Journal of Sustainable Marketing

Sustainable Tourism: Context and Relevance

Sustainable tourism can be defined as tourism that seeks to balance the needs of tourists, the environment, and host communities, ensuring that tourism activities benefit present and future generations (Streimikiene et al., 2020). It is centred on three core pillars: environmental protection, social equity, and economic development (Bramwell & Lane, 1993). While earlier sustainable tourism research often relied on the Triple Bottom Line (Elkington, 1997), or the Pyramid of Corporate Social Responsibility (Carroll, 2021) constructs, more recent approaches have evolved to integrate frameworks such as Environmental, Social, and Governance (ESG) (e.g., Mishra & Pandey, 2025; Putzer & Posza, 2024) and the United Nations’ Sustainable Development Goals (SDGs) (e.g., Fauzi, 2023; Nunkoo et al., 2021). According to the literature, sustainable tourism aims to minimize the negative environmental and social impacts of tourism while fostering long-term benefits for local communities (Hunter, 2002). At the same time, the tourism sector plays a critical role in achieving the SDGs, particularly Goal 8 (Decent Work and Economic Growth), Goal 12 (Responsible Consumption and Production), and Goal 13 (Climate Action), as reported by the United Nations World Tourism Organization (United Nations World Tourism Organization, 2018). These goals align with the principles of sustainable tourism by promoting resource efficiency, reducing environmental impact, and encouraging responsible consumption and production patterns in tourism activities. Tourism is often linked to overtourism, which leads to overcrowding and environmental degradation and raises serious concerns for communities (Damnjanovic, 2021). This makes it even more crucial to integrate sustainable practices within the industry to ensure that tourism continues to be a driver of social and economic development without compromising environmental sustainability and sustainable tourism equity (Arya et al., 2024). While several tourism groups and enterprises are doing so, the fact is that much more could be done. According to the literature (e.g., Budeanu et al., 2016), sustainable tourism offers opportunities to enhance environmental conservation and social equity, but it also faces several challenges. The overuse of resources, pollution, and the unequal sharing of tourism's benefits can hold back meaningful progress. To address these challenges, tourism policies must adopt a different approach, such as sustainable frameworks that focus on efficient resource management and community involvement (e.g., Shoeb-Ur-Rahman et al., 2019). At the same time, technological innovations such as AI offer significant opportunities to support the transformation of tourism towards sustainability by improving efficiency and enhancing environmental stewardship (Loureiro & Nascimento, 2021).

Artificial Intelligence: Definitions and Applications

AI refers to the simulation of human intelligence processes by machines, particularly computer systems (Russell & Norvig, 2021). These processes include learning (acquiring and applying knowledge), reasoning (solving problems), and self-correction. According to the authors, AI is not new and is an interdisciplinary field encompassing subfields such as machine learning, natural language processing, and neural networks, which equip machines with the ability to handle tasks traditionally associated with human cognition. AI has already demonstrated its revolutionary potential across various industries, and tourism is no exception (e.g., Gretzel et al., 2015; Suanpang & Pothipassa, 2024). The social and economic impacts of intelligent automation on the tourism sector, local communities, and the broader economy become significant as conventional hospitality and tourism tasks are increasingly performed by intelligent machines (Tong et al., 2022).
AI applications in tourism are diverse and have already begun to reshape the industry in various ways, including customer experience, operational efficiency, and resource management (Majid et al., 2023). When it comes to customer experience, AI-driven tools such as chatbots and virtual assistants are transforming customer service by offering personalized recommendations tailored to travelers' preferences and past behaviors (e.g., Gretzel et al., 2015; Samala et al., 2020). Additionally, AI is being used to suggest destinations, accommodations, and activities tailored to individual tastes. Regarding operational efficiency, AI-driven tools help businesses optimize operations, from dynamic pricing models that adjust rates in real time to resource management systems that improve logistics (Cristian & Tileagă, 2024). In the context of sustainable resource management, AI systems can analyze data to optimize energy consumption, manage waste, and reduce carbon footprints (e.g., Patrichi, 2024). For example, AI can predict tourist arrivals, allowing for better management of transportation, accommodations, and tourist sites to reduce overcrowding and minimize environmental degradation.
Despite its revolutionary potential, implementing AI in tourism presents several challenges. One key issue is the high initial cost of adopting AI technologies, which may be a barrier for smaller businesses (Wang et al., 2025). Additionally, according to the same authors, AI systems require large datasets to function effectively, raising concerns about data privacy and security. The collection and use of personal data for AI-driven applications raise critical privacy concerns, necessitating stringent safeguards to ensure responsible, ethical practices (Olawade et al., 2024). There are several concerns about the social implications of AI, including job displacement and inequality in access to AI-driven solutions (e.g., Subaveerapandiyan & Shimray, 2024). From an environmental perspective, energy consumption poses a significant challenge for the deployment of AI systems, as the high computational demands of advanced models impose substantial energy requirements on data centers and AI infrastructure (Olawade et al., 2024).

