Are all Generative AI Chatbots the Same? Analysing the Reliability of Hotel Recommendations

Are all Generative AI Chatbots the Same? Analysing the Reliability of Hotel Recommendations | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article Are all Generative AI Chatbots the Same? Analysing the Reliability of Hotel Recommendations Julia Marti-Ochoa, Eva Martin-Fuentes, Berta Ferrer-Rosell, Juho Pesonen This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-7687581/v1 This work is licensed under a CC BY 4.0 License Status: Under Review Version 1 posted 9 You are reading this latest preprint version Abstract The rise of generative AI tools like ChatGPT, Claude, Gemini, DeepSeek and Grok is transforming the way users interact with digital information, particularly in the global hospitality industry. This study evaluates hotel recommendations generated by these AI chatbots across the top 10 most visited cities worldwide. A comprehensive comparative analysis is conducted to evaluate whether these tools provide reliable and unbiased suggestions by comparing their outputs with verified real hotel data, including price, hotel category, and scores from Booking.com and TripAdvisor. The findings reveal a significant difference between AI-generated data and actual real-world values, especially in pricing. ChatGPT consistently recommends higher-category hotels but often underestimates scores and prices. Gemini achieves the closest alignment with star ratings. DeepSeek and Grok present increasingly promising multimodal capabilities. The study highlights the potential and current limitations of AI-driven hotel recommendations, offering strategic insights for hospitality businesses that are adapting to rapidly changing AI-driven search behaviour. Artificial Intelligence (AI) Hospitality Prompt large language models (LLM) Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 1. Introduction The main reason generative AI (GAI) chatbots such as ChatGPT’s comprehensive recommendations have a positive impact on users is that they perceive them as being accurate and reliable (Kim et al., 2023 ). GAI can provide users with quick, reliable, and interactive responses (Mladenović et al., 2024 ), but these responses also have to be accurate and reliable. GAI chatbots have a tendency to sometimes provide incorrect information, which can directly affect the user experience and decision-making (Kim et al., 2025 ). These systems also have inherent biases that can affect the recommendation results and user behaviour (Deldjoo, 2024 ). Thus, the reliability of generative AI chatbots has become an important research avenue to improve outputs and the user experience. Current research on the reliability of generative AI, particularly large language models (LLMs), primarily revolves around improving their consistency (Kim et al., 2023 ), accuracy (Bareh, 2025 ), trustworthiness (Lu & Feng, 2024 ), and robustness (Masciari et al., 2024 ) in diverse applications, as well as developing robust methodologies for their evaluation (Illicki, 2022). Researchers have especially noted that feedback generated by AI chatbots can be inaccurate and vary significantly from real-world values (Liebenow et al., 2025 ). Thus, there have been calls for greater understanding of AI accuracy and reliability to build trustworthy AI systems (Petkovic, 2023 ). In recent years, consumers have increasingly turned to AI-based tools to assist them in decision-making processes, especially in complex and information-rich domains such as travel and hospitality (Bhaskar & Sharma, 2022 ). Instead of relying exclusively on traditional search engines or static reviews (Wong, Lian, & Sun, 2023 ), users now interact with conversational platforms that provide direct and personalized responses to queries such as ‘Where should I stay in Tokyo for under $ 150?’, or ‘What is the best hotel near the Eiffel Tower with a terrace view?’. This shift reflects a broader transformation in how people access, filter, and act upon digital content (Bulchand-Gidumal, 2022 ; Popesku, 2019 ). AI systems based on LLMs are particularly effective in this context because they go beyond simple keyword matching and incorporate contextual understanding, tone, and even intent (Vashishth, Sharma, & Sharma, 2024). This capability enables them to emulate the role of a human travel advisor by dynamically interpreting user preferences and offering recommendations tailored to such constraints as budget, travel style, or timing (Pillai & Sivathanu, 2020 ; Buhalis & Moldavska, 2022 ). As a result, these systems have the potential to transform traditional pathways for searching for tourism information, affecting not only user experiences but also the visibility and competitiveness of businesses in the sector (Lee, Kwon, & Back, 2021 ; Štilić, Nicić, & Puška, 2023 ). Considering the above, the aim of this study is to gain a deeper understanding of how reliable and accurate different GAI chatbots are in the hospitality sector, particularly in generating hotel recommendations for users. Whereas most of the earlier research has focused mainly on ChatGPT, currently the most popular generative AI chatbot (Bailyn, 2025 ), we examine and compare the other most popular platforms and their performance. Thus, our research questions are as follow: What kind of differences exist between platforms in hotel recommendations provided by ChatGPT, Claude, Gemini, Grok, and DeepSeek for the world’s top 10 tourist destinations? What factors most define the recommendations of each GAI model (e.g., hotel category, average price, Booking.com and Tripadvisor scores)? How reliable are hotel recommendations by GAI chatbots compared to real-world data from Booking.com and TripAdvisor? Is there any bias in the recommendations of these chatbots towards hotels with specific characteristics, such as higher star ratings or prices? 2. Literature review 2.1 Generative AI Chatbots Generative AI chatbots based on LLMs, like ChatGPT, Claude and Gemini, for instance, have transformed the way we access and interact with information (Ray, 2024 ). LLMs are designed to process vast amounts of textual data, enabling them to generate coherent responses, analyse complex contexts, and even simulate conversational understanding. Researchers have often adopted a comparative approach to GAI chatbots to understand how accurate and reliable they are. There are various benchmarks that implement accuracy metrics to analyse the performance of GAI chatbots or compare their recommendations with traditional recommendation systems (Di Palma et al., 2023 ). Comparisons are also important to understand and optimize performance. Comparing different chatbots can provide insights into how AI is inclined towards certain underlying recommendation approaches (Di Palma et al., 2023 ). Currently, there are many different GAI chatbots such as ChatGPT, Gemini, Claude, DeepSeek, and Grok. While ChatGPT has become the most popular platform, it has its limitations. It is possible that different platforms can excel in different kinds of recommendations, as all chatbots are fed with different training material and have different algorithms. However, these kinds of comparisons are rarely conducted, especially in tourism and hospitality. ChatGPT, OpenAI’s leading LLM, has been consistently improved since the launch of ChatGPT-2 in 2019. Each version has advanced, particularly in language handling and complex interactions. ChatGPT-4o now includes real-time internet search capabilities (OpenAI, 2024 ). This marks a significant advancement, as it is no longer limited to data up to 2021; it can now access links from websites, blogs, newspapers, and more. OpenAI emphasizes that this feature allows ChatGPT to provide live updates on the weather, stock prices, sports schedules, news, and other timely topics (OpenAI, 2024 ). This advancement could benefit fields such as education, research, tourism, and more, offering users a richer and more personalized experience. Leading tourism scholars and industry professionals are analysing the potential impacts of ChatGPT on areas such as customer service, travel planning, and destination marketing within the tourism and hospitality sectors (Sigala et al., 2024 ). Gemini, meanwhile, developed by Google DeepMind, represents a significant evolution in the field of LLMs, integrating advanced capabilities in natural language understanding and contextual processing (Labrague, 2024 ). Designed as a direct competitor to ChatGPT, Gemini offers unique functionalities such as high-precision text generation, real-time analysis, and seamless integration with other tools in Google’s ecosystem, including Google Search and Workspace (Imran & Almusharraf, 2024 ). In the tourism sector, Gemini holds the potential to redefine personalized recommendations, leveraging its ability to interpret large volumes of historical and real-time data (Viglia et al., 2024 ). By utilizing geolocated searches, Gemini can suggest accommodation options that align user preferences with contextual factors such as availability, weather, and user reviews (Sigala et al., 2024 ). The increasing integration of generative AI tools in hospitality has already been seen to have a transformative impact on customer services, highlighting both their promise and the challenges of their implementation (Dwivedi et al., 2024 ). Claude, developed by Anthropic, introduces an innovative approach to LLM design by prioritizing ethics and safety in human-AI interactions (Priyanshu et al., 2024 ). Unlike other chatbots, Claude is designed to be more interpretative and less prone to generating inaccurate or harmful responses, making it a reliable tool for sectors like hospitality (Chester Business School, University of Chester, United Kingdom & Leong, 2024 ; Ilieva et al., 2024 ). This model is optimized for deep contextual understanding and delivers responses in a natural and empathetic manner (Wu et al., 2023 ). DeepSeek, co-founded by the Chinese hedge fund High-Flyer, represents a new frontier in large multimodal models (LMMs), designed for real-world applications by combining advanced language and vision capabilities. A key feature of DeepSeek is its rigorous training strategy, which emphasizes the balance between vision and language modalities, ensuring that both are optimized without compromising LLM capabilities (Wu et al., 2024 ). Additionally, the model employs instruction-tuning datasets and a use-case taxonomy, enabling robust performance across diverse tasks, including text and image-based contexts (Lu et al., 2024 ). Another AI tool created by xAI, Grok, stands out for its seamless integration with X (previously known as Twitter) and its emphasis on promoting freedom of speech (Edson de Carvalho Souza, 2025). Grok 3 represents a significant leap forward in the evolution of AI chatbots, incorporating advanced reasoning capabilities and a minimal censorship approach in its responses. Additionally, its integration with DeepSearch, an AI-based search engine, enhances the accuracy and depth of the information generated (Fernández, 2025 ). In certain domains, Grok has been rated as the lowest-quality model with regard to medical information (Malak & Şahin, 2024 ). Ensuring the responsible adoption of AI in tourism is crucial for balancing operational efficiency with essential human interaction, ethical considerations, and service quality (Tussyadiah, 2020 ). Tools like ChatGPT, Claude AI, and Gemini AI can enhance efficiency, but their implementation must prioritize authenticity and user trust to avoid potential drawbacks (Saleh, 2025 ). To achieve this, reliability plays a key role. 2.2 Reliability of LLM Recommendations The use of Large Language Models (LLMs) for recommendation tasks is rapidly evolving, particularly in consumer-oriented industries such as tourism and hospitality. Unlike traditional recommender systems based on collaborative filtering or content-based algorithms, LLMs generate recommendations through natural language understanding and contextual analysis (Remountakis, Kotis, & Kourtzis, 2023). These capabilities allow nuanced interactions with users, offering suggestions inferred from implicit preferences and contextual cues. Recent research has explored LLM-based explainable recommender systems, highlighting their potential to integrate multi-source data such as real-time reviews, price fluctuations, and location-specific events (Said, 2025 ). For example, studies on hotel hospitality show how ChatGPT-driven recommenders can enhance upselling strategies by tailoring persuasive recommendation messages (Remountakis, Kotis, Kourtzis, & Tsekouras, 2023 ). However, concerns remain around transparency and algorithmic bias in LLM-generated recommendations. Popularity bias and opacity can lead to the disproportionate visibility of high-priced or highly reviewed properties, leading to market inequalities (Hasan et al., 2024 ; Dang & Li, 2025 ). These issues align with broader challenges in explainable AI (XAI), where a lack of interpretability reduces user trust (de Cerqueira et al., 2025 ). In tourism, trust is central to recommendation acceptance. Explainable designs, such as those implemented in ExplainableTrip, demonstrate how providing clear rationales (e.g., “this hotel matches your budget and has a 4.8 cleanliness score”) improves user engagement (Chitaliya et al., 2025 ). Without such mechanisms, LLM recommendations risk being perceived as opaque and unreliable. Ethical considerations are equally critical. Scholars argue that the responsible adoption of AI-driven recommendation systems must prioritize fairness, accountability, and transparency to prevent manipulative or biased practices (Hudders et al., 2021; Nouriinanloo, 2024). Ensuring these safeguards is key to integrating LLM recommenders effectively in tourism and hospitality. LLMs can deliver highly personalized recommendations, but they may also introduce algorithmic biases, such as popularity bias, by disproportionately favouring well-known options, regardless of suitability (Lichtenberg et al., 2024 ; Kowald, 2025 ). Similar patterns emerge in healthcare, where both accuracy and fairness are critical: despite their robust performance in clinical tasks, LLMs may propagate demographic biases, adversely affecting certain patient groups (Omar et al., 2025 ; Schmidgall et al., 2024 ). Given the limited academic exploration of such issues in tourism, this sector emerges as a crucial domain for investigating LLM accuracy and bias. 3. Data & Methodology The study focused on prompts related to hotel recommendations, as results for other aspects of hospitality and tourism varied significantly depending on the geographic location of the IP address. This was verified by the authors using a VPN to simulate different user locations across all continents. In contrast, responses to accommodation-related queries remained consistent across locations, suggesting greater stability in that content domain. To ensure neutrality, the prompt selected was: “Which hotels would you recommend in [location]?”. This query was posed to the five AI chatbots: Claude 3.5 Sonnet, ChatGPT-4o, Gemini 1.5 Flash, DeepThink 1.5, and Grok 3. In order to minimize potential bias in the responses, no user history or personalized settings were active on any of the platforms at the time of testing. The sample included the top 10 most-visited cities worldwide (Istanbul, London, Dubai, Antalya, Paris, Hong Kong, Bangkok, New York, Cancun, and Mecca), as identified from Enterat.com (Enternat, 2024 ). The question was submitted to the first three AI chatbots (ChatGPT-4o, Gemini 1.5 Flash, and Claude 3.5 Sonnet) in December 2024 and to the other two (DeepThink R1 and Grok 3) in February 2025. After receiving the initial list of hotels, a follow-up prompt was submitted to each model: “Can you create a table including the following characteristics for each hotel: star rating, the average price per night for a stay from 6 to 7 September for two people, the Booking.com score, and the TripAdvisor score?”. To standardize the analysis, the average price per night for the recommended hotels was calculated for a weekend stay, for a double room, specifically from 6 September to 7 September 2025. This timeframe was chosen to reflect typical low-season demand period, offering a consistent basis for price comparison across all cities and chatbots. Subsequently, verification was conducted manually between January and February 2025 to confirm the accuracy of the hotel information provided by the AI chatbots. This verification involved ensuring that the recommended hotels existed, matched the descriptions given by the chatbots, and provided legitimate booking options. Table 1 Number of Hotel Recommendations by AI chatbots Across the Top 10 Most-Visited Cities ChatGPT 4.o Gemini 1.5 Flash Claude 3.5 Sonnet DeepThink R1 Grok 3 ANTALYA 7 10 12 15 12 BANGKOK 10 10 12 15 12 CANCUN 7 10 12 9 12 DUBAI 9 10 12 15 12 HONG KONG 10 10 12 9 12 ISTAMBUL 7 10 12 15 12 MECCA 7 5 12 7 12 LONDON 9 10 12 15 12 NYC 10 5 12 16 12 PARIS 9 9 12 15 12 Total 85 89 120 131 120 Table 1 presents the number of hotel recommendations provided by different AI chatbots across the 10 most-visited cities worldwide. DeepThink R1 stands out as the most generous model, offering 131 recommendations, followed by Claude 3.5 Sonnet and Grok 3, each with 120 suggestions. Gemini 1.5 Flash provides 89 recommendations, while ChatGPT-4o suggests the fewest, with 85. These differences suggest that each AI model follows a different logic when generating hotel recommendations, leading to varying outputs even when provided with the same input. Figures 1 , 2 , 3 , 4 and 5 are screenshots of the output given by each of the AI tools. It is worth noting that among the five AI tools evaluated, only ChatGPT and Grok provided direct links to the external sources used to generate their recommendations, offering greater transparency regarding the origin of the information. The other chatbots did not disclose the sources underlying their outputs. Table 2 shows that Grok 3 provided more external source links (196) compared to ChatGPT-4o (125), with notable differences in categories such as news websites (Grok: 40 vs. ChatGPT: 9) and social media (Grok: 18, not present in ChatGPT). ChatGPT primarily referenced travel websites (78) and relied more greatly on established platforms, with companies mentioned 10 or more times including TripAdvisor (34), Hotels.com (22), Conde Nast Traveler (15), and Booking.com (13). In contrast, Grok’s most frequent references (10 each) were travelandleisure.com and cntraveler.com, indicating a more diversified but less platform-centric link pattern. Table 2 Categorized external sources linked by ChatGPT-4o and Grok 3 across 10 hotel recommendation queries ChatGPT-4o Grok 3 Travel website 78 76 News website 9 40 Travel magazine 20 32 Social media - 18 Lifestyle blog 12 10 Luxury travel website - 8 Luxury hotel website 4 7 Luxury travel blog 2 5 Total 125 196 To analyse the alignment between AI-generated hotel recommendations and real-world data, two complementary statistical techniques were applied. First, Pearson correlation coefficients were calculated between each pair of variables (AI-supplied and real values) (Sedgwick, 2012 ) or hotel category shown by stars from 1 to 5, average prices per night, Booking.com scores, and TripAdvisor ratings. This allowed for the identification of directional relationships and the level of agreement between AI outputs and verified data. Second, paired-sample T-tests were conducted to assess whether the mean differences between AI-generated and real values were statistically significant (Afifah et al., 2022 ). These tests provide insights into potential systematic biases introduced by each AI tool. Additionally, to compare the real values for hotels recommended by different AI chatbots, Welch ANOVA tests were conducted to detect intergroup differences, followed by Games-Howell or Bonferroni post-hoc tests, depending on variable characteristics, to identify which tools significantly differed from each other (Agbangba et al., 2024 ; Liu, 2015 ). This multi-step statistical approach ensures a robust and nuanced evaluation of the performance and biases of each generative AI chatbot. 