Travel AI Chatbots Are More Efficient, Not More Trusted — And the Gap Is the Business Problem

Mario Gavira, chief marketing officer of Travelier, told a closing panel at Phocuswright Europe 2026 that AI is delivering real operational efficiency and cost reduction across travel and hospitality — but has not yet produced a consumer-facing breakthrough on the scale of email, mobile apps, or Google Maps. His specific point: despite two years of hype, conversational AI chatbots have not become the dominant travel search and booking channel some predicted, because “chatbot usage has not translated into meaningful conversions.” Trust, panelists said, was the conference’s second-biggest theme after AI itself — often reduced to a simpler question of whether the technology just works reliably every time.

Timothy O’Neil-Dunne of T2Impact added a structural counterpoint aimed at AI-native travel startups: incumbent platforms still hold the advantage in customer data, service infrastructure, and competitive positioning — advantages new entrants haven’t found a way around yet. The panel also flagged European AI regulatory requirements and workforce-composition effects as open questions, without resolving either.

For operators, the practical implication is that deploying an AI booking assistant and improving conversion are two different projects, and only one of them is mostly a software purchase. A useful, if non-hospitality, parallel comes from Salesforce’s July 16 post on marketing data trust: it argues that “access to AI is no longer the differentiator among enterprises — confidence and governance are,” and that Salesforce has positioned trust ahead of innovation and customer success as a stated company value. That’s a marketing-data-governance argument, not a travel one, but it names the same mechanism the Phocuswright panel was circling — that guests and buyers are increasingly asking what happens to their data once it touches an AI system, and that answering that question is the slower, harder half of the deployment, not the model integration itself. For hospitality specifically, that suggests the conversion gap won’t close by tuning the chatbot; it closes by making the AI’s data handling and reliability visible to the guest.