Intersection of AI and Sustainable Tourism Development

AI plays a key role in advancing sustainable tourism by helping to use resources more efficiently, reduce waste, and improve the overall performance of tourism operations. At the same time, AI can analyze large amounts of data, enabling it to predict patterns and trends, helping businesses and policymakers make data-driven decisions aligned with sustainability objectives. Several examples illustrate how AI is already being used to promote sustainability in tourism, such as tourism forecasting, sustainable transport solutions, and smart destinations (e.g., Majid et al., 2023). Tourism forecasting is highly important, and AI-driven systems are being used to predict visitor numbers to popular destinations, enabling better crowd management and reducing the environmental strain caused by overtourism (Shafiee, 2024). At the same time, this can have an impact on smart destinations, since in smart cities and destinations, AI is being applied to monitor environmental factors, such as air and water quality, and to manage waste. Additionally, AI can benefit inclusive tourism by providing a personalized, tailored experience that accounts for individual preferences and disabilities (e.g., Suanpang & Pothipassa, 2024).
In another stream, governments and tourism authorities can leverage AI to create and implement more effective policies for sustainable tourism (Afaq et al., 2024). By incorporating AI into the policymaking process, they can improve the targeting of sustainability measures, monitor the success of sustainability initiatives, and adapt strategies based on real-time data, with a real impact on overcoming overtourism, for instance (Siddik et al., 2025). AI-powered initiatives, such as chatbots, can also help monitor tourist behavior, ensuring that sustainability initiatives are followed and identifying areas where additional interventions may be necessary (Majid et al., 2024).
Despite its revolutionary potential to create a more efficient, responsible, and sustainable tourism sector, the intersection of AI and sustainable tourism also presents a challenging academic avenue, given the rapid expansion of research in this field.

Selection of Bibliometric Databases

To understand the interconnectivity among the themes of AI, tourism, and sustainability, we searched the Scopus and Web of Science (WoS) databases for relevant literature. Scopus and WoS are the two most widely used databases for bibliometric analysis (Verma et al., 2021; Blanco-Moreno et al., 2023; Yong-Hak, 2013), Scopus provides broader coverage, advanced search filters, and data analysis tools to improve data management. On the other hand, the WoS database is recognized as a reputable source for academic publishing and citation records, offering extensive access to globally renowned research (Birkle et al., 2020).

Search Strategy

The underlying premise of the applied methodology is grounded in the principle that keywords in scientific literature serve as direct indicators of the primary research topics and, consequently, reflect the nature of the data used in the study (Ordóñez-Martínez et al., 2024). In general, keywords are selected by authors to accurately reflect the study's central themes. All research paper titles, abstracts, or keywords containing the search terms were evaluated.

The initial search string included keywords such as ”artificial intelligence”, “sustainability”, and ”tourism”. Synonyms for artificial intelligence (AI) and Smart tourism are used with Boolean operators like “OR” to retrieve the complete set of papers. TheBoolean operator ”AND” is used to obtain the intersection of papers that cover topics related to artificial intelligence, sustainability, and tourism (Table 1).
During the document scanning process, new keywords such as ”smart tourism” and ”ecotourism” (2nd phase) and “green tourism” and “regenerative tourism” (3rd phase) were incorporated into the search strategy. Smart tourism focuses on integrating technology to enhance tourism experiences and management (Buhalis & Amaranggana, 2013), while ecotourism, green tourism, and regenerative tourism are primarily associated with the ecological and sustainable aspects of tourism (Pavlidis et al., 2022). These keywords were selected for their conceptual relevance in the sustainable tourism literature, thereby ensuring a more comprehensive coverage of the field without altering the original scope of the analysis.
Table 2 presents the annual distribution of publications on the topic of analysis with records identified through database searching. The data reveal a significant increase in the number of articles published, particularly from 2020, indicating a growing academic and practical interest in this emerging interdisciplinary field.