4. Results To assess the reliability and potential biases in hotel recommendations generated by AI chatbots, we conducted a detailed statistical analysis comparing the outputs of each model with real-world hotel data. The correlation between the real variable and the AI variable for each of the items analysed (hotel category, price, score on Booking.com (BK) and score on TripAdvisor (TA)was carried out to find out the extent to which both variables of the pair were correlated. Moreover, we wished to analyse how different the means were, so a T-test was carried out for each pair of variables, within each AI tool. It could be that the correlation between the pairs is strong and significant, which would mean that both variables (AI and real) go in the same direction (when one increases the other also increases) and thus there is not much bias. It is important to note that not all AI-generated outputs contained complete information for all variables (e.g., some recommendations lacked price or review scores), resulting in slight variations in the sample size (N) across different tests. Table 3 Paired samples for the ChatGPT-4o model Pairs N Mean Std. Dev. Paired samples Corr. Paired mean diff. AI stars 81 4.58 0.69 0.073 -0.383* Real stars 81 4.96 0.19 AI price 82 609.09 393.81 0.408* -318.98* Real price 82 928.06 828.55 AI BK score 81 8.78 0.68 0.263** -0.23** Real BK score 81 9.01 0.39 AI TA score 83 4.41 0.35 0.236** -0.25* Real TA score 83 4.66 0.27 * sig. <0.001; ** sig. < 0.05; Price in € As shown in Table 3 , ChatGPT recommends hotels with lower values than they actually score, for all four variables: stars, price, and Booking.com and TripAdvisor scores. Only in the case of the Price variable is correlation statistically significant (40.8%; p-value < 0.001) meaning that when the price given by AI increases, the real price also increases. For the other variables, there is no significant correlation between what is suggested by ChatGPT and the real data. Given that the prices analysed were static for a fixed date, the observed correlation does not indicate price increases over time but rather reflects logical consistency between the AI-generated outputs and real hotel price data. Considering the mean difference, it can be observed that in all four pairs, the difference between pairs of variables is statistically significant (at p-value < 0.001 or at p-value < 0.05). Therefore, it seems there is a degree of bias between the data provided by ChatGPT and the real data. Table 4 Paired samples for the Gemini model Pairs N Mean Std. Dev. Paired samples Corr. Paired mean diff. AI stars 84 4.67 0.84 0.851* 0.012 Real stars 84 4.65 0.89 AI price 80 424.69 198.87 -0.098 -686,51** Real price 80 1111.20 2990.32 AI BK score 84 8.47 1.33 0.675* -0.24** Real BK score 84 8.71 0.92 AI TA score 84 4.29 0.52 0.364* -0.18** Real TA score 84 4.47 0.57 * sig. <0.001; ** sig. < 0.05; Price in € Gemini (Table 4 ) suggests hotels with values closer to reality, regarding stars and scores on Booking.com and TripAdvisor. However, regarding prices, Gemini recommends much lower ones (€424 vs €1,111). We also computed the correlation between what AI suggested and real data, and it can be observed that for the stars variable, there is a strong correlation (85.1%; p < 0.001). We can also find statistically significant correlations between Gemini and the reality for Booking.com and TripAdvisor scores. As already pointed out, these results mean that there is not a significant bias between real and AI-generated data. Regarding the paired mean differences, those of price, Booking.com score and TripAdvisor score are significant but at 0.05. Table 5 Paired samples for the Claude model Pairs N Mean Std. Dev. Paired samples Corr. Paired mean diff. AI stars 113 3.81 0.96 0.672* -0.566* Real stars 113 4.38 0.86 AI price 114 349.63 325.66 0.465* -211.736** Real price 114 561.37 875.47 AI BK score 111 8.57 0.53 0.564* 0.241* Real BK score 111 8.33 0.93 AI TA score 117 4.16 0.28 0.470* -0.247* Real TA score 117 4.41 0.51 * sig. <0.001; ** sig. <0.05 Price in € Claude (Table 5 ) shows a statistically significant correlation between the tool and the reality for all four variables. This means that when Claude data increases, the real data also increases. However, when considering the paired mean difference, we find that the difference in the case of stars, and the differences in the case of Booking.com and TripAdvisor scores, are significant (p-value < 0.001). Table 6 Paired samples for the Grok model Pairs N Mean Std. Dev. Paired samples Corr. Paired mean diff. AI stars 119 4.33 0.82 0.789* 0.067 Real stars 119 4.26 1.19 AI price 115 259.63 211.40 0.500* -392.770* Real price 115 652.40 692.34 AI BK score 117 8.84 0.39 0.600* 0.174 Real BK score 117 8.67 1.09 AI TA score 117 4.33 0.25 0.372* -0.188* Real TA score 117 4.52 0.47 * sig. <0.001; ** sig. <0.05 Price in € Within the Grok model, it can be seen (Table 6 ) that correlations are all statistically significant and strong, especially for hotel category. Thus, real data and data provided by the tool go in the same direction. Therefore, in this case, the difference in means for the hotel category variable and for the Booking.com score is not significant. In the case of average price per night in a double room from 6 to 7 September, Grok data supplies much lower prices than they actually are. Table 7 Paired samples for the DeepSeek model Pairs N Mean Std. Dev. Paired samples Corr. Paired mean diff. AI stars 124 4.28 0.87 0.751* 0.000 Real stars 124 4.28 1.23 AI price 125 410.89 309.79 0.483* -301.59* Real price 125 712.48 817.06 AI BK score 125 8.936 0.46 0.660* 0.218* Real BK score 125 8.718 0.659 AI TA score 128 4.315 0.29 0.395* -0.216* Real TA score 128 4.531 0.36 * sig. <0.001; ** sig. <0.05; Price in € DeepSeek (Table 7 ) also shows statistically significant correlations for all four variables. The stars variable is the strongest (lower bias between real and DeepSeek data), while that of TripAdvisor scores the weakest. Regarding the paired mean difference, it can be observed there is no difference in the case of the stars variable, but there are statistically significant mean differences for the other three variables. Table 8 Paired samples for the entire sample Pairs N Mean Std. Dev. Paired samples Corr. Paired mean diff. AI stars 521 4.30 0.89 0.693* -0.165* Real stars 521 4.46 1.02 AI price 516 397.28 314.27 0.226* -364.51* Real price 516 761.78 1396.49 AI BK score 518 8.74 0.73 0.510* 0.069** Real BK score 518 8.67 0.89 AI TA score 529 4.29 0.35 0.388* -0.218* Real TA score 529 4.51 0.46 * sig. <0.001; ** sig. <0.05; Price in € Considering the whole sample of hotels suggested by AI tools (Table 8 ), the results reveal significant correlations between AI tools and real data, but only that of the hotel category variable is the strongest, which would indicate that the bias between AI and real data is lower. Regarding the difference between pairs, that of Booking.com score is almost insignificant, and in fact, the correlation between the two variables is rather strong and significant. Having correlated and compared the results between real and AI recommendations, we now draw from the hotels suggested by AI tools but with the real (human) collected data of each variable and compare each variable across the AI tools. Table 9 Hotel category stars (actual data) descriptives by AI tool AI tool N Mean Std. Dev. Min. Max. ChatGPT 81 4.96 0.190 4 5 Gemini 84 4.65 0.898 1 5 Claude 113 4.38 0.859 1 5 Grok 119 4.26 1.196 0 5 DeepSeek 124 4.28 1.233 0 5 Total 521 4.46 1.021 0 5 Table 9 shows the descriptives for the real data variable of stars per AI tool. It can be observed that ChatGPT is the AI tool that recommends the highest (4.96) and Grok the lowest (4.26) hotel category. However, in all five tools, the mean number of hotel stars is above 4, meaning that in general, AI tools recommend top-of-the-range hotels. In this case we carried out a robust test of equality of means with Welch statistics. The results show there is a statistically significant difference between AI tool (groups) means. For the hotel category variable, we then carried out Games-Howell post-hoc test to compare the means of each AI tool with the others, and it seems that ChatGPT is the only AI tool to differ significantly from the others (Table 10 ). Table 10 Multiple comparison per variable and AI tool Games-Howell test (stars dependent variable) Bonferroni test (price dependent variable) Games-Howell test (Booking.com score dependent variable) Games-Howell test (TripAdvisor score dependent variable) Name tool (i) Name tool (j) Mean difference (i-j) Std. Error Mean difference (i-j) Std. Error Mean difference (i-j) Std. Error Mean difference (i-j) Std. Error ChatGPT Gemini 0.308** 0.100 183.518* 44.747 0.296 0.109 0.1924** 0.069 Claude 0.582* 0.084 259.039* 41.595 0.673* 0.098 0.2524* 0.056 Grok 0.702* 0.112 343.573* 41.596 0.334** 0.109 0.1413 0.053 DeepSeek 0.681* 0.113 197.886* 40.865 0.289* 0.073 0.1314* 0.044 Gemini ChatGPT -0.308** 0.100 -183.518* 44.747 -0.296 0.109 -0.1924** 0.069 Claude 0.274 0.127 75.521 41.314 0.377** 0.133 0.0600 0.078 Grok 0.394 0.147 160.056* 41.315 0.042 0.142 -0.0511 0.076 DeepSeek 0.373 0.148 14.368 40.579 -0.007 0.116 -0.0610 0.070 Claude ChatGPT -0.582* 0.084 -259.039* 41.596 -0.673* 0.098 − .02524* 0.056 Gemini -0.274 0.127 -75.522 41.314 -0.377** 0.133 -0.0600 0.078 Grok 0.120 0.136 84.534 37.878 -0.335 0.134 -0.1111 0.064 DeepSeek 0.098 0.137 -61.153 37.075 -0.384** 0.106 -0.1210 0.057 Grok ChatGPT -0.702* 0.112 -343.573* 41.596 -0.338** 0.109 -0.1413 0.053 Gemini -0.394 0.147 -160.056* 41.315 -0.042 0.142 0.0511 0.076 Claude -0.120 0.136 -84.534 37.878 .0335 0.134 0.1111 0.064 DeepSeek -0.022 0.156 -145.687* 37.075 -0.049 0.117 -0.0099 0.054 DeepSeek ChatGPT -0.681* 0.113 -197.886* 40.865 -0.289* 0.073 -0.1314** 0.044 Gemini -0.373 0.148 -14.368 40.579 0.007 0.116 0.0610 0.070 Claude -0.098 0.137 61.153 37.075 0.384** 0.106 0.1210 0.057 Grok 0.022 0.156 145.687* 37.075 0.049 0.117 0.0099 0.054 * sig: <0.001; ** sig: <0.05 Table 11 PRICE descriptives by AI tool AI tool N Mean Std. Dev. Min. Max. ChatGPT 82 928.06 828.551 147 4047 Gemini 80 1111.20 2990.316 57 26849 Claude 114 561.37 875.472 21 7625 Grok 115 652.40 692.342 23 4480 DeepSeek 125 712.48 817.062 30 4480 Total 516 761.78 1396.495 21 26849 Price in € Regarding the price variable (Table 11 ), Gemini recommends much more expensive hotels than the other tools (mean €1,111.20), and Claude suggests the lowest priced establishments (mean €561.37). In this case, for the price variable, we carried out a Bonferroni test to analyse the significance of price mean differences between AI tools, as the assumption of homogeneity of variances was confirmed by Levene’s test. Regarding the multiple comparison of the price variable, Table 10 shows that the differences are statistically significant for the majority of tools (groups). As in the case of the stars variable, ChatGPT also shows significant differences from all the other AI tools. In this case, however, we applied the Games-Howell test due to the violation of the homogeneity of variances, as indicated by Levene’s test. Table 12 Booking.com score descriptives by AI tool AI tool N Mean Std. Dev. Min. Max. ChatGPT 81 9.006 0.386 7.90 10.00 Gemini 84 8.710 0.919 5.50 9.70 Claude 111 8.333 0.927 3.60 9.70 Grok 117 8.668 1.086 0.00 9.70 DeepSeek 125 8.718 0.659 6.20 9.70 Total 518 8.668 0.868 0.00 10.00 With regard to the Booking.com score (Table 12 ), the means are quite similar for all five AI tools, but some statistical differences arise. In this case, for the Booking.com score variable, again we carried out a Games-Howell test, as was indicated by Levene’s test. ChatGPT presents significant differences from Claude and DeepSeek in particular, but also from Grok (at 0.05 level). There are also some statistical differences between DeepSeek and Claude. Table 13 TripAdvisor score descriptives by AI tool AI tool N Mean Std. Dev. Min. Max. ChatGPT 83 4.663 0.272 4.0 5.0 Gemini 84 4.470 0.572 2.0 5.0 Claude 117 4.410 0.511 1.5 5.0 Grok 117 4.521 0.471 1.0 5.0 DeepSeek 128 4.531 0.365 3.0 5.0 Total 529 4.513 0.455 1.0 5.0 Regarding the score on TripAdvisor, Table 13 shows the descriptives for each AI tool and it can be observed that the means are very similar. For the TripAdvisor score variable, again we carried out a Games-Howell post-hoc test to compare the means of the AI tools (Table 9 ). The results show that, again, ChatGPT presents significant differences from the other AI tools, except Grok. It seems that ChatGPT and Grok are more prone than other chatbots to recommending hotels that have high-scoring user reviews. 5. Conclusions This study confirms that chatbots can generate hotel recommendations that align with real-world data to varying degrees. While all tools showed statistically significant correlations with real information on hotels, the degree of accuracy, reliability and bias differed across platforms. ChatGPT, for example, consistently recommended higher-category hotels but tended to underestimate both prices and user scores. Gemini performed well in terms of accuracy for star ratings and review scores, although it also underestimated prices. Claude displayed the strongest correlations across all variables and suggested significantly lower hotel categories. Grok and DeepSeek maintained relatively balanced responses, with DeepSeek showing minimal bias in star ratings but underestimating prices. These findings underscore that no single LLM performs optimally across all evaluation dimensions. Therefore, when leveraging AI tools for travel recommendations, users should select the model that best matches their strategic needs, whether that is luxury targeting, price-sensitive suggestions, reliability or balanced accuracy. 5.1 Theoretical and methodological contribution This study contributes to the ongoing discussion concerning the reliability of generative AI chatbots by being the first to compare the accuracy and biases of different chatbots. This comparison provides unique insights into LLM training and use. We also contribute to the emerging literature on travel-related LLM recommendations by providing insights into how and why these chatbots generate hotel suggestions. The observed correlations with Booking.com and TripAdvisor indicate that data from these platforms have influenced LLM training, reinforcing their importance for hotels aiming to be represented in AI-driven searches. As customers increasingly turn to LLMs to plan their travel, understanding this connection becomes critical for both research and practice in the tourism and hospitality sectors. Importantly, this article also offers a replicable and scalable methodological framework to analyse and compare evolving chatbots. This framework can be applied in future studies to test new AI agents, track improvements over time, and explore different contexts such as other tourism segments, geographic regions, or languages. By continuously monitoring model behaviour, tourism professionals and researchers can ensure responsible, effective, and user-centred AI deployment in hospitality. Given the novelty of this topic, this study addresses a significant gap in the literature by being the first to systematically compare multiple generative AI chatbots in the tourism sector. The proposed methodology allows transparent evaluation of these black-box systems, offering a starting point to better understand their internal decision-making processes. As chatbot outputs vary notably across platforms and sometimes diverge from real-world data, this research demonstrates that the choice of chatbot significantly affects the information retrieved by users, an essential insight for both researchers and practitioners. The originality and replicability of the proposed framework provide a strong foundation for future academic work, especially as the pace of chatbot development accelerates. Moreover, this study highlights the urgent need for more comparative evaluations to ensure the reliability and fairness of recommendations in high-stakes domains like tourism. 5.2 Practical contribution Until now, companies aiming for top positions on Google needed to conduct a keyword study specific to their market niche. With AI chatbots, however, this strategy shifts towards analysing a set of semantically related words tied to those keywords, which then become a series of values that allow the algorithm to better calculate how to respond to a query. This approach adapts strategies to search intent rather than relying on exact keywords. As AI continues to evolve, the integration of real-time searches, personalization, and data memory will define how hotels are discovered and recommended. In the future, paid AI placements could emerge, as brands seek visibility within generative responses. Thus, understanding how these tools interpret, prioritize, and segment recommendations is critical for digital competitiveness in the tourism sector (Dwivedi et al., 2024 ) From a strategic standpoint, hospitality businesses should start adapting their content and SEO strategies to AI-based platforms, applying Generative Engine Optimization (GEO), Answering Engine Optimization (AEO) and advanced prompting engineering techniques like Chain-of-Thought (CoT) and Tree-of-Thought (ToT) (Aggarwal et al., 2024 ; Saleh, 2025 ). CoT allows chatbots to generate step-by-step reasoning sequences, while ToT expands this logic by introducing multiple branches of thought and intermediate evaluations to explore more deliberate and optimal reasoning paths (Zhang et al., 2024 ). Importantly, effective prompt engineering by companies can influence how LLMs interpret and prioritize information, ultimately shaping the recommendations presented to customers. The evolution of these chatbots will depend on their ability to adapt to new environments and to address ethical and technical challenges (Edson de Carvalho Souza, 2025). These aforementioned searches enable a more precise segmentation based on user intent, thanks to the ability of AI chatbots’ ability to process complex queries and respond in a personalized manner. This represents a significant strategic shift for businesses as well as for tourism and hospitality destinations. The capacity to provide recommendations or resolve questions in real time could also reduce the need for traditional chatbots, as AI can handle complex, contextual conversations. Brands that adapt to this shift by creating high-quality content and using data ethically will gain an advantage in this new digital era. AI Chatbot Ads, similar to Google Ads, may appear in the future. However, there is also a risk that recommendations could increasingly be influenced by paid advertising or ranking mechanisms that remain opaque, raising important questions about transparency and trust in AI-driven outputs. Advertising on AI platforms could become less intrusive and more focused on delivering value at the right moment. The chatbot’s memory feature allows conversational AI to collect data that can be used to create more detailed user profiles, enabling even more targeted segmentation for future campaigns. 5.3 Limitations and future research Our research is primarily descriptive and constrained by the emerging nature of generative AI tools, which are still undergoing rapid development. Future improvements may significantly change how they perform. Additionally, only one version per tool was tested. Paid versions could differ in their responses and capabilities. Finally, the pace at which these chatbots evolve presents a challenge, as research in this area requires constant updates to remain relevant, accurate and reliable. Future research could compare results with other AI tools that have search access, such as Perplexity, or that conduct searches in different languages. Another promising line of research would be to analyse the performance of the new AI agents that each LLM is progressively launching, such as ChatGPT’s agents, which became available in Europe on 17 July 2025. These findings can be especially useful for hotel managers, tourism marketers, and destination management organizations (DMOs), who can use this information to select the AI chatbot that best aligns with their branding and pricing strategy. Moreover, researchers and developers working on AI-based recommender systems can apply the proposed methodology to assess model performance and fairness across new platforms and languages. Declarations Author Contribution JM-O: conceptualization and writing the main manuscript textEM-F: conceptualization and data analysisBF-R: data analysis and results interpretationJP: manuscript revision References Afifah, S., Mudzakir, A., & Nandiyanto, A. B. D. (2022). How to Calculate Paired Sample t-Test using SPSS Software: From Step-by-Step Processing for Users to the Practical Examples in the Analysis of the Effect of Application Anti-Fire Bamboo Teaching Materials on Student Learning Outcomes. Indonesian Journal of Teaching in Science , 2 (1), 81–92. https://doi.org/10.17509/ijotis.v2i1.45895 Agbangba, C. E., Sacla Aide, E., Honfo, H., & Glèlè Kakai, R. (2024). On the use of post-hoc tests in environmental and biological sciences: A critical review. Heliyon , 10 (3), e25131. https://doi.org/10.1016/j.heliyon.2024.e25131 Aggarwal, P., Murahari, V., Rajpurohit, T., Kalyan, A., Narasimhan, K., & Deshpande, A. (2024). GEO: Generative Engine Optimization. Proceedings of the 30th ACM SIGKDD Conference on Knowledge Discovery and Data Mining , 5–16. https://doi.org/10.1145/3637528.3671900 Bailyn, E. (2025). Top Generative AI Chatbots by Market Share – August 2025 . https://firstpagesage.com/reports/top-generative-ai-chatbots/ Bareh, C. K. (2025). A qualitative assessment of the accuracy of AI-LLM in academic research. AI and Ethics , 1–20. https://doi.org/10.1007/s43681-025-00730-8 Bhaskar, P., & Sharma, K. D. (2022). A critical insight into the role of artificial intelligence (AI) in tourism and hospitality industries. Pacific Business Review International, 15 (3). Buhalis, D., & Moldavska, I. (2022). Voice assistants in hospitality: using artificial intelligence for customer service. Journal of Hospitality and Tourism Technology, 13 (3), 386-403. https://doi.org/10.1108/JHTT-03-2021-0104 Bulchand-Gidumal, J. (2022). Impact of artificial intelligence in travel, tourism, and hospitality. In Handbook of e-Tourism (pp. 1943-1962). Cham: Springer International Publishing. https://doi.org/10.1007/978-3-030-48652-5_110 Chester Business School, University of Chester, United Kingdom, & Leong, K. (2024). An exploratory study of AI participants’ views on using AI in business hotels. Journal of Applied Sciences in Travel and Hospitality , 7 (2), 69–83. https://doi.org/10.31940/jasth.v7i2.69-83 Chitaliya, H., Jain, D., Tiwari, S., Khati, G. S., Sankhe, D., Kanani, P., & Bhanja, M. (2025). ExplainableTrip: An XAI-Driven Personalized Travel Recommendation System Using Gradient Boosting and SHAP. Ingénierie des Systèmes d’Information, 30 (5). https://doi.org/10.18280/isi.300507 Dang, Q., & Li, G. (2025). Unveiling trust in AI: the interplay of antecedents, consequences, and cultural dynamics. AI & Society. https://doi.org/10.1007/s00146-025-02477-6 de Cerqueira, J. S., Kemell, K. K., Rousi, R., Xi, N., Hamari, J., & Abrahamsson, P. (2025). Mapping trustworthiness in large language models: A bibliometric analysis bridging theory to practice. https://doi.org/10.48550/arXiv.2503.0478 Deldjoo, Y. (2024). Understanding biases in ChatGPT-based recommender systems: Provider fairness, temporal stability, and recency. ACM Transactions on Recommender Systems . https://doi.org/10.1145/3690655 Di Palma, D., Biancofiore, G. M., Anelli, V. W., Narducci, F., Di Noia, T., & Di Sciascio, E. (2023). Evaluating chatgpt as a recommender system: A rigorous approach. arXiv preprint arXiv:2309.03613. https://doi.org/10.48550/arXiv.2309.03613 Dwivedi, Y. K., Pandey, N., Currie, W., & Micu, A. (2024). Leveraging ChatGPT and other generative artificial intelligence (AI)-based applications in the hospitality and tourism industry: Practices, challenges and research agenda. International Journal of Contemporary Hospitality Management , 36 (1), 1–12. https://doi.org/10.1108/IJCHM-05-2023-0686 Edson de Carvalho Souza, M. (2025). Grok, Gemini, ChatGPT and DeepSeek: Comparison and Applications in Conversational Artificial Intelligence . https://doi.org/10.5281/ZENODO.14885243 Enternat. (2024). Ciudades más visitadas del mundo 2025 (ranking de turismo) . https://www.enterat.com/ocio/ciudades-mas-visitadas-mundo.php Fernández, Y. (2025). Grok 3: Qué es y principales novedades del nuevo modelo de inteligencia artificial de xAI [Xataka]. https://www.xataka.com/basics/grok-3-que-principales-novedades-nuevo-modelo-inteligencia-artificial-xai Hasan, E., Rahman, M., Ding, C., Huang, J., & Raza, S. (2024). Based recommender systems: A survey of approaches, challenges and future perspectives. ACM Computing Surveys. https://doi.org/10.1145/3742421 Ilieva, G., Yankova, T., & Klisarova-Belcheva, S. (2024). Effects of Generative AI in Tourism Industry. Information , 15 (11), 671. https://doi.org/10.3390/info15110671 Imran, M., & Almusharraf, N. (2024). Google Gemini as a next generation AI educational tool: A review of emerging educational technology. Smart Learning Environments , 11 (1), 22. https://doi.org/10.1186/s40561-024-00310-z Islam, R., & Ahmed, I. (2024). Gemini-the most powerful LLM: Myth or Truth. 