GENERATIVE AI IN TOURISM: OPPORTUNITIES, RISKS, AND IMPLICATIONS FOR AFRICAN AND SUB-SAHARAN TOURISM DESTINATIONS

Abstract

Generative artificial intelligence (GenAI) is changing how tourists search for information, construct itineraries, evaluate destinations, and interact with tourism businesses. Recent tourism research has expanded rapidly, but the evidence base remains uneven across regions and applications. This literature review synthesizes recent scholarship on the opportunities and risks of GenAI in tourism and develops a specific discussion of implications for African and Sub-Saharan destinations. A structured narrative review approach was used, with targeted searches of academic and publisher databases and supplementary searches of authoritative industry, policy, and preprint sources. Literature was organized thematically around four opportunity areas: personalization, conversational travel planning, destination marketing and content generation, and operational efficiency; and four risk areas: algorithmic concentration of destination attention, trust and information-quality problems, ethical and environmental concerns, and dependence on digital infrastructure and data quality. The review finds that GenAI can lower information-search costs, support personalized travel planning, expand the capacity of destination marketers to create content, and improve service efficiency. At the same time, recent evidence indicates that AI recommendations can reproduce established tourism imaginaries, while inaccuracies, privacy concerns, limited transparency, and overreliance may weaken trust. For Africa and Sub-Saharan Africa, these issues intersect with uneven digital infrastructure, skills, organizational capacity, and data availability. The review therefore argues that GenAI should be understood not only as a tourism-service innovation but also as a potential mechanism of destination visibility and exclusion. Rather than claiming that African destinations are already systematically under-recommended, the review identifies algorithmic under-representation as an empirically testable risk. Four research priorities are proposed: auditing AI representations of African destinations, testing the relationship between AI visibility and tourism demand, examining GenAI adoption among African tourism SMEs, and developing responsible and locally grounded AI practices for destination marketing.