A recent systematic review by Majid et al. (2023) showed a notable increase in scholarly attention to intelligent automation in sustainable tourism, with a significant surge starting in 2015. The rapid advancement in AI capabilities has garnered increased scholarly attention and interest in the last decade. Other literature reviews have also analyzed papers published since 2015 and observed a significant increase in articles focused on smart tourism (Liu et al., 2023) or digital tourism research (Luo et al., 2022).

Inclusion and Exclusion Criteria

Inclusion and exclusion criteria were applied to the search results, which helped refine the selection process and extract the most relevant articles for the literature review. Criteria for including and excluding literature were established based on relevance, publication date, and quality, using the eligibility and rejection parameters outlined in Table 3. By applying these criteria, we ensured that only relevant, credible, and current sources were included.

The review includes literature published from 2015 to March 2024 and was limited to articles written in English to ensure consistency in language. The literature search generated 408 records. The bibliometric study was complemented by a thorough peer review of the selected scientific articles, identifying their application areas, methodologies, and the data/information used in their development. We eliminated duplicate and non-topic-related articles. For example, some articles referred to AI as 'Assessment Indicator' or 'Appreciative Inquiry,' terms that are unrelated to AI. After removing duplicate references and applying selection criteria, this search approach produced a final set of 76 records, the full texts of which were retrieved and evaluated for inclusion in the final analysis.

Data Analysis Plan

VOSviewer was selected for this study to perform the bibliometric analysis and visualize bibliometric networks. This open-source tool is widely recognized among scholars for visualizing literature maps, providing graphical representations of relationships and patterns between studies, such as keywords, co-authorship, and citation networks (Figueroa-Domecq et al., 2015).
We employed four different bibliometric analyses: co-authorship, citation, co-citation, and co-word. Co-authorship analysis allows us to identify the research groups actively contributing to the field of study (Ordóñez-Martínez et al., 2024). Citation analysis is a bibliometric approach used to explore the quantitative patterns and relationships among academic publications (Shiau et al., 2023). This study explores citation patterns across academic literature to identify the most highly cited and impactful papers. Co-citation analysis measures how frequently two articles are cited together within a collection of documents (Liu et al., 2023). Lastly, co-word analysis evaluates the frequency and patterns of keyword co-occurrence in a collection of documents (Liu et al., 2023). A co-word network or map can help identify clusters of frequently co-occurring words that represent particular topics or themes in a research field (Wider et al., 2023). By using this method, we can map connections among publications and gain insights into foundational works, dominant themes, ongoing trends, and emerging directions in sustainable AI tourism.

Co-authorship Analysis

Table 4 presents the top 10 most relevant authors based on the total number of citations in the database. The author with the highest number of citations (12,499) is Balsalobre-Lorente, Daniel, who also has the highest G-index (119) and H-index (59). The most contributed topics from 2019 to 2023 are Greenhouse Gas Emissions, Carbon Dioxide, and the Environmental Kuznets Curve. The author with the oldest publication is Ronan De Kervenoael (2020).

Countries' affiliations were assessed using the Environmental Performance Index Block20242024. The EPI evaluates 180 countries using 58 performance indicators across 11 issue categories, encompassing climate change performance, environmental health, and ecosystem vitality Block20242024. Table 5 shows that the majority of the most relevant countries in the database score above 50, indicating they are leaders in environmental performance. Despite these challenges, these nations are studying AI's role in sustainable tourism, showing their aspiration to transition towards a more sustainable future. Additionally, the countries' Gross Domestic Products (GDPs) were assessed, as they are among the most well-known indicators of a country's level of development. Vietnam and India are the countries with the lowest GDP and are also classified as laggards in the EPI, indicating that less developed countries tend to exhibit lower levels of concern for sustainability. Finally, the United Kingdom holds the highest EPI score (72.6) and has one of the longest coastlines (12,429 km), which may reflect a strong concern for plastic pollution in oceans. The United Kingdom also has one of the highest GDP per capita (48,866.6 US$), suggesting that more developed countries tend to demonstrate greater environmental concern.