2024 5th Information Communication Technologies Conference (ICTC) , 303–308. https://doi.org/10.1109/ICTC61510.2024.10602253 Kim, J., Kim, J. H., Kim, C., & Park, J. (2023). Decisions with ChatGPT: Reexamining choice overload in ChatGPT recommendations. Journal of Retailing and Consumer Services, 75 . https://doi.org/10.1016/j.jretconser.2023.103494 Kim, J. H., Kim, J., Park, J., Kim, C., Jhang, J., & King, B. (2025). When ChatGPT gives incorrect answers: The impact of inaccurate information by generative AI on tourism decision-making. Journal of Travel Research, 64 (1), 51–73. https://doi.org/10.1177/0047287523121299 Kowald, D. (2025). Investigating Popularity Bias Amplification in Recommender Systems Employed in the Entertainment Domain. https://doi.org/10.48550/arXiv.2504.04752 Labrague, L. J. (2024). Utilizing Artificial Intelligence–Based Tools for Addressing Clinical Queries: ChatGPT Versus Google Gemini. Journal of Nursing Education , 63 (8), 556–559. https://doi.org/10.3928/01484834-20240426-01 Lee, M., Kwon, W., & Back, K. J. (2021). Artificial intelligence for hospitality big data analytics: Developing a prediction model of restaurant review helpfulness for customer decision-making. International Journal of Contemporary Hospitality Management, 33 (6), 2117-2136. https://doi.org/10.1108/IJCHM-06-2020-0587 Lichtenberg, J. M., Buchholz, A., & Schwöbel, P. (2024). Large language models as recommender systems: A study of popularity bias. arXiv preprint arXiv:2406.01285. Liu, H. (2015). Comparing Welch’s ANOVA, a Kruskal-Wallis test and traditional ANOVA in case of Heterogeneity of Variance . Liebenow, L. W., Schmidt, F. T., Meyer, J., & Fleckenstein, J. (2025). Self-assessment accuracy in the age of artificial intelligence: Differential effects of LLM-generated feedback. Computers & Education , 105385. https://doi.org/110.1016/j.compedu.2025.105385 Lu, P., & Feng, X. (2024, May). Personalized recommendation algorithm in AI-assisted learning system. In 2024 6th International Conference on Communications, Information System and Computer Engineering (CISCE) (pp. 1289–1294). IEEE. https://doi.org/10.1109/CISCE62493.2024.10653402 Lu, H., Liu, W., Zhang, B., Wang, B., Dong, K., Liu, B., Sun, J., Ren, T., Li, Z., Yang, H., Sun, Y., Deng, C., Xu, H., Xie, Z., & Ruan, C. (2024). DeepSeek-VL: Towards Real-World Vision-Language Understanding (arXiv:2403.05525). arXiv. https://doi.org/10.48550/arXiv.2403.05525 Malak, A., & Şahin, M. F. (2024). How Useful are Current Chatbots Regarding Urology Patient Information? Comparison of the Ten Most Popular Chatbots’ Responses About Female Urinary Incontinence. Journal of Medical Systems , 48 (1), 102. https://doi.org/10.1007/s10916-024-02125-4 Masciari, E., Umair, A., & Ullah, M. H. (2024). A systematic literature review on AI-based recommendation systems and their ethical considerations. IEEE Access, 12 , 121223–121241. https://doi.org/10.1109/ACCESS.2024.3451054 Mladenović, D., Beheshti, M., Kolar, T., Ismagilova, E., & Dwivedi, Y. K. (2024). Synthetic WOM? The emergence of generative artificial intelligence-induced recommendations. Journal of Computer Information Systems , 1–18. https://doi.org/10.1080/08874417.2024.2409252 Omar, M., Sorin, V., Agbareia, R., Apakama, D. U., Soroush, A., & Sakhuja,A. (2025). Evaluating and addressing demographic disparities in medical large language models: a systematic review. International Journal for Equity in Health, 24 (1), 57. https://doi.org/10.1186/s12939-025-02419-0 OpenAI. (2024). Introducing ChatGPT search . https://openai.com/index/introducing-chatgpt-search/ Pande, A., Patil, R., Mukkemwar, R., Panchal, R., & Bhoite, S. (2024). Comprehensive Study of Google Gemini and Text Generating Models: Understanding Capabilities and Performance . Petkovic, D. (2023). It is not “Accuracy vs. Explainability”—we need both for trustworthy AI systems. IEEE Transactions on Technology and Society, 4 (1), 46–53. https://doi.org/10.1109/TTS.2023.3239921 Pillai, R., & Sivathanu, B. (2020). Adoption of AI-based chatbots for hospitality and tourism. International Journal of Contemporary Hospitality Management, 32 (10), 3199-3226. https://doi.org/10.1108/IJCHM-04-2020-0259 Priyanshu, A., Maurya, Y., & Hong, Z. (2024). AI Governance and Accountability: An Analysis of Anthropic’s Claude (arXiv:2407.01557). arXiv. https://doi.org/10.48550/arXiv.2407.01557 Popesku, J. (2019). Current applications of artificial intelligence in tourism and hospitality. In Sinteza 2019-International Scientific Conference on Information Technology and Data Related Research (84-90). Singidunum University. https://doi.org/10.15308/Sinteza-2019-84-90 Ray, P. P. (2024). A deep introspection into the role of ChatGPT for transforming hospitality, leisure, sport, and tourism education. Journal of Hospitality, Leisure, Sport & Tourism Education , 35 , 100504. https://doi.org/10.1016/j.jhlste.2024.100504 Remountakis, M., Kotis, K., Kourtzis, B., & Tsekouras, G. E. (2023). ChatGPT and persuasive technologies for the management and delivery of personalized recommendations in hotel hospitality. https://doi.org/10.48550/arXiv.2307.14298 Reuters. (2025). Chinese universities launch DeepSeek courses to capitalise on AI boom . https://www.reuters.com/technology/artificial-intelligence/chinese-universities-launch-deepseek-courses-capitalise-ai-boom-2025-02-21/?utm_source=chatgpt.com Saleh, M. I. (2025). Generative artificial intelligence in hospitality and tourism: Future capabilities, AI prompts and real-world applications. Journal of Hospitality Marketing & Management , 1–32. https://doi.org/10.1080/19368623.2025.2458603 Said, A. (2025). On explaining recommendations with Large Language Models: A review. Frontiers in Big Data, 7, 1505284. https://doi.org/10.3389/fdata.2024.1505284 Schmidgall, S., Harris, C., Essien, I., Olshvang, D., Rahman, T., Kim, J. W., & Chellappa, R. (2024). Evaluation and mitigation of cognitive biases in medical language models. npj Digital Medicine, 7 (1), 295. https://doi.org/10.1038/s41746-024-01283-6 Sedgwick, P. (2012). Pearson’s correlation coefficient. BMJ , 345 (jul04 1), e4483–e4483. https://doi.org/10.1136/bmj.e4483 Sigala, M., Ooi, K.-B., Tan, G. W.-H., Aw, E. C.-X., Buhalis, D., Cham, T.-H., Chen, M.-M., Dwivedi, Y. K., Gretzel, U., Inversini, A., Jung, T., Law, R., & Ye, I. H. (2024). Understanding the impact of ChatGPT on tourism and hospitality: Trends, prospects and research agenda. Journal of Hospitality and Tourism Management , 60 , 384–390. https://doi.org/10.1016/j.jhtm.2024.08.004 Štilić, A., Nicić, M., & Puška, A. (2023). Check-in to the future: Exploring the impact of contemporary information technologies and artificial intelligence on the hotel industry. Turističko poslovanje, (31). https://doi.org/10.5937/turpos0-43739 Tussyadiah, I. (2020). A review of research into automation in tourism: Launching the Annals of Tourism Research Curated Collection on Artificial Intelligence and Robotics in Tourism. Annals of Tourism Research , 81 , 102883. https://doi.org/10.1016/j.annals.2020.102883 Vashishth, T. K., Sharma, V., Sharma, K. K., Kumar, B., Kumar, A., & Panwar, R. (2024). Artificial intelligence (AI)–powered chatbots: Providing instant support and personalized recommendations to guests 24/7. In Technology and Luxury Hospitality (211-236). Routledge. https://doi.org/10.4324/9781003488248-15 Viglia, G., Adler, S. J., Miltgen, C. L., & Sarstedt, M. (2024). The use of synthetic data in tourism. Annals of Tourism Research , 108 , 103819. https://doi.org/10.1016/j.annals.2024.103819 Wong, I. A., Lian, Q. L., & Sun, D. (2023). Autonomous travel decision-making: An early glimpse into ChatGPT and generative AI. Journal of Hospitality and Tourism Management, 56, 253-263. https://doi.org/10.1016/j.jhtm.2023.06.022 Wu, S., Koo, M., Blum, L., Black, A., Kao, L., Scalzo, F., & Kurtz, I. (2023). A Comparative Study of Open-Source Large Language Models, GPT-4 and Claude 2: Multiple-Choice Test Taking in Nephrology (arXiv:2308.04709). arXiv. https://doi.org/10.48550/arXiv.2308.04709 Wu, Z., Chen, X., Pan, Z., Liu, X., Liu, W., Dai, D., Gao, H., Ma, Y., Wu, C., Wang, B., Xie, Z., Wu, Y., Hu, K., Wang, J., Sun, Y., Li, Y., Piao, Y., Guan, K., Liu, A., … Ruan, C. (2024). DeepSeek-VL2: Mixture-of-Experts Vision-Language Models for Advanced Multimodal Understanding (arXiv:2412.10302). arXiv. https://doi.org/10.48550/arXiv.2412.10302 Zhang, X., Du, C., Pang, T., Liu, Q., Gao, W., & Lin, M. (2024). Chain of Preference Optimization: Improving Chain-of-Thought Reasoning in LLMs . Additional Declarations No competing interests reported. Cite Share Download PDF Status: Under Review Version 1 posted Editorial decision: Revision requested 17 Feb, 2026 Reviews received at journal 16 Feb, 2026 Reviewers agreed at journal 04 Feb, 2026 Reviews received at journal 24 Nov, 2025 Reviewers agreed at journal 05 Nov, 2025 Reviewers invited by journal 04 Nov, 2025 Editor assigned by journal 10 Oct, 2025 Submission checks completed at journal 25 Sep, 2025 First submitted to journal 22 Sep, 2025 You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-7687581","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":540277953,"identity":"d53257b2-7e11-4b78-816d-a7b1f2eb4ad3","order_by":0,"name":"Julia Marti-Ochoa","email":"","orcid":"","institution":"University of Lleida","correspondingAuthor":false,"prefix":"","firstName":"Julia","middleName":"","lastName":"Marti-Ochoa","suffix":""},{"id":540277954,"identity":"3486218b-7351-46df-a4ac-4f357db36eb0","order_by":1,"name":"Eva Martin-Fuentes","email":"","orcid":"","institution":"University of Lleida","correspondingAuthor":false,"prefix":"","firstName":"Eva","middleName":"","lastName":"Martin-Fuentes","suffix":""},{"id":540277955,"identity":"636ee2ab-bcb4-430b-95db-e7b97b0961cd","order_by":2,"name":"Berta Ferrer-Rosell","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA5klEQVRIie2OvYrCQBSFzzCQaQ7YpgjZV5iQIoTguygLsdFim2U7lYVU2X4gL+E7BLSz3k5BWBsLxcbCYpdkC6tJSov5mvvD/bgHcDieEN2WsC1nyKCPIhZA3AzCQBLtpqfyd99DSdTXcX+DDpOqXl+HRUao+ri3KWm5iZYldBxs89dqVkwI5pE92HcuFsR9bMhYzoqa8NGh7H7E8g49NxxcZdoo6tLxxROfhB75pJSiUWj/kpa5qAKtI0MvFuV2Qo/TN2NTErXG5fShX3zKA27vWThQm9XZpvzHe+i97nOHw+FwdPELBgs7xrlGoaoAAAAASUVORK5CYII=","orcid":"","institution":"University of Lleida","correspondingAuthor":true,"prefix":"","firstName":"Berta","middleName":"","lastName":"Ferrer-Rosell","suffix":""},{"id":540277958,"identity":"0a9bb36e-07e7-4dfb-b445-bc895fcf2729","order_by":3,"name":"Juho Pesonen","email":"","orcid":"","institution":"University of Eastern Finland","correspondingAuthor":false,"prefix":"","firstName":"Juho","middleName":"","lastName":"Pesonen","suffix":""}],"badges":[],"createdAt":"2025-09-23 01:54:34","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-7687581/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-7687581/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":96242852,"identity":"4e093918-3092-4f70-82f8-650784fd9d8b","added_by":"auto","created_at":"2025-11-19 07:14:37","extension":"docx","order_by":0,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":2899315,"visible":true,"origin":"","legend":"","description":"","filename":"anonymousmanuscript.docx","url":"https://assets-eu.researchsquare.com/files/rs-7687581/v1/6023562df2d65110952251fe.docx"},{"id":95907508,"identity":"9bc7cc0c-1e57-45db-9863-dcf4f0f726d7","added_by":"auto","created_at":"2025-11-14 09:39:39","extension":"json","order_by":1,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":5251,"visible":true,"origin":"","legend":"","description":"","filename":"91683582fdf04e68ba2ecd9cc763f878.json","url":"https://assets-eu.researchsquare.com/files/rs-7687581/v1/67023ff63698fc80446bab73.json"},{"id":95907527,"identity":"7ee26d28-12ef-4176-bc8c-083e1c86998d","added_by":"auto","created_at":"2025-11-14 09:39:48","extension":"xml","order_by":2,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":190903,"visible":true,"origin":"","legend":"","description":"","filename":"91683582fdf04e68ba2ecd9cc763f8781enriched.xml","url":"https://assets-eu.researchsquare.com/files/rs-7687581/v1/0faffa1bc14a7ff1529e0b17.xml"},{"id":95907521,"identity":"73c8847d-0263-455e-aa51-b3efb052b4cc","added_by":"auto","created_at":"2025-11-14 09:39:41","extension":"png","order_by":3,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":779641,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage1.png","url":"https://assets-eu.researchsquare.com/files/rs-7687581/v1/a5bdf2e9a625d239a1f73b2e.png"},{"id":95907523,"identity":"34d99188-7854-489c-8a67-03ec4a9b0d0a","added_by":"auto","created_at":"2025-11-14 09:39:41","extension":"png","order_by":4,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":460925,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage2.png","url":"https://assets-eu.researchsquare.com/files/rs-7687581/v1/2d32f5cd2b16a559a326acd7.png"},{"id":96243711,"identity":"c283a5f2-ebfd-4481-9499-67d7c06f64a1","added_by":"auto","created_at":"2025-11-19 07:16:52","extension":"png","order_by":5,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":920592,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage3.png","url":"https://assets-eu.researchsquare.com/files/rs-7687581/v1/e63b4e55c81c127f8a6d1efe.png"},{"id":95907526,"identity":"c1bbfe26-88fa-4d91-b36b-9f49c20552de","added_by":"auto","created_at":"2025-11-14 09:39:48","extension":"png","order_by":6,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":290721,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage4.png","url":"https://assets-eu.researchsquare.com/files/rs-7687581/v1/b0a1ef9d6ebfa12474a5fc0f.png"},{"id":95907514,"identity":"f527f249-a970-49f3-8532-07a45396f33f","added_by":"auto","created_at":"2025-11-14 09:39:40","extension":"png","order_by":7,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":336022,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage5.png","url":"https://assets-eu.researchsquare.com/files/rs-7687581/v1/955153fde057b09cc335ccc2.png"},{"id":96242847,"identity":"658531a0-eaa6-4f68-9464-fd1088315c4c","added_by":"auto","created_at":"2025-11-19 07:14:36","extension":"png","order_by":8,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":94129,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage1.png","url":"https://assets-eu.researchsquare.com/files/rs-7687581/v1/7c9d11b4da43ceaa117c3d19.png"},{"id":95907516,"identity":"7e870964-b2b5-49c8-8d5b-26467f2d3de9","added_by":"auto","created_at":"2025-11-14 09:39:40","extension":"png","order_by":9,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":78285,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage2.png","url":"https://assets-eu.researchsquare.com/files/rs-7687581/v1/39832ea7e31e3f3aacfde332.png"},{"id":95907517,"identity":"622a7e6f-700e-4562-adbb-22201d3a5c3c","added_by":"auto","created_at":"2025-11-14 09:39:40","extension":"png","order_by":10,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":120928,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage3.png","url":"https://assets-eu.researchsquare.com/files/rs-7687581/v1/c6e7f053aca35f88b487015f.png"},{"id":95907518,"identity":"1a500878-d5a7-48f4-894a-eaf043d14ba3","added_by":"auto","created_at":"2025-11-14 09:39:40","extension":"png","order_by":11,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":51142,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage4.png","url":"https://assets-eu.researchsquare.com/files/rs-7687581/v1/9e3f90c79f908f6173ca45d7.png"},{"id":96243235,"identity":"e6cf4d2c-bed9-4417-8f5c-82437a5d1cf1","added_by":"auto","created_at":"2025-11-19 07:15:53","extension":"png","order_by":12,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":58423,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage5.png","url":"https://assets-eu.researchsquare.com/files/rs-7687581/v1/8a8ba519101dad7713f9ac04.png"},{"id":95907519,"identity":"edb2d046-a486-4a78-8d62-50b684a55dde","added_by":"auto","created_at":"2025-11-14 09:39:40","extension":"xml","order_by":13,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":189699,"visible":true,"origin":"","legend":"","description":"","filename":"91683582fdf04e68ba2ecd9cc763f8781structuring.xml","url":"https://assets-eu.researchsquare.com/files/rs-7687581/v1/857f07d91a81c97306c13261.xml"},{"id":95907520,"identity":"bcb67a58-4005-452c-8a19-50951ad5eb67","added_by":"auto","created_at":"2025-11-14 09:39:40","extension":"html","order_by":14,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":195287,"visible":true,"origin":"","legend":"","description":"","filename":"earlyproof.html","url":"https://assets-eu.researchsquare.com/files/rs-7687581/v1/5e88608a922623ffc6eddf57.html"},{"id":95907507,"identity":"b6d4c078-2f16-4401-a95b-a83a5c4e4353","added_by":"auto","created_at":"2025-11-14 09:39:39","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":715951,"visible":true,"origin":"","legend":"\u003cp\u003eChatGPT-40 – AI-generated hotel recommendations for New York City\u003c/p\u003e","description":"","filename":"floatimage1.png","url":"https://assets-eu.researchsquare.com/files/rs-7687581/v1/05e4e609f314c4364e272151.png"},{"id":96242152,"identity":"953491e0-d4a9-44e0-bf46-8c3ad8ecd8ea","added_by":"auto","created_at":"2025-11-19 07:12:13","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":426295,"visible":true,"origin":"","legend":"\u003cp\u003eGemini 1.5 Flash – AI-generated hotel recommendations for New York City\u003c/p\u003e","description":"","filename":"floatimage2.png","url":"https://assets-eu.researchsquare.com/files/rs-7687581/v1/add4ad9027693c411d8b348f.png"},{"id":95907509,"identity":"dd656332-ef17-45b4-b29a-e89e86405a35","added_by":"auto","created_at":"2025-11-14 09:39:40","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":920592,"visible":true,"origin":"","legend":"\u003cp\u003eClaude 3.5 Sonnet: AI-generated hotel recommendations for New York City\u003c/p\u003e","description":"","filename":"floatimage3.png","url":"https://assets-eu.researchsquare.com/files/rs-7687581/v1/fdf23a58e300a1163f7dbd65.png"},{"id":95907522,"identity":"e7251dda-67bc-4169-9205-39fe5b556514","added_by":"auto","created_at":"2025-11-14 09:39:41","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":290721,"visible":true,"origin":"","legend":"\u003cp\u003eDeepThink R1: AI-generated hotel recommendations for New York City\u003c/p\u003e","description":"","filename":"floatimage4.png","url":"https://assets-eu.researchsquare.com/files/rs-7687581/v1/765740149ae925fbd84c5fc6.png"},{"id":95907511,"identity":"ff0afef3-5360-4d53-b3a0-b2db20be526d","added_by":"auto","created_at":"2025-11-14 09:39:40","extension":"png","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":336022,"visible":true,"origin":"","legend":"\u003cp\u003eGrok 3: AI-generated hotel recommendations for New York City\u003c/p\u003e","description":"","filename":"floatimage5.png","url":"https://assets-eu.researchsquare.com/files/rs-7687581/v1/9e142451d9aa0c3107b2ec2c.png"},{"id":96452916,"identity":"2f9ea2d6-f009-4360-847c-6e09f7604dd5","added_by":"auto","created_at":"2025-11-21 09:53:55","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":3781940,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7687581/v1/c835ef00-5963-417e-9c90-d02bc5f5cb50.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"Are all Generative AI Chatbots the Same? Analysing the Reliability of Hotel Recommendations","fulltext":[{"header":"1. Introduction","content":"\u003cp\u003eThe main reason generative AI (GAI) chatbots such as ChatGPT\u0026rsquo;s comprehensive recommendations have a positive impact on users is that they perceive them as being accurate and reliable (Kim et al., \u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e2023\u003c/span\u003e). GAI can provide users with quick, reliable, and interactive responses (Mladenović et al., \u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e2024\u003c/span\u003e), but these responses also have to be accurate and reliable. GAI chatbots have a tendency to sometimes provide incorrect information, which can directly affect the user experience and decision-making (Kim et al., \u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e2025\u003c/span\u003e). These systems also have inherent biases that can affect the recommendation results and user behaviour (Deldjoo, \u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e2024\u003c/span\u003e). Thus, the reliability of generative AI chatbots has become an important research avenue to improve outputs and the user experience.