing, automated guest communication and more personalised service. Yet many hotels are discovering that implementing AI successfully is far more difficult than simply adding a chatbot or revenue tool to their technology stack. The article argues that the biggest obstacle is not the capability of AI, but the fragmented and outdated infrastructure still used by many hotel businesses. Without connected systems and reliable data flows, even advanced AI tools often fail to deliver meaningful operational or commercial improvements.

Key takeaways

• Legacy PMS platforms remain a bottleneck: Many hotels still rely on outdated property management systems and disconnected software that were never designed for today’s real-time, data-driven operating environment.
• AI is only as good as the data behind it: Revenue management, guest messaging and automation tools depend on accurate and connected data sources. Fragmented systems often result in incomplete or unreliable AI outputs.
• Disconnected systems reduce automation benefits: Hotels frequently deploy AI tools on top of existing infrastructure without proper integration. This creates silos that still require manual work from staff, limiting efficiency gains.
• Guest experience can suffer when AI lacks context: Chatbots and AI assistants that cannot access reservation details or guest history may provide inaccurate responses, forcing guests to repeat requests to hotel staff.
• Modern cloud-based systems are becoming essential: Hotels looking to benefit from AI may first need to modernise their core technology stack, particularly PMS and integration layers that allow data to move across departments more effectively.
• Gradual transformation often works better: The article suggests that phased technology upgrades are usually more practical than large-scale overhauls, helping hotels improve compatibility while reducing operational disruption.
• Staff adoption matters as much as technology: Training hotel teams to use AI tools consistently and correctly remains a critical part of successful implementation strategies.
• AI should solve operational problems, not follow hype: Hotels tend to achieve stronger results when AI is introduced with a clear business objective, such as reducing response times, improving pricing accuracy or streamlining check-in processes.

Source: Hotel Management

Enjoying this analysis? Hospitality.today delivers daily insights on hotel distribution, AI trends, and travel commerce — straight to your inbox. Subscribe for free at Hospitality.today →