Why AI Has Not Solved the Hospitality Connectivity Problem

The tools are not the problem. The infrastructure underneath them is. AI agents need clean data, clear context, and well-structured access to backend systems. But most hospitality stacks were built for a different era, and simply layering AI on top of them is not enough.

This white paper breaks down:

• Why this moment is different from previous technology cycles, and why the stakes are higher than most operators realise

• Why high AI usage is not translating into measurable returns, and the three root causes behind that gap

• What good connectivity actually looks like when AI agents, not humans, are the consumer of your systems

If you are an operator trying to figure out where to invest, or a technologist trying to turn AI pilots into production outcomes, this paper is for you.

5 Key Takeaways:

1. By 2027, AI agents will generate more internet traffic than humans — and over 95% of AI-driven traffic in 2025 was concentrated in retail, streaming, and travel and hospitality. If your systems are not accessible to agents, you are invisible to a growing share of demand.

2. Only 12% of organisations are seeing measurable returns from AI — not because the tools do not work, but because the foundations required to unlock value are missing.

3. AI does not fix bad data. It amplifies it. Hotels already miss up to 20% of potential revenue due to data fragmentation. Layer AI on top of that and the problems do not disappear — they scale.

4. Wrapping your existing APIs in MCP is not a strategy. Traditional APIs were built for engineers, not agents. Without the right abstraction layer, you are pushing complexity, cost, and hallucination risk into the model.

5. In a world where writing code is no longer the differentiator, domain expertise is. The organisations winning with AI are not the ones with the most tools — they are the ones who understand what to expose, how to structure it, and what good looks like from a business perspective.