\u003c/p\u003e\u003cp\u003eCurrent research on the reliability of generative AI, particularly large language models (LLMs), primarily revolves around improving their consistency (Kim et al., \u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e2023\u003c/span\u003e), accuracy (Bareh, \u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e2025\u003c/span\u003e), trustworthiness (Lu \u0026amp; Feng, \u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e2024\u003c/span\u003e), and robustness (Masciari et al., \u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e2024\u003c/span\u003e) in diverse applications, as well as developing robust methodologies for their evaluation (Illicki, 2022). Researchers have especially noted that feedback generated by AI chatbots can be inaccurate and vary significantly from real-world values (Liebenow et al., \u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e2025\u003c/span\u003e). Thus, there have been calls for greater understanding of AI accuracy and reliability to build trustworthy AI systems (Petkovic, \u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e2023\u003c/span\u003e).\u003c/p\u003e\u003cp\u003eIn recent years, consumers have increasingly turned to AI-based tools to assist them in decision-making processes, especially in complex and information-rich domains such as travel and hospitality (Bhaskar \u0026amp; Sharma, \u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e2022\u003c/span\u003e). Instead of relying exclusively on traditional search engines or static reviews (Wong, Lian, \u0026amp; Sun, \u003cspan citationid=\"CR55\" class=\"CitationRef\"\u003e2023\u003c/span\u003e), users now interact with conversational platforms that provide direct and personalized responses to queries such as \u0026lsquo;Where should I stay in Tokyo for under \u003cspan\u003e$\u003c/span\u003e150?\u0026rsquo;, or \u0026lsquo;What is the best hotel near the Eiffel Tower with a terrace view?\u0026rsquo;. This shift reflects a broader transformation in how people access, filter, and act upon digital content (Bulchand-Gidumal, \u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e2022\u003c/span\u003e; Popesku, \u003cspan citationid=\"CR42\" class=\"CitationRef\"\u003e2019\u003c/span\u003e).\u003c/p\u003e\u003cp\u003eAI systems based on LLMs are particularly effective in this context because they go beyond simple keyword matching and incorporate contextual understanding, tone, and even intent (Vashishth, Sharma, \u0026amp; Sharma, 2024). This capability enables them to emulate the role of a human travel advisor by dynamically interpreting user preferences and offering recommendations tailored to such constraints as budget, travel style, or timing (Pillai \u0026amp; Sivathanu, \u003cspan citationid=\"CR40\" class=\"CitationRef\"\u003e2020\u003c/span\u003e; Buhalis \u0026amp; Moldavska, \u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e2022\u003c/span\u003e). As a result, these systems have the potential to transform traditional pathways for searching for tourism information, affecting not only user experiences but also the visibility and competitiveness of businesses in the sector (Lee, Kwon, \u0026amp; Back, \u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e2021\u003c/span\u003e; Štilić, Nicić, \u0026amp; Puška, \u003cspan citationid=\"CR51\" class=\"CitationRef\"\u003e2023\u003c/span\u003e).\u003c/p\u003e\u003cp\u003eConsidering the above, the aim of this study is to gain a deeper understanding of how reliable and accurate different GAI chatbots are in the hospitality sector, particularly in generating hotel recommendations for users. Whereas most of the earlier research has focused mainly on ChatGPT, currently the most popular generative AI chatbot (Bailyn, \u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e2025\u003c/span\u003e), we examine and compare the other most popular platforms and their performance.\u003c/p\u003e\u003cp\u003eThus, our research questions are as follow: What kind of differences exist between platforms in hotel recommendations provided by ChatGPT, Claude, Gemini, Grok, and DeepSeek for the world\u0026rsquo;s top 10 tourist destinations? What factors most define the recommendations of each GAI model (e.g., hotel category, average price, Booking.com and Tripadvisor scores)? How reliable are hotel recommendations by GAI chatbots compared to real-world data from Booking.com and TripAdvisor? Is there any bias in the recommendations of these chatbots towards hotels with specific characteristics, such as higher star ratings or prices?\u003c/p\u003e"},{"header":"2. Literature review","content":"\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e\u003ch2\u003e2.1 Generative AI Chatbots\u003c/h2\u003e\u003cp\u003eGenerative AI chatbots based on LLMs, like ChatGPT, Claude and Gemini, for instance, have transformed the way we access and interact with information (Ray, \u003cspan citationid=\"CR43\" class=\"CitationRef\"\u003e2024\u003c/span\u003e). LLMs are designed to process vast amounts of textual data, enabling them to generate coherent responses, analyse complex contexts, and even simulate conversational understanding. Researchers have often adopted a comparative approach to GAI chatbots to understand how accurate and reliable they are. There are various benchmarks that implement accuracy metrics to analyse the performance of GAI chatbots or compare their recommendations with traditional recommendation systems (Di Palma et al., \u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e2023\u003c/span\u003e). Comparisons are also important to understand and optimize performance. Comparing different chatbots can provide insights into how AI is inclined towards certain underlying recommendation approaches (Di Palma et al., \u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e2023\u003c/span\u003e).\u003c/p\u003e\u003cp\u003eCurrently, there are many different GAI chatbots such as ChatGPT, Gemini, Claude, DeepSeek, and Grok. While ChatGPT has become the most popular platform, it has its limitations. It is possible that different platforms can excel in different kinds of recommendations, as all chatbots are fed with different training material and have different algorithms. However, these kinds of comparisons are rarely conducted, especially in tourism and hospitality.\u003c/p\u003e\u003cp\u003eChatGPT, OpenAI\u0026rsquo;s leading LLM, has been consistently improved since the launch of ChatGPT-2 in 2019. Each version has advanced, particularly in language handling and complex interactions. ChatGPT-4o now includes real-time internet search capabilities (OpenAI, \u003cspan citationid=\"CR37\" class=\"CitationRef\"\u003e2024\u003c/span\u003e). This marks a significant advancement, as it is no longer limited to data up to 2021; it can now access links from websites, blogs, newspapers, and more. OpenAI emphasizes that this feature allows ChatGPT to provide live updates on the weather, stock prices, sports schedules, news, and other timely topics (OpenAI, \u003cspan citationid=\"CR37\" class=\"CitationRef\"\u003e2024\u003c/span\u003e).\u003c/p\u003e\u003cp\u003eThis advancement could benefit fields such as education, research, tourism, and more, offering users a richer and more personalized experience. Leading tourism scholars and industry professionals are analysing the potential impacts of ChatGPT on areas such as customer service, travel planning, and destination marketing within the tourism and hospitality sectors (Sigala et al., \u003cspan citationid=\"CR50\" class=\"CitationRef\"\u003e2024\u003c/span\u003e). Gemini, meanwhile, developed by Google DeepMind, represents a significant evolution in the field of LLMs, integrating advanced capabilities in natural language understanding and contextual processing (Labrague, \u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e2024\u003c/span\u003e). Designed as a direct competitor to ChatGPT, Gemini offers unique functionalities such as high-precision text generation, real-time analysis, and seamless integration with other tools in Google\u0026rsquo;s ecosystem, including Google Search and Workspace (Imran \u0026amp; Almusharraf, \u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e2024\u003c/span\u003e). In the tourism sector, Gemini holds the potential to redefine personalized recommendations, leveraging its ability to interpret large volumes of historical and real-time data (Viglia et al., \u003cspan citationid=\"CR54\" class=\"CitationRef\"\u003e2024\u003c/span\u003e). By utilizing geolocated searches, Gemini can suggest accommodation options that align user preferences with contextual factors such as availability, weather, and user reviews (Sigala et al., \u003cspan citationid=\"CR50\" class=\"CitationRef\"\u003e2024\u003c/span\u003e). The increasing integration of generative AI tools in hospitality has already been seen to have a transformative impact on customer services, highlighting both their promise and the challenges of their implementation (Dwivedi et al., \u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e2024\u003c/span\u003e).\u003c/p\u003e\u003cp\u003eClaude, developed by Anthropic, introduces an innovative approach to LLM design by prioritizing ethics and safety in human-AI interactions (Priyanshu et al., \u003cspan citationid=\"CR41\" class=\"CitationRef\"\u003e2024\u003c/span\u003e). Unlike other chatbots, Claude is designed to be more interpretative and less prone to generating inaccurate or harmful responses, making it a reliable tool for sectors like hospitality (Chester Business School, University of Chester, United Kingdom \u0026amp; Leong, \u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e2024\u003c/span\u003e; Ilieva et al., \u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e2024\u003c/span\u003e). This model is optimized for deep contextual understanding and delivers responses in a natural and empathetic manner (Wu et al., \u003cspan citationid=\"CR56\" class=\"CitationRef\"\u003e2023\u003c/span\u003e).\u003c/p\u003e\u003cp\u003eDeepSeek, co-founded by the Chinese hedge fund High-Flyer, represents a new frontier in large multimodal models (LMMs), designed for real-world applications by combining advanced language and vision capabilities. A key feature of DeepSeek is its rigorous training strategy, which emphasizes the balance between vision and language modalities, ensuring that both are optimized without compromising LLM capabilities (Wu et al., \u003cspan citationid=\"CR57\" class=\"CitationRef\"\u003e2024\u003c/span\u003e). Additionally, the model employs instruction-tuning datasets and a use-case taxonomy, enabling robust performance across diverse tasks, including text and image-based contexts (Lu et al., \u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e2024\u003c/span\u003e).\u003c/p\u003e\u003cp\u003eAnother AI tool created by xAI, Grok, stands out for its seamless integration with X (previously known as Twitter) and its emphasis on promoting freedom of speech (Edson de Carvalho Souza, 2025). Grok 3 represents a significant leap forward in the evolution of AI chatbots, incorporating advanced reasoning capabilities and a minimal censorship approach in its responses. Additionally, its integration with DeepSearch, an AI-based search engine, enhances the accuracy and depth of the information generated (Fern\u0026aacute;ndez, \u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e2025\u003c/span\u003e). In certain domains, Grok has been rated as the lowest-quality model with regard to medical information (Malak \u0026amp; Şahin, \u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e2024\u003c/span\u003e).\u003c/p\u003e\u003cp\u003eEnsuring the responsible adoption of AI in tourism is crucial for balancing operational efficiency with essential human interaction, ethical considerations, and service quality (Tussyadiah, \u003cspan citationid=\"CR52\" class=\"CitationRef\"\u003e2020\u003c/span\u003e). Tools like ChatGPT, Claude AI, and Gemini AI can enhance efficiency, but their implementation must prioritize authenticity and user trust to avoid potential drawbacks (Saleh, \u003cspan citationid=\"CR46\" class=\"CitationRef\"\u003e2025\u003c/span\u003e). To achieve this, reliability plays a key role.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec4\" class=\"Section2\"\u003e\u003ch2\u003e2.2 Reliability of LLM Recommendations\u003c/h2\u003e\u003cp\u003eThe use of Large Language Models (LLMs) for recommendation tasks is rapidly evolving, particularly in consumer-oriented industries such as tourism and hospitality. Unlike traditional recommender systems based on collaborative filtering or content-based algorithms, LLMs generate recommendations through natural language understanding and contextual analysis (Remountakis, Kotis, \u0026amp; Kourtzis, 2023). These capabilities allow nuanced interactions with users, offering suggestions inferred from implicit preferences and contextual cues.\u003c/p\u003e\u003cp\u003eRecent research has explored LLM-based explainable recommender systems, highlighting their potential to integrate multi-source data such as real-time reviews, price fluctuations, and location-specific events (Said, \u003cspan citationid=\"CR47\" class=\"CitationRef\"\u003e2025\u003c/span\u003e). For example, studies on hotel hospitality show how ChatGPT-driven recommenders can enhance upselling strategies by tailoring persuasive recommendation messages (Remountakis, Kotis, Kourtzis, \u0026amp; Tsekouras, \u003cspan citationid=\"CR44\" class=\"CitationRef\"\u003e2023\u003c/span\u003e).\u003c/p\u003e\u003cp\u003eHowever, concerns remain around transparency and algorithmic bias in LLM-generated recommendations. Popularity bias and opacity can lead to the disproportionate visibility of high-priced or highly reviewed properties, leading to market inequalities (Hasan et al., \u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e2024\u003c/span\u003e; Dang \u0026amp; Li, \u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e2025\u003c/span\u003e). These issues align with broader challenges in explainable AI (XAI), where a lack of interpretability reduces user trust (de Cerqueira et al., \u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e2025\u003c/span\u003e).\u003c/p\u003e\u003cp\u003eIn tourism, trust is central to recommendation acceptance. Explainable designs, such as those implemented in ExplainableTrip, demonstrate how providing clear rationales (e.g., \u0026ldquo;this hotel matches your budget and has a 4.8 cleanliness score\u0026rdquo;) improves user engagement (Chitaliya et al., \u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e2025\u003c/span\u003e). Without such mechanisms, LLM recommendations risk being perceived as opaque and unreliable.\u003c/p\u003e\u003cp\u003eEthical considerations are equally critical. Scholars argue that the responsible adoption of AI-driven recommendation systems must prioritize fairness, accountability, and transparency to prevent manipulative or biased practices (Hudders et al., 2021; Nouriinanloo, 2024). Ensuring these safeguards is key to integrating LLM recommenders effectively in tourism and hospitality.\u003c/p\u003e\u003cp\u003eLLMs can deliver highly personalized recommendations, but they may also introduce algorithmic biases, such as popularity bias, by disproportionately favouring well-known options, regardless of suitability (Lichtenberg et al., \u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e2024\u003c/span\u003e; Kowald, \u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e2025\u003c/span\u003e). Similar patterns emerge in healthcare, where both accuracy and fairness are critical: despite their robust performance in clinical tasks, LLMs may propagate demographic biases, adversely affecting certain patient groups (Omar et al., \u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e2025\u003c/span\u003e; Schmidgall et al., \u003cspan citationid=\"CR48\" class=\"CitationRef\"\u003e2024\u003c/span\u003e). Given the limited academic exploration of such issues in tourism, this sector emerges as a crucial domain for investigating LLM accuracy and bias.\u003c/p\u003e\u003c/div\u003e"},{"header":"3. Data \u0026 Methodology","content":"\u003cp\u003eThe study focused on prompts related to hotel recommendations, as results for other aspects of hospitality and tourism varied significantly depending on the geographic location of the IP address. This was verified by the authors using a VPN to simulate different user locations across all continents. In contrast, responses to accommodation-related queries remained consistent across locations, suggesting greater stability in that content domain.\u003c/p\u003e\u003cp\u003eTo ensure neutrality, the prompt selected was: \u003cem\u003e\u0026ldquo;Which hotels would you recommend in [location]?\u0026rdquo;.\u003c/em\u003e This query was posed to the five AI chatbots: Claude 3.5 Sonnet, ChatGPT-4o, Gemini 1.5 Flash, DeepThink 1.5, and Grok 3. In order to minimize potential bias in the responses, no user history or personalized settings were active on any of the platforms at the time of testing. The sample included the top 10 most-visited cities worldwide (Istanbul, London, Dubai, Antalya, Paris, Hong Kong, Bangkok, New York, Cancun, and Mecca), as identified from Enterat.com (Enternat, \u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e2024\u003c/span\u003e). The question was submitted to the first three AI chatbots (ChatGPT-4o, Gemini 1.5 Flash, and Claude 3.5 Sonnet) in December 2024 and to the other two (DeepThink R1 and Grok 3) in February 2025.\u003c/p\u003e\u003cp\u003eAfter receiving the initial list of hotels, a follow-up prompt was submitted to each model: \u0026ldquo;Can you create a table including the following characteristics for each hotel: star rating, the average price per night for a stay from 6 to 7 September for two people, the Booking.com score, and the TripAdvisor score?\u0026rdquo;. To standardize the analysis, the average price per night for the recommended hotels was calculated for a weekend stay, for a double room, specifically from 6 September to 7 September 2025. This timeframe was chosen to reflect typical low-season demand period, offering a consistent basis for price comparison across all cities and chatbots.\u003c/p\u003e\u003cp\u003eSubsequently, verification was conducted manually between January and February 2025 to confirm the accuracy of the hotel information provided by the AI chatbots. This verification involved ensuring that the recommended hotels existed, matched the descriptions given by the chatbots, and provided legitimate booking options.