What AI has actually changed, and what it has not

TechnologyAugust 2026Insight 04

82% of hotels expect AI use to grow this year, and 85% will put at least 5% of IT budget behind it. Only about a third have it embedded across most functions.

A global survey of more than 400 hotel technology decision makers, published by Canary Technologies in March 2026, found 82% expecting AI use to increase across their organisation within the year, 85% allocating at least 5% of IT budget to it, and 71% describing the impact on the industry as significant or transformative. Separate owner research puts the share of hotel owners using AI in some form as high as 98%.

Depth is another matter. Roughly a third have it embedded across most functions. Most adoption clusters in three places: guest messaging, revenue management and back office automation. Industry analyses report RevPAR improvements in the region of 10 to 15% where AI-driven revenue management replaces traditional methods.

The part worth understanding is what this does to a career, because it is not what people assume. AI has not removed the revenue manager. It has removed the part of the revenue manager's day that consisted of rebuilding a spreadsheet. What is left is judgement: which segments to protect, when the model is wrong, what a competitor's rate move actually signals, and whether a discount buys demand or simply rents it from a channel that was going to book anyway.

That shifts the skill premium upward rather than downward. A person who can operate the tool but cannot read the output is now less useful, not more, because the tool has made producing output cheap and interpreting it scarce.

The same logic applies to a small operator. Independent properties are reporting measurable gains from scheduling and upselling automation, which means the technology is no longer a large-chain advantage. The lasting advantage is knowing which line in the statement to interrogate when a number moves.