This morning I read a panel on Hospitalitynet.
Fourteen experts. One question.
Will AI Search benefit the direct online channel or the OTAs?
Read it yourself: https://www.hospitalitynet.org/panel/digital-marketing-in-hospitality/will-ai-search-benefit-the-direct-online-channel-or-the-otas
Fourteen people. Serious credentials. Real answers.
Good question. Missing floor.
Max Starkov, one of the most respected voices in hospitality digital strategy, told the panel to invest in AEO and GEO. Get your rates into AI platforms.
Nobody in the discussion questioned what AI already believes about the properties in question.
Simone Puorto noted that OTAs are often the source layer feeding these AI systems.
Nobody followed that thread.
AEO gets you found.
GEO gets you mentioned.
Neither one fixes the OTA description that AI returns as an answer.
Both assume the thing being found is right. That assumption never got checked.
The question assumed AI already knows the right things about your property. That the knowledge is accurate. That it reflects your positioning. That when a traveler asks an AI system to recommend a luxury resort, the version of your property that comes back is yours.
Nobody asked whether that was true.
AI did not learn about your property only from you. It learned from everything it could find. OTA listings. Aggregator copy. Review summaries. Travel blog coverage. Generic hospitality content written by someone optimizing for clicks, not for your brand.
Your language may be in there. Your positioning may be in there. But it is sitting beside OTA copy, aggregator descriptions, and syndicated content written by people with no responsibility for your brand.
That is not control. That is contamination.
So you invest in AEO. You build your GEO content strategy. You get found. You get mentioned.
AI recommends your property in someone else’s words.
There is a discipline for fixing the formation layer before the retrieval competition begins.
Knowledge Formation Optimization.
KFO.
It is not a companion to AEO and GEO. It is what should have come first.
The premise underneath it was not examined.
Fourteen experts answered what they were asked.
Nobody asked what AI had already decided before the question was written.
Nobody asked if it was right.
They optimized the retrieval.
They called it a strategy.