\u003c/p\u003e\u003cp\u003e\u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab1\" border=\"1\"\u003e\u003ccaption language=\"En\"\u003e\u003cdiv class=\"CaptionNumber\"\u003eTable 1\u003c/div\u003e\u003cdiv class=\"CaptionContent\"\u003e\u003cp\u003eNumber of Hotel Recommendations by AI chatbots Across the Top 10 Most-Visited Cities\u003c/p\u003e\u003c/div\u003e\u003c/caption\u003e\u003ccolgroup cols=\"6\"\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e\u003cthead\u003e\u003ctr\u003e\u003cth align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/th\u003e\u003cth align=\"left\" colname=\"c2\"\u003e\u003cp\u003eChatGPT 4.o\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c3\"\u003e\u003cp\u003eGemini 1.5 Flash\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c4\"\u003e\u003cp\u003eClaude 3.5 Sonnet\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c5\"\u003e\u003cp\u003eDeepThink R1\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c6\"\u003e\u003cp\u003eGrok 3\u003c/p\u003e\u003c/th\u003e\u003c/tr\u003e\u003c/thead\u003e\u003ctbody\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eANTALYA\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e7\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e10\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e12\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e15\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e\u003cp\u003e12\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eBANGKOK\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e10\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e10\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e12\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e15\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e\u003cp\u003e12\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eCANCUN\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e7\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e10\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e12\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e9\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e\u003cp\u003e12\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eDUBAI\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e9\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e10\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e12\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e15\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e\u003cp\u003e12\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eHONG KONG\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e10\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e10\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e12\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e9\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e\u003cp\u003e12\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eISTAMBUL\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e7\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e10\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e12\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e15\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e\u003cp\u003e12\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eMECCA\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e7\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e5\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e12\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e7\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e\u003cp\u003e12\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eLONDON\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e9\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e10\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e12\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e15\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e\u003cp\u003e12\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eNYC\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e10\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e5\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e12\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e16\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e\u003cp\u003e12\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003ePARIS\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e9\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e9\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e12\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e15\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e\u003cp\u003e12\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eTotal\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e85\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e89\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e120\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e131\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e\u003cp\u003e120\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003c/tbody\u003e\u003c/colgroup\u003e\u003c/table\u003e\u003c/div\u003e\u003c/p\u003e\u003cp\u003eTable\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e presents the number of hotel recommendations provided by different AI chatbots across the 10 most-visited cities worldwide. DeepThink R1 stands out as the most generous model, offering 131 recommendations, followed by Claude 3.5 Sonnet and Grok 3, each with 120 suggestions. Gemini 1.5 Flash provides 89 recommendations, while ChatGPT-4o suggests the fewest, with 85. These differences suggest that each AI model follows a different logic when generating hotel recommendations, leading to varying outputs even when provided with the same input. Figures\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e, \u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e, \u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003e, \u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003e and \u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003e are screenshots of the output given by each of the AI tools.\u003c/p\u003e\u003cp\u003eIt is worth noting that among the five AI tools evaluated, only ChatGPT and Grok provided direct links to the external sources used to generate their recommendations, offering greater transparency regarding the origin of the information. The other chatbots did not disclose the sources underlying their outputs. Table\u0026nbsp;\u003cspan refid=\"Tab2\" class=\"InternalRef\"\u003e2\u003c/span\u003e shows that Grok 3 provided more external source links (196) compared to ChatGPT-4o (125), with notable differences in categories such as news websites (Grok: 40 vs. ChatGPT: 9) and social media (Grok: 18, not present in ChatGPT). ChatGPT primarily referenced travel websites (78) and relied more greatly on established platforms, with companies mentioned 10 or more times including TripAdvisor (34), Hotels.com (22), Conde Nast Traveler (15), and Booking.com (13). In contrast, Grok\u0026rsquo;s most frequent references (10 each) were travelandleisure.com and cntraveler.com, indicating a more diversified but less platform-centric link pattern.\u003c/p\u003e\u003cp\u003e\u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab2\" border=\"1\"\u003e\u003ccaption language=\"En\"\u003e\u003cdiv class=\"CaptionNumber\"\u003eTable 2\u003c/div\u003e\u003cdiv class=\"CaptionContent\"\u003e\u003cp\u003eCategorized external sources linked by ChatGPT-4o and Grok 3 across 10 hotel recommendation queries\u003c/p\u003e\u003c/div\u003e\u003c/caption\u003e\u003ccolgroup cols=\"3\"\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e\u003cthead\u003e\u003ctr\u003e\u003cth align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/th\u003e\u003cth align=\"left\" colname=\"c2\"\u003e\u003cp\u003eChatGPT-4o\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c3\"\u003e\u003cp\u003eGrok 3\u003c/p\u003e\u003c/th\u003e\u003c/tr\u003e\u003c/thead\u003e\u003ctbody\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eTravel website\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e78\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e76\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eNews website\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e9\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e40\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eTravel magazine\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e20\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e32\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eSocial media\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e-\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e18\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eLifestyle blog\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e12\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e10\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eLuxury travel website\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e-\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e8\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eLuxury hotel website\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e4\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e7\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eLuxury travel blog\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e2\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e5\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eTotal\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e125\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e196\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003c/tbody\u003e\u003c/colgroup\u003e\u003c/table\u003e\u003c/div\u003e\u003c/p\u003e\u003cp\u003eTo analyse the alignment between AI-generated hotel recommendations and real-world data, two complementary statistical techniques were applied. First, Pearson correlation coefficients were calculated between each pair of variables (AI-supplied and real values) (Sedgwick, \u003cspan citationid=\"CR49\" class=\"CitationRef\"\u003e2012\u003c/span\u003e) or hotel category shown by stars from 1 to 5, average prices per night, Booking.com scores, and TripAdvisor ratings. This allowed for the identification of directional relationships and the level of agreement between AI outputs and verified data. Second, paired-sample T-tests were conducted to assess whether the mean differences between AI-generated and real values were statistically significant (Afifah et al., \u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e2022\u003c/span\u003e). These tests provide insights into potential systematic biases introduced by each AI tool. Additionally, to compare the real values for hotels recommended by different AI chatbots, Welch ANOVA tests were conducted to detect intergroup differences, followed by Games-Howell or Bonferroni post-hoc tests, depending on variable characteristics, to identify which tools significantly differed from each other (Agbangba et al., \u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2024\u003c/span\u003e; Liu, \u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e2015\u003c/span\u003e). This multi-step statistical approach ensures a robust and nuanced evaluation of the performance and biases of each generative AI chatbot.\u003c/p\u003e"},{"header":"4. Results","content":"\u003cp\u003eTo assess the reliability and potential biases in hotel recommendations generated by AI chatbots, we conducted a detailed statistical analysis comparing the outputs of each model with real-world hotel data.\u003c/p\u003e\u003cp\u003eThe correlation between the real variable and the AI variable for each of the items analysed (hotel category, price, score on Booking.com (BK) and score on TripAdvisor (TA)was carried out to find out the extent to which both variables of the pair were correlated. Moreover, we wished to analyse how different the means were, so a T-test was carried out for each pair of variables, within each AI tool.\u003c/p\u003e\u003cp\u003eIt could be that the correlation between the pairs is strong and significant, which would mean that both variables (AI and real) go in the same direction (when one increases the other also increases) and thus there is not much bias. It is important to note that not all AI-generated outputs contained complete information for all variables (e.g., some recommendations lacked price or review scores), resulting in slight variations in the sample size (N) across different tests.\u003c/p\u003e\u003cp\u003e\u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab3\" border=\"1\"\u003e\u003ccaption language=\"En\"\u003e\u003cdiv class=\"CaptionNumber\"\u003eTable 3\u003c/div\u003e\u003cdiv class=\"CaptionContent\"\u003e\u003cp\u003ePaired samples for the ChatGPT-4o model\u003c/p\u003e\u003c/div\u003e\u003c/caption\u003e\u003ccolgroup cols=\"6\"\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e\u003cthead\u003e\u003ctr\u003e\u003cth align=\"left\" colname=\"c1\"\u003e\u003cp\u003ePairs\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c2\"\u003e\u003cp\u003eN\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c3\"\u003e\u003cp\u003eMean\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c4\"\u003e\u003cp\u003eStd. Dev.\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c5\"\u003e\u003cp\u003ePaired samples Corr.\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c6\"\u003e\u003cp\u003ePaired mean diff.\u003c/p\u003e\u003c/th\u003e\u003c/tr\u003e\u003c/thead\u003e\u003ctbody\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eAI stars\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e81\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e4.58\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.69\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\" morerows=\"1\" rowspan=\"2\"\u003e\u003cp\u003e0.073\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\" morerows=\"1\" rowspan=\"2\"\u003e\u003cp\u003e-0.383*\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eReal stars\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e81\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e4.96\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.19\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eAI price\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e82\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e609.09\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e393.81\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\" morerows=\"1\" rowspan=\"2\"\u003e\u003cp\u003e0.408*\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\" morerows=\"1\" rowspan=\"2\"\u003e\u003cp\u003e-318.98*\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eReal price\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e82\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e928.06\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e828.55\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eAI BK score\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e81\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e8.78\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.68\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\" morerows=\"1\" rowspan=\"2\"\u003e\u003cp\u003e0.263**\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\" morerows=\"1\" rowspan=\"2\"\u003e\u003cp\u003e-0.23**\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eReal BK score\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e81\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e9.01\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.39\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eAI TA score\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e83\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e4.41\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.35\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\" morerows=\"1\" rowspan=\"2\"\u003e\u003cp\u003e0.236**\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\" morerows=\"1\" rowspan=\"2\"\u003e\u003cp\u003e-0.25*\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eReal TA score\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e83\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e4.66\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.27\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003c/tbody\u003e\u003c/colgroup\u003e\u003ctfoot\u003e\u003ctr\u003e\u003ctd colspan=\"6\"\u003e* sig. \u0026lt;0.001; ** sig. \u0026lt; 0.05; Price in \u0026euro;\u003c/td\u003e\u003c/tr\u003e\u003c/tfoot\u003e\u003c/table\u003e\u003c/div\u003e\u003c/p\u003e\u003cp\u003eAs shown in Table\u0026nbsp;\u003cspan refid=\"Tab3\" class=\"InternalRef\"\u003e3\u003c/span\u003e, ChatGPT recommends hotels with lower values than they actually score, for all four variables: stars, price, and Booking.com and TripAdvisor scores. Only in the case of the Price variable is correlation statistically significant (40.8%; p-value\u0026thinsp;\u0026lt;\u0026thinsp;0.001) meaning that when the price given by AI increases, the real price also increases. For the other variables, there is no significant correlation between what is suggested by ChatGPT and the real data. Given that the prices analysed were static for a fixed date, the observed correlation does not indicate price increases over time but rather reflects logical consistency between the AI-generated outputs and real hotel price data.\u003c/p\u003e\u003cp\u003eConsidering the mean difference, it can be observed that in all four pairs, the difference between pairs of variables is statistically significant (at p-value\u0026thinsp;\u0026lt;\u0026thinsp;0.001 or at p-value\u0026thinsp;\u0026lt;\u0026thinsp;0.05). Therefore, it seems there is a degree of bias between the data provided by ChatGPT and the real data.\u003c/p\u003e\u003cp\u003e\u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab4\" border=\"1\"\u003e\u003ccaption language=\"En\"\u003e\u003cdiv class=\"CaptionNumber\"\u003eTable 4\u003c/div\u003e\u003cdiv class=\"CaptionContent\"\u003e\u003cp\u003ePaired samples for the Gemini model\u003c/p\u003e\u003c/div\u003e\u003c/caption\u003e\u003ccolgroup cols=\"6\"\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e\u003cthead\u003e\u003ctr\u003e\u003cth align=\"left\" colname=\"c1\"\u003e\u003cp\u003ePairs\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c2\"\u003e\u003cp\u003eN\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c3\"\u003e\u003cp\u003eMean\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c4\"\u003e\u003cp\u003eStd. Dev.\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c5\"\u003e\u003cp\u003ePaired samples Corr.\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c6\"\u003e\u003cp\u003ePaired mean diff.\u003c/p\u003e\u003c/th\u003e\u003c/tr\u003e\u003c/thead\u003e\u003ctbody\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eAI stars\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e84\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e4.67\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.84\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\" morerows=\"1\" rowspan=\"2\"\u003e\u003cp\u003e0.851*\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\" morerows=\"1\" rowspan=\"2\"\u003e\u003cp\u003e0.012\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eReal stars\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e84\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e4.65\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.89\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eAI price\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e80\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e424.69\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e198.87\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\" morerows=\"1\" rowspan=\"2\"\u003e\u003cp\u003e-0.098\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\" morerows=\"1\" rowspan=\"2\"\u003e\u003cp\u003e-686,51**\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eReal price\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e80\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e1111.20\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e2990.32\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eAI BK score\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e84\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e8.47\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e1.33\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\" morerows=\"1\" rowspan=\"2\"\u003e\u003cp\u003e0.675*\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\" morerows=\"1\" rowspan=\"2\"\u003e\u003cp\u003e-0.24**\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eReal BK score\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e84\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e8.71\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.92\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eAI TA score\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e84\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e4.29\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.52\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\" morerows=\"1\" rowspan=\"2\"\u003e\u003cp\u003e0.364*\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\" morerows=\"1\" rowspan=\"2\"\u003e\u003cp\u003e-0.18**\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eReal TA score\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e84\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e4.47\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.57\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003c/tbody\u003e\u003c/colgroup\u003e\u003ctfoot\u003e\u003ctr\u003e\u003ctd colspan=\"6\"\u003e* sig. \u0026lt;0.001; ** sig. \u0026lt; 0.05; Price in \u0026euro;\u003c/td\u003e\u003c/tr\u003e\u003c/tfoot\u003e\u003c/table\u003e\u003c/div\u003e\u003c/p\u003e\u003cp\u003eGemini (Table\u0026nbsp;\u003cspan refid=\"Tab4\" class=\"InternalRef\"\u003e4\u003c/span\u003e) suggests hotels with values closer to reality, regarding stars and scores on Booking.com and TripAdvisor. However, regarding prices, Gemini recommends much lower ones (\u0026euro;424 vs \u0026euro;1,111). We also computed the correlation between what AI suggested and real data, and it can be observed that for the stars variable, there is a strong correlation (85.1%; p\u0026thinsp;\u0026lt;\u0026thinsp;0.001). We can also find statistically significant correlations between Gemini and the reality for Booking.com and TripAdvisor scores. As already pointed out, these results mean that there is not a significant bias between real and AI-generated data.\u003c/p\u003e\u003cp\u003eRegarding the paired mean differences, those of price, Booking.com score and TripAdvisor score are significant but at 0.05.\u003c/p\u003e\u003cp\u003e\u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab5\" border=\"1\"\u003e\u003ccaption language=\"En\"\u003e\u003cdiv class=\"CaptionNumber\"\u003eTable 5\u003c/div\u003e\u003cdiv class=\"CaptionContent\"\u003e\u003cp\u003ePaired samples for the Claude model\u003c/p\u003e\u003c/div\u003e\u003c/caption\u003e\u003ccolgroup cols=\"6\"\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e\u003cthead\u003e\u003ctr\u003e\u003cth align=\"left\" colname=\"c1\"\u003e\u003cp\u003ePairs\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c2\"\u003e\u003cp\u003eN\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c3\"\u003e\u003cp\u003eMean\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c4\"\u003e\u003cp\u003eStd. Dev.\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c5\"\u003e\u003cp\u003ePaired samples Corr.\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c6\"\u003e\u003cp\u003ePaired mean diff.