Table 6 presents the 10 most cited articles from the 10 organizations included in the database. Marketing, Management, and Economics are the primary fields specializing in studying the impact of AI on sustainable tourism development. However, there is a growing interest in emerging areas such as Finance, Business Administration, and Computational Science and Technology.

Citation and Co-citation Analysis

Table 7 presents the three most relevant sources based on the maximum number of articles published in different journals. The most cited sources in the field are published in Tourism Management, Sustainability, and Resources Policy. Among these, sustainability stands out as the journal with the highest number of documents, totaling 17, while the majority of other sources have contributed only one document each. Tourism Management is a leading academic journal that explores the fields of travel and tourism, with particular emphasis on management, planning, and policy. Sustainability is an internationally recognized, peer-reviewed, open-access journal that publishes research covering the environmental, cultural, economic, and social aspects of human sustainability. Resources Policy is a global journal dedicated to the economic and policy dimensions of mineral and fossil fuel extraction, production, and use. Additionally, the most relevant journals exhibit strong performance across bibliometric indicators such as the H-index and G-index. This underscores the overall influence and quality of the leading sources within this field.

Figure 2 presents the co-citation network analysis of authors. Five main clusters were identified based on citation counts and strong relationships among articles.

1. In cluster one (red), the authors primarily focused on using machine learning and deep learning techniques to collect objective, multi-perspective data. They also integrated these within the Green Internet of Things to align ecological impact assessment with ecotourism development. This includes authors such as Rashid Mehmood, Ana Ramón-Rodríguez, Hui Li, and Jingyu Li. The authors applied qualitative studies by developing models and frameworks such as the Framework of Long-term Competitiveness (Moreno-Izquierdo et al., 2018), the Framework for Smarter Sustainable Tourism (Alsahafi et al., 2023), and the Spatial Spillover Effect Model (Gong et al., 2023).

2. In cluster two (green), authors explored smart tourism and the impact of AI, with a particular focus on eco-hospitality. This includes authors such as Dimitrios Buhalis, Pasquale del Vecchio, and Ioannis Vardopoulos. The authors aimed to shed new light on smart tourism by developing a conceptual framework (Buhalis et al., 2024) and a circular economy business model (del Vecchio et al., 2021). Additionally, the authors contributed to a deeper understanding of economic and developmental issues in a tourism-specialized context by developing a SWOT (Strengths, Weaknesses, Opportunities, and Threats) matrix within the smart city framework (Vardopoulos et al., 2023).

3. Cluster three (blue) discusses human-robot interaction for innovative sustainable tourism and hospitality. This includes authors such as Iis Tussyadiah, Edwin Goh, and Subhankar Das. Adopting an empirical approach, the authors analyzed various variables and their relationships, such as empathy from service robots, perceived value, tangibility of the service, and the intention to use robots. Research shows that AI-powered human-machine interaction devices significantly improve service quality, boosting tourist satisfaction and loyalty (Van et al., 2020).

4. In cluster four (orange), the authors focused on the transport field within the tourism sector, specifically public transport and space tourism. This includes authors like C. Michael Hall and Myung Ja Kim. The authors applied the Value-Attitude-Behavior theory and developed an integrated research model to examine relationships among variables related to AI (e.g., AI awareness) and sustainability (e.g., sustainable mobility). The studies showed how important values and awareness of AI are in shaping people's attitudes and both social and personal norms, which in turn encourage sustainable behaviors (Kim et al., 2024).

5. In cluster five (purple), the authors discuss sustainable tourism development in rural or urbanized areas with AI methods. This includes authors like Jaffar Abbas, Yue Sun, and Hongmei Yang. Some models included in the studies are a tourism-transportation-based Spatial Durbin Model (Gong et al., 2023) and a tourism-environment framework (Balsalobre-Lorente et al., 2023). Some of the findings indicate that urbanization, natural resources, and tourism contribute positively to CO2 emissions, while information and communication technology reduces emissions (Balsalobre-Lorente et al., 2023).

