Before the booking: AI-mediated discovery, demand allocation, and destination competitiveness in tourism markets

AI-mediated search systems are introducing a new intermediation layer at the discovery stage of travel planning by generating bounded recommendation sets in response to natural language queries. Unlike traditional search engines, which present ranked lists for user-driven exploration, these systems produce pre-constructed choice sets that may partially substitute for travelers’ own search processes. This paper argues that this shift raises underexamined economic questions concerning consumer welfare, pricing transparency, market access, vertical distribution structure, and the allocation of tourism demand across destinations. We advance the central proposition that AI-mediated discovery functions as a demand-allocation mechanism, shaping where visitor demand is directed, how concentrated it becomes, and how much spending is retained within host economies. We also introduce constraint amplification, whereby more specific traveler queries may produce narrower and less price-diverse outputs. The paper develops a conceptual framework for these dynamics and outlines a research agenda for tourism economics.