\u003c/p\u003e\u003c/th\u003e\u003c/tr\u003e\u003c/thead\u003e\u003ctbody\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eAI stars\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e113\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e3.81\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.96\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\" morerows=\"1\" rowspan=\"2\"\u003e\u003cp\u003e0.672*\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\" morerows=\"1\" rowspan=\"2\"\u003e\u003cp\u003e-0.566*\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eReal stars\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e113\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e4.38\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.86\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eAI price\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e114\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e349.63\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e325.66\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\" morerows=\"1\" rowspan=\"2\"\u003e\u003cp\u003e0.465*\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\" morerows=\"1\" rowspan=\"2\"\u003e\u003cp\u003e-211.736**\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eReal price\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e114\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e561.37\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e875.47\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eAI BK score\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e111\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e8.57\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.53\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\" morerows=\"1\" rowspan=\"2\"\u003e\u003cp\u003e0.564*\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\" morerows=\"1\" rowspan=\"2\"\u003e\u003cp\u003e0.241*\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eReal BK score\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e111\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e8.33\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.93\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eAI TA score\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e117\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e4.16\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.28\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\" morerows=\"1\" rowspan=\"2\"\u003e\u003cp\u003e0.470*\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\" morerows=\"1\" rowspan=\"2\"\u003e\u003cp\u003e-0.247*\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eReal TA score\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e117\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e4.41\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.51\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003c/tbody\u003e\u003c/colgroup\u003e\u003ctfoot\u003e\u003ctr\u003e\u003ctd colspan=\"6\"\u003e* sig. \u0026lt;0.001; ** sig. \u0026lt;0.05 Price in \u0026euro;\u003c/td\u003e\u003c/tr\u003e\u003c/tfoot\u003e\u003c/table\u003e\u003c/div\u003e\u003c/p\u003e\u003cp\u003eClaude (Table\u0026nbsp;\u003cspan refid=\"Tab5\" class=\"InternalRef\"\u003e5\u003c/span\u003e) shows a statistically significant correlation between the tool and the reality for all four variables. This means that when Claude data increases, the real data also increases. However, when considering the paired mean difference, we find that the difference in the case of stars, and the differences in the case of Booking.com and TripAdvisor scores, are significant (p-value\u0026thinsp;\u0026lt;\u0026thinsp;0.001).\u003c/p\u003e\u003cp\u003e\u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab6\" border=\"1\"\u003e\u003ccaption language=\"En\"\u003e\u003cdiv class=\"CaptionNumber\"\u003eTable 6\u003c/div\u003e\u003cdiv class=\"CaptionContent\"\u003e\u003cp\u003ePaired samples for the Grok model\u003c/p\u003e\u003c/div\u003e\u003c/caption\u003e\u003ccolgroup cols=\"6\"\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e\u003cthead\u003e\u003ctr\u003e\u003cth align=\"left\" colname=\"c1\"\u003e\u003cp\u003ePairs\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c2\"\u003e\u003cp\u003eN\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c3\"\u003e\u003cp\u003eMean\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c4\"\u003e\u003cp\u003eStd. Dev.\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c5\"\u003e\u003cp\u003ePaired samples Corr.\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c6\"\u003e\u003cp\u003ePaired mean diff.\u003c/p\u003e\u003c/th\u003e\u003c/tr\u003e\u003c/thead\u003e\u003ctbody\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eAI stars\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e119\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e4.33\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.82\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\" morerows=\"1\" rowspan=\"2\"\u003e\u003cp\u003e0.789*\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\" morerows=\"1\" rowspan=\"2\"\u003e\u003cp\u003e0.067\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eReal stars\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e119\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e4.26\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e1.19\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eAI price\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e115\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e259.63\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e211.40\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\" morerows=\"1\" rowspan=\"2\"\u003e\u003cp\u003e0.500*\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\" morerows=\"1\" rowspan=\"2\"\u003e\u003cp\u003e-392.770*\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eReal price\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e115\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e652.40\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e692.34\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eAI BK score\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e117\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e8.84\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.39\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\" morerows=\"1\" rowspan=\"2\"\u003e\u003cp\u003e0.600*\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\" morerows=\"1\" rowspan=\"2\"\u003e\u003cp\u003e0.174\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eReal BK score\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e117\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e8.67\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e1.09\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eAI TA score\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e117\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e4.33\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.25\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\" morerows=\"1\" rowspan=\"2\"\u003e\u003cp\u003e0.372*\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\" morerows=\"1\" rowspan=\"2\"\u003e\u003cp\u003e-0.188*\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eReal TA score\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e117\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e4.52\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.47\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003c/tbody\u003e\u003c/colgroup\u003e\u003ctfoot\u003e\u003ctr\u003e\u003ctd colspan=\"6\"\u003e* sig. \u0026lt;0.001; ** sig. \u0026lt;0.05 Price in \u0026euro;\u003c/td\u003e\u003c/tr\u003e\u003c/tfoot\u003e\u003c/table\u003e\u003c/div\u003e\u003c/p\u003e\u003cp\u003eWithin the Grok model, it can be seen (Table\u0026nbsp;\u003cspan refid=\"Tab6\" class=\"InternalRef\"\u003e6\u003c/span\u003e) that correlations are all statistically significant and strong, especially for hotel category. Thus, real data and data provided by the tool go in the same direction. Therefore, in this case, the difference in means for the hotel category variable and for the Booking.com score is not significant. In the case of average price per night in a double room from 6 to 7 September, Grok data supplies much lower prices than they actually are.\u003c/p\u003e\u003cp\u003e\u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab7\" border=\"1\"\u003e\u003ccaption language=\"En\"\u003e\u003cdiv class=\"CaptionNumber\"\u003eTable 7\u003c/div\u003e\u003cdiv class=\"CaptionContent\"\u003e\u003cp\u003ePaired samples for the DeepSeek model\u003c/p\u003e\u003c/div\u003e\u003c/caption\u003e\u003ccolgroup cols=\"6\"\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e\u003cthead\u003e\u003ctr\u003e\u003cth align=\"left\" colname=\"c1\"\u003e\u003cp\u003ePairs\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c2\"\u003e\u003cp\u003eN\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c3\"\u003e\u003cp\u003eMean\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c4\"\u003e\u003cp\u003eStd. Dev.\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c5\"\u003e\u003cp\u003ePaired samples Corr.\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c6\"\u003e\u003cp\u003ePaired mean diff.\u003c/p\u003e\u003c/th\u003e\u003c/tr\u003e\u003c/thead\u003e\u003ctbody\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eAI stars\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e124\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e4.28\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.87\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\" morerows=\"1\" rowspan=\"2\"\u003e\u003cp\u003e0.751*\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\" morerows=\"1\" rowspan=\"2\"\u003e\u003cp\u003e0.000\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eReal stars\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e124\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e4.28\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e1.23\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eAI price\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e125\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e410.89\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e309.79\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\" morerows=\"1\" rowspan=\"2\"\u003e\u003cp\u003e0.483*\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\" morerows=\"1\" rowspan=\"2\"\u003e\u003cp\u003e-301.59*\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eReal price\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e125\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e712.48\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e817.06\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eAI BK score\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e125\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e8.936\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.46\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\" morerows=\"1\" rowspan=\"2\"\u003e\u003cp\u003e0.660*\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\" morerows=\"1\" rowspan=\"2\"\u003e\u003cp\u003e0.218*\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eReal BK score\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e125\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e8.718\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.659\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eAI TA score\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e128\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e4.315\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.29\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\" morerows=\"1\" rowspan=\"2\"\u003e\u003cp\u003e0.395*\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\" morerows=\"1\" rowspan=\"2\"\u003e\u003cp\u003e-0.216*\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eReal TA score\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e128\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e4.531\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.36\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003c/tbody\u003e\u003c/colgroup\u003e\u003ctfoot\u003e\u003ctr\u003e\u003ctd colspan=\"6\"\u003e* sig. \u0026lt;0.001; ** sig. \u0026lt;0.05; Price in \u0026euro;\u003c/td\u003e\u003c/tr\u003e\u003c/tfoot\u003e\u003c/table\u003e\u003c/div\u003e\u003c/p\u003e\u003cp\u003eDeepSeek (Table\u0026nbsp;\u003cspan refid=\"Tab7\" class=\"InternalRef\"\u003e7\u003c/span\u003e) also shows statistically significant correlations for all four variables. The stars variable is the strongest (lower bias between real and DeepSeek data), while that of TripAdvisor scores the weakest. Regarding the paired mean difference, it can be observed there is no difference in the case of the stars variable, but there are statistically significant mean differences for the other three variables.\u003c/p\u003e\u003cp\u003e\u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab8\" border=\"1\"\u003e\u003ccaption language=\"En\"\u003e\u003cdiv class=\"CaptionNumber\"\u003eTable 8\u003c/div\u003e\u003cdiv class=\"CaptionContent\"\u003e\u003cp\u003ePaired samples for the entire sample\u003c/p\u003e\u003c/div\u003e\u003c/caption\u003e\u003ccolgroup cols=\"6\"\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e\u003cthead\u003e\u003ctr\u003e\u003cth align=\"left\" colname=\"c1\"\u003e\u003cp\u003ePairs\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c2\"\u003e\u003cp\u003eN\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c3\"\u003e\u003cp\u003eMean\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c4\"\u003e\u003cp\u003eStd. Dev.\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c5\"\u003e\u003cp\u003ePaired samples Corr.\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c6\"\u003e\u003cp\u003ePaired mean diff.\u003c/p\u003e\u003c/th\u003e\u003c/tr\u003e\u003c/thead\u003e\u003ctbody\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eAI stars\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e521\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e4.30\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.89\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\" morerows=\"1\" rowspan=\"2\"\u003e\u003cp\u003e0.693*\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\" morerows=\"1\" rowspan=\"2\"\u003e\u003cp\u003e-0.165*\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eReal stars\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e521\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e4.46\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e1.02\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eAI price\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e516\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e397.28\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e314.27\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\" morerows=\"1\" rowspan=\"2\"\u003e\u003cp\u003e0.226*\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\" morerows=\"1\" rowspan=\"2\"\u003e\u003cp\u003e-364.51*\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eReal price\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e516\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e761.78\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e1396.49\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eAI BK score\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e518\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e8.74\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.73\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\" morerows=\"1\" rowspan=\"2\"\u003e\u003cp\u003e0.510*\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\" morerows=\"1\" rowspan=\"2\"\u003e\u003cp\u003e0.069**\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eReal BK score\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e518\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e8.67\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.89\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eAI TA score\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e529\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e4.29\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.35\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\" morerows=\"1\" rowspan=\"2\"\u003e\u003cp\u003e0.388*\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\" morerows=\"1\" rowspan=\"2\"\u003e\u003cp\u003e-0.218*\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eReal TA score\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e529\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e4.51\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.46\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003c/tbody\u003e\u003c/colgroup\u003e\u003ctfoot\u003e\u003ctr\u003e\u003ctd colspan=\"6\"\u003e* sig. \u0026lt;0.001; ** sig. \u0026lt;0.05; Price in \u0026euro;\u003c/td\u003e\u003c/tr\u003e\u003c/tfoot\u003e\u003c/table\u003e\u003c/div\u003e\u003c/p\u003e\u003cp\u003eConsidering the whole sample of hotels suggested by AI tools (Table\u0026nbsp;\u003cspan refid=\"Tab8\" class=\"InternalRef\"\u003e8\u003c/span\u003e), the results reveal significant correlations between AI tools and real data, but only that of the hotel category variable is the strongest, which would indicate that the bias between AI and real data is lower. Regarding the difference between pairs, that of Booking.com score is almost insignificant, and in fact, the correlation between the two variables is rather strong and significant.\u003c/p\u003e\u003cp\u003eHaving correlated and compared the results between real and AI recommendations, we now draw from the hotels suggested by AI tools but with the real (human) collected data of each variable and compare each variable across the AI tools.\u003c/p\u003e\u003cp\u003e\u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab9\" border=\"1\"\u003e\u003ccaption language=\"En\"\u003e\u003cdiv class=\"CaptionNumber\"\u003eTable 9\u003c/div\u003e\u003cdiv class=\"CaptionContent\"\u003e\u003cp\u003eHotel category stars (actual data) descriptives by AI tool\u003c/p\u003e\u003c/div\u003e\u003c/caption\u003e\u003ccolgroup cols=\"6\"\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e\u003cthead\u003e\u003ctr\u003e\u003cth align=\"left\" colname=\"c1\"\u003e\u003cp\u003eAI tool\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c2\"\u003e\u003cp\u003eN\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c3\"\u003e\u003cp\u003eMean\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c4\"\u003e\u003cp\u003eStd. Dev.\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c5\"\u003e\u003cp\u003eMin.\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c6\"\u003e\u003cp\u003eMax.\u003c/p\u003e\u003c/th\u003e\u003c/tr\u003e\u003c/thead\u003e\u003ctbody\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eChatGPT\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e81\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e4.96\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.190\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e4\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e\u003cp\u003e5\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eGemini\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e84\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e4.65\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.898\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e1\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e\u003cp\u003e5\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eClaude\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e113\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e4.38\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.859\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e1\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e\u003cp\u003e5\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eGrok\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e119\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e4.26\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e1.196\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e0\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e\u003cp\u003e5\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eDeepSeek\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e124\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e4.28\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e1.233\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e0\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e\u003cp\u003e5\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eTotal\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e521\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e4.46\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e1.021\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e0\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e\u003cp\u003e5\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003c/tbody\u003e\u003c/colgroup\u003e\u003c/table\u003e\u003c/div\u003e\u003c/p\u003e\u003cp\u003eTable\u0026nbsp;\u003cspan refid=\"Tab9\" class=\"InternalRef\"\u003e9\u003c/span\u003e shows the descriptives for the real data variable of stars per AI tool. It can be observed that ChatGPT is the AI tool that recommends the highest (4.96) and Grok the lowest (4.26) hotel category. However, in all five tools, the mean number of hotel stars is above 4, meaning that in general, AI tools recommend top-of-the-range hotels. In this case we carried out a robust test of equality of means with Welch statistics. The results show there is a statistically significant difference between AI tool (groups) means. For the hotel category variable, we then carried out Games-Howell post-hoc test to compare the means of each AI tool with the others, and it seems that ChatGPT is the only AI tool to differ significantly from the others (Table\u0026nbsp;\u003cspan refid=\"Tab10\" class=\"InternalRef\"\u003e10\u003c/span\u003e).