The majority of the articles were published in 2023 and primarily use an exploratory approach, suggesting that the topic is still emerging and remains underexplored empirically. Clusters three and four were the most empirically oriented, emphasizing data-driven analyses and the examination of measurable variables. This focus reflects a strong commitment to anchoring theoretical models in observable phenomena.

Keywords Cluster Analysis

Table 8 shows the top 10 keywords that co-occur most frequently. Artificial intelligence, Tourism, and Sustainable Development were the most relevant keywords, with 23 and 16 occurrences respectively, and a total link strength exceeding 200. The relevance of the keyword ”ecotourism” is also evident, with a total link strength of 153, a term that was incorporated during the second phase of the search strategy. Figure 3 shows how the keywords from the selected articles are linked based on co-citations. The grouping of keywords by subject matter gives rise to the following clusters:
In cluster one (purple), dedicated to AI, terms such as “artificial intelligence”, “deep learning”, “economic development”, “tourism economics” and “governance approach” are highlighted, underscoring the importance of artificial intelligence in driving economic development in tourism and the role of governance in promoting tourism sustainability (Tong et al., 2022; Kim et al., 2023).
Cluster two (light blue) focuses on Tourism, Sustainability, and Technology. This cluster addresses sustainability and technology within the tourism sector, specifically the hospitality and aviation industries, as seen by the terms “tourism”, “hospitality”, “aviation”, “hotel industry”, and “sustainability”. The inclusion of terms such as ”technology,” ”digital transformation,” and ”artificial intelligence” underscores the articles' strong technological focus and explores various domains within AI. Kashem et al. (2023) highlighted the potential of AI and blockchain to balance economy-driven tourism with sustainability-oriented, tech-enhanced tourism.
Cluster three (grey) addresses machine learning in the sustainable development of the tourism industry. Terms like ”smart tourism”, “destination”, “ice-snow tourism” and “cross-border tourism” highlight the different types of tourism analyzed, as do the case studies of specific regions, as mentioned, such as ”Adriatic-Ionian tourism”. The Adriatic-Ionian area presents a clear vocation for tourism, with distinctive cultural heritage, and has been identified as a priority for developing innovative strategies to enhance competitiveness (Ndou et al., 2023).
Cluster four (red) focuses on ecotourism, sustainable tourism, and virtual reality. The presence of terms such as ”artificial neural networks”, ”heritage sites”, and ”cultural heritages” suggests the use of neural networks as a methodological tool, with a focus on UNESCO World Heritage Sites. By leveraging AI capabilities, Artificial Neural Networks can assist in the classification, preservation, and sustainable development of UNESCO World Heritage Sites (Bozkurt & Şeker, 2023).
Cluster five (dark green) addresses sustainability in tourism development using AI. Terms such as “greenhouse gases”, “environmental impact”, and “carbon footprint” underscore the growing concern for the environmental field of sustainability in tourism, while “economic and social effects” reflect a focus on the economic and social practices of tourism companies (Zopiatis et al., 2020; Miedes-Ugarte et al., 2020).

Although ethical, regulatory, and legal issues did not emerge as dominant keywords in the bibliometric analysis, the growing integration of AI and advanced technologies in the tourism sector highlights the critical need for further research in this area. As the digital landscape evolves, ethical concerns —including fairness, transparency, accountability, and data privacy —are increasingly central to responsible technology deployment. While AI offers significant potential to enhance experiences, efficiency, and personalization within the tourism sector, its ethical implications cannot be ignored. Deploying these technologies in an ethically responsible, privacy-conscious, and socially accountable manner will be crucial for fostering a future that is both technologically advanced and ethically robust (Chon & Hao, 2024).