\u003c/p\u003e\u003cp\u003e\u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab10\" border=\"1\"\u003e\u003ccaption language=\"En\"\u003e\u003cdiv class=\"CaptionNumber\"\u003eTable 10\u003c/div\u003e\u003cdiv class=\"CaptionContent\"\u003e\u003cp\u003eMultiple comparison per variable and AI tool\u003c/p\u003e\u003c/div\u003e\u003c/caption\u003e\u003ccolgroup cols=\"10\"\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c7\" colnum=\"7\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c8\" colnum=\"8\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c9\" colnum=\"9\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c10\" colnum=\"10\"\u003e\u003c/div\u003e\u003cthead\u003e\u003ctr\u003e\u003cth align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/th\u003e\u003cth align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/th\u003e\u003cth align=\"left\" colspan=\"2\" nameend=\"c4\" namest=\"c3\"\u003e\u003cp\u003eGames-Howell test (stars dependent variable)\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colspan=\"2\" nameend=\"c6\" namest=\"c5\"\u003e\u003cp\u003eBonferroni test (price dependent variable)\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colspan=\"2\" nameend=\"c8\" namest=\"c7\"\u003e\u003cp\u003eGames-Howell test (Booking.com score dependent variable)\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colspan=\"2\" nameend=\"c10\" namest=\"c9\"\u003e\u003cp\u003eGames-Howell test (TripAdvisor score dependent variable)\u003c/p\u003e\u003c/th\u003e\u003c/tr\u003e\u003c/thead\u003e\u003ctbody\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eName tool (i)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eName tool (j)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003eMean\u003c/p\u003e\u003cp\u003edifference\u003c/p\u003e\u003cp\u003e(i-j)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003eStd.\u003c/p\u003e\u003cp\u003eError\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003eMean\u003c/p\u003e\u003cp\u003edifference\u003c/p\u003e\u003cp\u003e(i-j)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003eStd.\u003c/p\u003e\u003cp\u003eError\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003eMean\u003c/p\u003e\u003cp\u003edifference\u003c/p\u003e\u003cp\u003e(i-j)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c8\"\u003e\u003cp\u003eStd.\u003c/p\u003e\u003cp\u003eError\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c9\"\u003e\u003cp\u003eMean difference\u003c/p\u003e\u003cp\u003e(i-j)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c10\"\u003e\u003cp\u003eStd.\u003c/p\u003e\u003cp\u003eError\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\" morerows=\"3\" rowspan=\"4\"\u003e\u003cp\u003eChatGPT\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eGemini\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e0.308**\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e0.100\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e183.518*\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e44.747\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e0.296\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c8\"\u003e\u003cp\u003e0.109\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c9\"\u003e\u003cp\u003e0.1924**\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c10\"\u003e\u003cp\u003e0.069\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eClaude\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e0.582*\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e0.084\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e259.039*\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e41.595\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e0.673*\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c8\"\u003e\u003cp\u003e0.098\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c9\"\u003e\u003cp\u003e0.2524*\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c10\"\u003e\u003cp\u003e0.056\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eGrok\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e0.702*\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e0.112\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e343.573*\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e41.596\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e0.334**\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c8\"\u003e\u003cp\u003e0.109\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c9\"\u003e\u003cp\u003e0.1413\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c10\"\u003e\u003cp\u003e0.053\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eDeepSeek\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e0.681*\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e0.113\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e197.886*\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e40.865\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e0.289*\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c8\"\u003e\u003cp\u003e0.073\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c9\"\u003e\u003cp\u003e0.1314*\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c10\"\u003e\u003cp\u003e0.044\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\" morerows=\"3\" rowspan=\"4\"\u003e\u003cp\u003eGemini\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eChatGPT\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e-0.308**\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e0.100\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e-183.518*\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e44.747\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e-0.296\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c8\"\u003e\u003cp\u003e0.109\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c9\"\u003e\u003cp\u003e-0.1924**\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c10\"\u003e\u003cp\u003e0.069\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eClaude\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e0.274\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e0.127\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e75.521\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e41.314\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e0.377**\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c8\"\u003e\u003cp\u003e0.133\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c9\"\u003e\u003cp\u003e0.0600\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c10\"\u003e\u003cp\u003e0.078\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eGrok\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e0.394\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e0.147\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e160.056*\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e41.315\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e0.042\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c8\"\u003e\u003cp\u003e0.142\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c9\"\u003e\u003cp\u003e-0.0511\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c10\"\u003e\u003cp\u003e0.076\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eDeepSeek\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e0.373\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e0.148\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e14.368\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e40.579\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e-0.007\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c8\"\u003e\u003cp\u003e0.116\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c9\"\u003e\u003cp\u003e-0.0610\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c10\"\u003e\u003cp\u003e0.070\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\" morerows=\"3\" rowspan=\"4\"\u003e\u003cp\u003eClaude\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eChatGPT\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e-0.582*\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e0.084\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e-259.039*\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e41.596\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e-0.673*\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c8\"\u003e\u003cp\u003e0.098\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c9\"\u003e\u003cp\u003e\u0026minus;\u0026thinsp;.02524*\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c10\"\u003e\u003cp\u003e0.056\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eGemini\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e-0.274\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e0.127\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e-75.522\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e41.314\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e-0.377**\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c8\"\u003e\u003cp\u003e0.133\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c9\"\u003e\u003cp\u003e-0.0600\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c10\"\u003e\u003cp\u003e0.078\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eGrok\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e0.120\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e0.136\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e84.534\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e37.878\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e-0.335\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c8\"\u003e\u003cp\u003e0.134\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c9\"\u003e\u003cp\u003e-0.1111\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c10\"\u003e\u003cp\u003e0.064\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eDeepSeek\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e0.098\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e0.137\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e-61.153\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e37.075\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e-0.384**\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c8\"\u003e\u003cp\u003e0.106\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c9\"\u003e\u003cp\u003e-0.1210\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c10\"\u003e\u003cp\u003e0.057\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\" morerows=\"3\" rowspan=\"4\"\u003e\u003cp\u003eGrok\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eChatGPT\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e-0.702*\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e0.112\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e-343.573*\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e41.596\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e-0.338**\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c8\"\u003e\u003cp\u003e0.109\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c9\"\u003e\u003cp\u003e-0.1413\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c10\"\u003e\u003cp\u003e0.053\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eGemini\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e-0.394\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e0.147\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e-160.056*\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e41.315\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e-0.042\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c8\"\u003e\u003cp\u003e0.142\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c9\"\u003e\u003cp\u003e0.0511\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c10\"\u003e\u003cp\u003e0.076\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eClaude\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e-0.120\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e0.136\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e-84.534\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e37.878\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e.0335\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c8\"\u003e\u003cp\u003e0.134\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c9\"\u003e\u003cp\u003e0.1111\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c10\"\u003e\u003cp\u003e0.064\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eDeepSeek\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e-0.022\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e0.156\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e-145.687*\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e37.075\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e-0.049\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c8\"\u003e\u003cp\u003e0.117\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c9\"\u003e\u003cp\u003e-0.0099\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c10\"\u003e\u003cp\u003e0.054\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\" morerows=\"3\" rowspan=\"4\"\u003e\u003cp\u003eDeepSeek\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eChatGPT\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e-0.681*\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e0.113\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e-197.886*\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e40.865\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e-0.289*\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c8\"\u003e\u003cp\u003e0.073\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c9\"\u003e\u003cp\u003e-0.1314**\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c10\"\u003e\u003cp\u003e0.044\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eGemini\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e-0.373\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e0.148\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e-14.368\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e40.579\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e0.007\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c8\"\u003e\u003cp\u003e0.116\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c9\"\u003e\u003cp\u003e0.0610\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c10\"\u003e\u003cp\u003e0.070\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eClaude\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e-0.098\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e0.137\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e61.153\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e37.075\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e0.384**\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c8\"\u003e\u003cp\u003e0.106\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c9\"\u003e\u003cp\u003e0.1210\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c10\"\u003e\u003cp\u003e0.057\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eGrok\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e0.022\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e0.156\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e145.687*\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e37.075\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e0.049\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c8\"\u003e\u003cp\u003e0.117\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c9\"\u003e\u003cp\u003e0.0099\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c10\"\u003e\u003cp\u003e0.054\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003c/tbody\u003e\u003c/colgroup\u003e\u003ctfoot\u003e\u003ctr\u003e\u003ctd colspan=\"10\"\u003e* sig: \u0026lt;0.001; ** sig: \u0026lt;0.05\u003c/td\u003e\u003c/tr\u003e\u003c/tfoot\u003e\u003c/table\u003e\u003c/div\u003e\u003c/p\u003e\u003cp\u003e\u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab11\" border=\"1\"\u003e\u003ccaption language=\"En\"\u003e\u003cdiv class=\"CaptionNumber\"\u003eTable 11\u003c/div\u003e\u003cdiv class=\"CaptionContent\"\u003e\u003cp\u003ePRICE descriptives by AI tool\u003c/p\u003e\u003c/div\u003e\u003c/caption\u003e\u003ccolgroup cols=\"6\"\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e\u003cthead\u003e\u003ctr\u003e\u003cth align=\"left\" colname=\"c1\"\u003e\u003cp\u003eAI tool\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c2\"\u003e\u003cp\u003eN\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c3\"\u003e\u003cp\u003eMean\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c4\"\u003e\u003cp\u003eStd. Dev.\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c5\"\u003e\u003cp\u003eMin.\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c6\"\u003e\u003cp\u003eMax.\u003c/p\u003e\u003c/th\u003e\u003c/tr\u003e\u003c/thead\u003e\u003ctbody\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eChatGPT\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e82\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e928.06\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e828.551\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e147\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e\u003cp\u003e4047\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eGemini\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e80\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e1111.20\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e2990.316\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e57\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e\u003cp\u003e26849\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eClaude\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e114\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e561.37\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e875.472\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e21\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e\u003cp\u003e7625\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eGrok\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e115\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e652.40\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e692.342\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e23\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e\u003cp\u003e4480\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eDeepSeek\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e125\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e712.48\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e817.062\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e30\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e\u003cp\u003e4480\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eTotal\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e516\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e761.78\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e1396.495\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e21\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e\u003cp\u003e26849\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003c/tbody\u003e\u003c/colgroup\u003e\u003ctfoot\u003e\u003ctr\u003e\u003ctd colspan=\"6\"\u003ePrice in \u0026euro;\u003c/td\u003e\u003c/tr\u003e\u003c/tfoot\u003e\u003c/table\u003e\u003c/div\u003e\u003c/p\u003e\u003cp\u003eRegarding the price variable (Table\u0026nbsp;\u003cspan refid=\"Tab11\" class=\"InternalRef\"\u003e11\u003c/span\u003e), Gemini recommends much more expensive hotels than the other tools (mean \u0026euro;1,111.20), and Claude suggests the lowest priced establishments (mean \u0026euro;561.37). In this case, for the price variable, we carried out a Bonferroni test to analyse the significance of price mean differences between AI tools, as the assumption of homogeneity of variances was confirmed by Levene\u0026rsquo;s test.\u003c/p\u003e\u003cp\u003eRegarding the multiple comparison of the price variable, Table\u0026nbsp;\u003cspan refid=\"Tab10\" class=\"InternalRef\"\u003e10\u003c/span\u003e shows that the differences are statistically significant for the majority of tools (groups). As in the case of the stars variable, ChatGPT also shows significant differences from all the other AI tools. In this case, however, we applied the Games-Howell test due to the violation of the homogeneity of variances, as indicated by Levene\u0026rsquo;s test.\u003c/p\u003e\u003cp\u003e\u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab12\" border=\"1\"\u003e\u003ccaption language=\"En\"\u003e\u003cdiv class=\"CaptionNumber\"\u003eTable 12\u003c/div\u003e\u003cdiv class=\"CaptionContent\"\u003e\u003cp\u003eBooking.com score descriptives by AI tool\u003c/p\u003e\u003c/div\u003e\u003c/caption\u003e\u003ccolgroup cols=\"6\"\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e\u003cthead\u003e\u003ctr\u003e\u003cth align=\"left\" colname=\"c1\"\u003e\u003cp\u003eAI tool\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c2\"\u003e\u003cp\u003eN\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c3\"\u003e\u003cp\u003eMean\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c4\"\u003e\u003cp\u003eStd. Dev.\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c5\"\u003e\u003cp\u003eMin.\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c6\"\u003e\u003cp\u003eMax.