Theoretical and practical implications

This study contributes to both theory and practice by advancing the understanding of AI applications within the sustainable tourism sector. From a theoretical perspective, this study is particularly relevant given the increasing urgency to adopt sustainable practices within the global tourism industry. AI presents a promising pathway to achieve these goals by providing advanced solutions to sustainability challenges such as resource management, environmental monitoring, and efficiency improvement. By systematically mapping the research landscape, this paper offers a valuable contribution to current trends, collaborative networks, and research gaps. From a practical standpoint, this study provides significant contributions to industry stakeholders seeking to adopt AI technologies. It identifies practical strategies for integrating AI into sustainable tourism practices, helping tourism businesses and policymakers navigate the complexities of AI implementation. The study emphasizes the importance of collaboration among technology providers, tourism operators, and government bodies to maximize AI's impact on sustainability goals, offering actionable recommendations to facilitate the transition to more intelligent, more sustainable tourism systems.

Recommendations for Stakeholders

Building on the findings of this bibliometric review, this section outlines tailored recommendations, each directly informed by the gaps and patterns identified in the analysis. This aims to support stakeholders in addressing the complexities of AI implementation and to contribute to the development of a more sustainable and intelligent tourism ecosystem:
Tourism Industry Professionals: The analysis revealed a scarcity of empirical studies on AI implementation, with existing research remaining largely conceptual. Given the lack of empirical studies on AI implementation, practitioners should prioritize pilot projects and applied initiatives that test AI-driven solutions for enhancing sustainability in tourism operations and infrastructure. This includes experimenting with intelligent automation to reduce resource consumption, adapting physical facilities to evolving environmental and social requirements, and implementing AI-assisted visitor management systems.
Policy Makers: The bibliometric clusters indicated a weak presence of governance and regulation themes in the current literature. Given the literature's limited focus on governance aspects, policy frameworks should incentivize the evidence-based adoption of AI while integrating environmental, social, and governance (ESG) principles. In addition, support for cross-sector collaboration and targeted funding for empirical research can help close the current gap between theory and practice.
Technology Providers: Our keyword analysis highlighted limited attention to sector-specific AI applications. Given the scarcity of sector-specific AI solutions, developers should focus on designing tools tailored to the tourism industry’s sustainability challenges. Emphasis should be placed on scalable applications that improve resource efficiency, foster social responsibility, and align with ethical standards. Academia and Researchers: The co-citation analysis demonstrated a concentration of publications in a single outlet and a predominance of exploratory approaches. The results clearly underscore the need for more empirical and applied research. Scholars should investigate how AI concretely contributes to sustainable practices, including the adaptation of tourism infrastructure, governance mechanisms, and consumer behavior, while also diversifying publication outlets to reduce the current concentration in a single journal.
Consumers: While consumer perspectives were underrepresented in the dataset, consumers can play a critical role by supporting tourism providers that adopt AI-based sustainable practices, thereby shaping demand for responsible and environmentally conscious travel.
Finally, AI implementation in sustainable tourism should follow ethical principles, ensuring benefits for all stakeholders without increasing inequalities or environmental harm. Concerns about data privacy, algorithmic bias, and equitable access to technology are pervasive throughout the literature and remain central to sustainable adoption. Addressing these issues is crucial to ensuring that AI-driven tourism benefits all stakeholders without exacerbating inequalities, environmental harm, and social exclusion.

Limitations

This study has some limitations, particularly concerning its focus and methodology. Firstly, it heavily depends on the databases used, such as Web of Science and Scopus, which may not fully represent non-English publications or content from smaller publishers. Secondly, it is crucial to acknowledge that keywords may not fully capture the methodological aspects of the analyzed articles. Keywords are often general and broad, potentially overlooking unconventional research areas. Furthermore, as the field evolves, keywords may change over time, influencing the interpretation of trends and data. Finally, this study focuses on articles published from 2015 to March 2024, potentially excluding earlier or very recent works, thereby limiting the scope of the analysis. By limiting the analysis to open-access publications, the sample is composed mainly of articles from Sustainability (MDPI), which may affect the overall representativeness of the findings. To address the study's limitations, several strategies were employed. First, Web of Science and Scopus were chosen as the primary databases due to their reputation as the most commonly used sources for bibliometric analysis, providing extensive access to globally recognized research. Keywords were carefully selected in three phases, with additional terms added after reviewing the articles, minimizing the risk of excluding relevant studies. Finally, the time frame (2015–2024) was chosen to maintain a forward-looking perspective, while acknowledging the influence of earlier research through citation analysis, ensuring that current trends and developments were captured. These measures helped provide a comprehensive and accurate analysis of the literature.