\u003c/p\u003e\u003c/th\u003e\u003c/tr\u003e\u003c/thead\u003e\u003ctbody\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eChatGPT\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e81\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e9.006\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.386\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e7.90\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e\u003cp\u003e10.00\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eGemini\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e84\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e8.710\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.919\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e5.50\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e\u003cp\u003e9.70\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eClaude\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e111\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e8.333\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.927\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e3.60\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e\u003cp\u003e9.70\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eGrok\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e117\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e8.668\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e1.086\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e0.00\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e\u003cp\u003e9.70\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eDeepSeek\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e125\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e8.718\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.659\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e6.20\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e\u003cp\u003e9.70\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eTotal\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e518\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e8.668\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.868\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e0.00\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e\u003cp\u003e10.00\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003c/tbody\u003e\u003c/colgroup\u003e\u003c/table\u003e\u003c/div\u003e\u003c/p\u003e\u003cp\u003eWith regard to the Booking.com score (Table\u0026nbsp;\u003cspan refid=\"Tab12\" class=\"InternalRef\"\u003e12\u003c/span\u003e), the means are quite similar for all five AI tools, but some statistical differences arise. In this case, for the Booking.com score variable, again we carried out a Games-Howell test, as was indicated by Levene\u0026rsquo;s test. ChatGPT presents significant differences from Claude and DeepSeek in particular, but also from Grok (at 0.05 level). There are also some statistical differences between DeepSeek and Claude.\u003c/p\u003e\u003cp\u003e\u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab13\" border=\"1\"\u003e\u003ccaption language=\"En\"\u003e\u003cdiv class=\"CaptionNumber\"\u003eTable 13\u003c/div\u003e\u003cdiv class=\"CaptionContent\"\u003e\u003cp\u003eTripAdvisor score descriptives by AI tool\u003c/p\u003e\u003c/div\u003e\u003c/caption\u003e\u003ccolgroup cols=\"6\"\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e\u003cthead\u003e\u003ctr\u003e\u003cth align=\"left\" colname=\"c1\"\u003e\u003cp\u003eAI tool\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c2\"\u003e\u003cp\u003eN\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c3\"\u003e\u003cp\u003eMean\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c4\"\u003e\u003cp\u003eStd. Dev.\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c5\"\u003e\u003cp\u003eMin.\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c6\"\u003e\u003cp\u003eMax.\u003c/p\u003e\u003c/th\u003e\u003c/tr\u003e\u003c/thead\u003e\u003ctbody\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eChatGPT\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e83\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e4.663\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.272\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e4.0\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e\u003cp\u003e5.0\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eGemini\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e84\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e4.470\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.572\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e2.0\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e\u003cp\u003e5.0\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eClaude\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e117\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e4.410\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.511\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e1.5\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e\u003cp\u003e5.0\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eGrok\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e117\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e4.521\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.471\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e1.0\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e\u003cp\u003e5.0\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eDeepSeek\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e128\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e4.531\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.365\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e3.0\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e\u003cp\u003e5.0\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eTotal\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e529\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e4.513\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.455\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e1.0\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e\u003cp\u003e5.0\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003c/tbody\u003e\u003c/colgroup\u003e\u003c/table\u003e\u003c/div\u003e\u003c/p\u003e\u003cp\u003eRegarding the score on TripAdvisor, Table\u0026nbsp;\u003cspan refid=\"Tab13\" class=\"InternalRef\"\u003e13\u003c/span\u003e shows the descriptives for each AI tool and it can be observed that the means are very similar. For the TripAdvisor score variable, again we carried out a Games-Howell post-hoc test to compare the means of the AI tools (Table\u0026nbsp;\u003cspan refid=\"Tab9\" class=\"InternalRef\"\u003e9\u003c/span\u003e). The results show that, again, ChatGPT presents significant differences from the other AI tools, except Grok. It seems that ChatGPT and Grok are more prone than other chatbots to recommending hotels that have high-scoring user reviews.\u003c/p\u003e"},{"header":"5. Conclusions","content":"\u003cp\u003eThis study confirms that chatbots can generate hotel recommendations that align with real-world data to varying degrees. While all tools showed statistically significant correlations with real information on hotels, the degree of accuracy, reliability and bias differed across platforms. ChatGPT, for example, consistently recommended higher-category hotels but tended to underestimate both prices and user scores. Gemini performed well in terms of accuracy for star ratings and review scores, although it also underestimated prices. Claude displayed the strongest correlations across all variables and suggested significantly lower hotel categories. Grok and DeepSeek maintained relatively balanced responses, with DeepSeek showing minimal bias in star ratings but underestimating prices.\u003c/p\u003e\u003cp\u003eThese findings underscore that no single LLM performs optimally across all evaluation dimensions. Therefore, when leveraging AI tools for travel recommendations, users should select the model that best matches their strategic needs, whether that is luxury targeting, price-sensitive suggestions, reliability or balanced accuracy.\u003c/p\u003e\u003cdiv id=\"Sec8\" class=\"Section2\"\u003e\u003ch2\u003e5.1 Theoretical and methodological contribution\u003c/h2\u003e\u003cp\u003eThis study contributes to the ongoing discussion concerning the reliability of generative AI chatbots by being the first to compare the accuracy and biases of different chatbots. This comparison provides unique insights into LLM training and use. We also contribute to the emerging literature on travel-related LLM recommendations by providing insights into how and why these chatbots generate hotel suggestions. The observed correlations with Booking.com and TripAdvisor indicate that data from these platforms have influenced LLM training, reinforcing their importance for hotels aiming to be represented in AI-driven searches. As customers increasingly turn to LLMs to plan their travel, understanding this connection becomes critical for both research and practice in the tourism and hospitality sectors.\u003c/p\u003e\u003cp\u003eImportantly, this article also offers a replicable and scalable methodological framework to analyse and compare evolving chatbots. This framework can be applied in future studies to test new AI agents, track improvements over time, and explore different contexts such as other tourism segments, geographic regions, or languages. By continuously monitoring model behaviour, tourism professionals and researchers can ensure responsible, effective, and user-centred AI deployment in hospitality.\u003c/p\u003e\u003cp\u003eGiven the novelty of this topic, this study addresses a significant gap in the literature by being the first to systematically compare multiple generative AI chatbots in the tourism sector. The proposed methodology allows transparent evaluation of these black-box systems, offering a starting point to better understand their internal decision-making processes. As chatbot outputs vary notably across platforms and sometimes diverge from real-world data, this research demonstrates that the choice of chatbot significantly affects the information retrieved by users, an essential insight for both researchers and practitioners. The originality and replicability of the proposed framework provide a strong foundation for future academic work, especially as the pace of chatbot development accelerates. Moreover, this study highlights the urgent need for more comparative evaluations to ensure the reliability and fairness of recommendations in high-stakes domains like tourism.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec9\" class=\"Section2\"\u003e\u003ch2\u003e5.2 Practical contribution\u003c/h2\u003e\u003cp\u003eUntil now, companies aiming for top positions on Google needed to conduct a keyword study specific to their market niche. With AI chatbots, however, this strategy shifts towards analysing a set of semantically related words tied to those keywords, which then become a series of values that allow the algorithm to better calculate how to respond to a query. This approach adapts strategies to search intent rather than relying on exact keywords.\u003c/p\u003e\u003cp\u003eAs AI continues to evolve, the integration of real-time searches, personalization, and data memory will define how hotels are discovered and recommended. In the future, paid AI placements could emerge, as brands seek visibility within generative responses. Thus, understanding how these tools interpret, prioritize, and segment recommendations is critical for digital competitiveness in the tourism sector (Dwivedi et al., \u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e2024\u003c/span\u003e)\u003c/p\u003e\u003cp\u003eFrom a strategic standpoint, hospitality businesses should start adapting their content and SEO strategies to AI-based platforms, applying Generative Engine Optimization (GEO), Answering Engine Optimization (AEO) and advanced prompting engineering techniques like Chain-of-Thought (CoT) and Tree-of-Thought (ToT) (Aggarwal et al., \u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e2024\u003c/span\u003e; Saleh, \u003cspan citationid=\"CR46\" class=\"CitationRef\"\u003e2025\u003c/span\u003e). CoT allows chatbots to generate step-by-step reasoning sequences, while ToT expands this logic by introducing multiple branches of thought and intermediate evaluations to explore more deliberate and optimal reasoning paths (Zhang et al., \u003cspan citationid=\"CR58\" class=\"CitationRef\"\u003e2024\u003c/span\u003e). Importantly, effective prompt engineering by companies can influence how LLMs interpret and prioritize information, ultimately shaping the recommendations presented to customers. The evolution of these chatbots will depend on their ability to adapt to new environments and to address ethical and technical challenges (Edson de Carvalho Souza, 2025).\u003c/p\u003e\u003cp\u003eThese aforementioned searches enable a more precise segmentation based on user intent, thanks to the ability of AI chatbots\u0026rsquo; ability to process complex queries and respond in a personalized manner. This represents a significant strategic shift for businesses as well as for tourism and hospitality destinations. The capacity to provide recommendations or resolve questions in real time could also reduce the need for traditional chatbots, as AI can handle complex, contextual conversations. Brands that adapt to this shift by creating high-quality content and using data ethically will gain an advantage in this new digital era. AI Chatbot Ads, similar to Google Ads, may appear in the future. However, there is also a risk that recommendations could increasingly be influenced by paid advertising or ranking mechanisms that remain opaque, raising important questions about transparency and trust in AI-driven outputs. Advertising on AI platforms could become less intrusive and more focused on delivering value at the right moment. The chatbot\u0026rsquo;s memory feature allows conversational AI to collect data that can be used to create more detailed user profiles, enabling even more targeted segmentation for future campaigns.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec10\" class=\"Section2\"\u003e\u003ch2\u003e5.3 Limitations and future research\u003c/h2\u003e\u003cp\u003eOur research is primarily descriptive and constrained by the emerging nature of generative AI tools, which are still undergoing rapid development. Future improvements may significantly change how they perform. Additionally, only one version per tool was tested. Paid versions could differ in their responses and capabilities. Finally, the pace at which these chatbots evolve presents a challenge, as research in this area requires constant updates to remain relevant, accurate and reliable. Future research could compare results with other AI tools that have search access, such as Perplexity, or that conduct searches in different languages. Another promising line of research would be to analyse the performance of the new AI agents that each LLM is progressively launching, such as ChatGPT\u0026rsquo;s agents, which became available in Europe on 17 July 2025.\u003c/p\u003e\u003cp\u003eThese findings can be especially useful for hotel managers, tourism marketers, and destination management organizations (DMOs), who can use this information to select the AI chatbot that best aligns with their branding and pricing strategy. Moreover, researchers and developers working on AI-based recommender systems can apply the proposed methodology to assess model performance and fairness across new platforms and languages.\u003c/p\u003e\u003c/div\u003e"},{"header":"Declarations","content":"\u003ch2\u003eAuthor Contribution\u003c/h2\u003e\u003cp\u003eJM-O: conceptualization and writing the main manuscript textEM-F: conceptualization and data analysisBF-R: data analysis and results interpretationJP: manuscript revision\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n\u003cli\u003eAfifah, S., Mudzakir, A., \u0026amp; Nandiyanto, A. B. D. (2022). How to Calculate Paired Sample t-Test using SPSS Software: From Step-by-Step Processing for Users to the Practical Examples in the Analysis of the Effect of Application Anti-Fire Bamboo Teaching Materials on Student Learning Outcomes. \u003cem\u003eIndonesian Journal of Teaching in Science\u003c/em\u003e, \u003cem\u003e2\u003c/em\u003e(1), 81\u0026ndash;92. https://doi.org/10.17509/ijotis.v2i1.45895\u003c/li\u003e\n\u003cli\u003eAgbangba, C. E., Sacla Aide, E., Honfo, H., \u0026amp; Gl\u0026egrave;l\u0026egrave; Kakai, R. (2024). On the use of post-hoc tests in environmental and biological sciences: A critical review. \u003cem\u003eHeliyon\u003c/em\u003e, \u003cem\u003e10\u003c/em\u003e(3), e25131. https://doi.org/10.1016/j.heliyon.2024.e25131\u003c/li\u003e\n\u003cli\u003eAggarwal, P., Murahari, V., Rajpurohit, T., Kalyan, A., Narasimhan, K., \u0026amp; Deshpande, A. (2024). GEO: Generative Engine Optimization. \u003cem\u003eProceedings of the 30th ACM SIGKDD Conference on Knowledge Discovery and Data Mining\u003c/em\u003e, 5\u0026ndash;16. https://doi.org/10.1145/3637528.3671900\u003c/li\u003e\n\u003cli\u003eBailyn, E. (2025). \u003cem\u003eTop Generative AI Chatbots by Market Share \u0026ndash; August 2025\u003c/em\u003e. https://firstpagesage.com/reports/top-generative-ai-chatbots/\u003c/li\u003e\n\u003cli\u003eBareh, C. K. (2025). A qualitative assessment of the accuracy of AI-LLM in academic research. \u003cem\u003eAI and Ethics\u003c/em\u003e, 1\u0026ndash;20. https://doi.org/10.1007/s43681-025-00730-8\u003c/li\u003e\n\u003cli\u003eBhaskar, P., \u0026amp; Sharma, K. D. (2022). A critical insight into the role of artificial intelligence (AI) in tourism and hospitality industries. \u003cem\u003ePacific Business Review International, 15\u003c/em\u003e(3).\u003c/li\u003e\n\u003cli\u003eBuhalis, D., \u0026amp; Moldavska, I. (2022). Voice assistants in hospitality: using artificial intelligence for customer service. \u003cem\u003eJournal of Hospitality and Tourism Technology, 13\u003c/em\u003e(3), 386-403. https://doi.org/10.1108/JHTT-03-2021-0104\u003c/li\u003e\n\u003cli\u003eBulchand-Gidumal, J. (2022). Impact of artificial intelligence in travel, tourism, and hospitality. In \u003cem\u003eHandbook of e-Tourism\u003c/em\u003e (pp. 1943-1962). Cham: Springer International Publishing. https://doi.org/10.1007/978-3-030-48652-5_110\u003c/li\u003e\n\u003cli\u003eChester Business School, University of Chester, United Kingdom, \u0026amp; Leong, K. (2024). An exploratory study of AI participants\u0026rsquo; views on using AI in business hotels. \u003cem\u003eJournal of Applied Sciences in Travel and Hospitality\u003c/em\u003e, \u003cem\u003e7\u003c/em\u003e(2), 69\u0026ndash;83. https://doi.org/10.31940/jasth.v7i2.69-83\u003c/li\u003e\n\u003cli\u003eChitaliya, H., Jain, D., Tiwari, S., Khati, G. S., Sankhe, D., Kanani, P., \u0026amp; Bhanja, M. (2025). ExplainableTrip: An XAI-Driven Personalized Travel Recommendation System Using Gradient Boosting and SHAP. \u003cem\u003eIng\u0026eacute;nierie des Syst\u0026egrave;mes d\u0026rsquo;Information, 30\u003c/em\u003e(5). https://doi.org/10.18280/isi.300507\u003c/li\u003e\n\u003cli\u003eDang, Q., \u0026amp; Li, G. (2025). Unveiling trust in AI: the interplay of antecedents, consequences, and cultural dynamics. \u003cem\u003eAI \u0026amp; Society.\u003c/em\u003e https://doi.org/10.1007/s00146-025-02477-6\u003c/li\u003e\n\u003cli\u003ede Cerqueira, J. S., Kemell, K. K., Rousi, R., Xi, N., Hamari, J., \u0026amp; Abrahamsson, P. (2025). Mapping trustworthiness in large language models: A bibliometric analysis bridging theory to practice. https://doi.org/10.48550/arXiv.2503.0478\u003c/li\u003e\n\u003cli\u003eDeldjoo, Y. (2024). Understanding biases in ChatGPT-based recommender systems: Provider fairness, temporal stability, and recency. \u003cem\u003eACM Transactions on Recommender Systems\u003c/em\u003e. https://doi.org/10.1145/3690655\u003c/li\u003e\n\u003cli\u003eDi Palma, D., Biancofiore, G. M., Anelli, V. W., Narducci, F., Di Noia, T., \u0026amp; Di Sciascio, E. (2023). Evaluating chatgpt as a recommender system: A rigorous approach. \u003cem\u003earXiv preprint\u003c/em\u003e arXiv:2309.03613. https://doi.org/10.48550/arXiv.2309.03613\u003c/li\u003e\n\u003cli\u003eDwivedi, Y. K., Pandey, N., Currie, W., \u0026amp; Micu, A. (2024). Leveraging ChatGPT and other generative artificial intelligence (AI)-based applications in the hospitality and tourism industry: Practices, challenges and research agenda. \u003cem\u003eInternational Journal of Contemporary Hospitality Management\u003c/em\u003e, \u003cem\u003e36\u003c/em\u003e(1), 1\u0026ndash;12. https://doi.org/10.1108/IJCHM-05-2023-0686\u003c/li\u003e\n\u003cli\u003eEdson de Carvalho Souza, M. (2025). \u003cem\u003eGrok, Gemini, ChatGPT and DeepSeek: Comparison and Applications in Conversational Artificial Intelligence\u003c/em\u003e. https://doi.org/10.5281/ZENODO.14885243\u003c/li\u003e\n\u003cli\u003eEnternat. (2024). \u003cem\u003eCiudades m\u0026aacute;s visitadas del mundo 2025 (ranking de turismo)\u003c/em\u003e. https://www.enterat.com/ocio/ciudades-mas-visitadas-mundo.php\u003c/li\u003e\n\u003cli\u003eFern\u0026aacute;ndez, Y. (2025). \u003cem\u003eGrok 3: Qu\u0026eacute; es y principales novedades del nuevo modelo de inteligencia artificial de xAI\u003c/em\u003e [Xataka]. https://www.xataka.com/basics/grok-3-que-principales-novedades-nuevo-modelo-inteligencia-artificial-xai\u003c/li\u003e\n\u003cli\u003eHasan, E., Rahman, M., Ding, C., Huang, J., \u0026amp; Raza, S. (2024). Based recommender systems: A survey of approaches, challenges and future perspectives. \u003cem\u003eACM Computing Surveys.\u003c/em\u003e https://doi.org/10.1145/3742421\u003c/li\u003e\n\u003cli\u003eIlieva, G., Yankova, T., \u0026amp; Klisarova-Belcheva, S. (2024). Effects of Generative AI in Tourism Industry. \u003cem\u003eInformation\u003c/em\u003e, \u003cem\u003e15\u003c/em\u003e(11), 671. https://doi.org/10.3390/info15110671\u003c/li\u003e\n\u003cli\u003eImran, M., \u0026amp; Almusharraf, N. (2024). Google Gemini as a next generation AI educational tool: A review of emerging educational technology. \u003cem\u003eSmart Learning Environments\u003c/em\u003e, \u003cem\u003e11\u003c/em\u003e(1), 22. https://doi.org/10.1186/s40561-024-00310-z\u003c/li\u003e\n\u003cli\u003eIslam, R., \u0026amp; Ahmed, I. (2024). Gemini-the most powerful LLM: Myth or Truth. \u003cem\u003e2024 5th Information Communication Technologies Conference (ICTC)\u003c/em\u003e, 303\u0026ndash;308. https://doi.org/10.1109/ICTC61510.2024.10602253\u003c/li\u003e\n\u003cli\u003eKim, J., Kim, J. H., Kim, C., \u0026amp; Park, J. (2023). Decisions with ChatGPT: Reexamining choice overload in ChatGPT recommendations. \u003cem\u003eJournal of Retailing and Consumer Services, 75\u003c/em\u003e. https://doi.org/10.1016/j.jretconser.2023.103494\u003c/li\u003e\n\u003cli\u003eKim, J. H., Kim, J., Park, J., Kim, C., Jhang, J., \u0026amp; King, B. (2025). When ChatGPT gives incorrect answers: The impact of inaccurate information by generative AI on tourism decision-making. \u003cem\u003eJournal of Travel Research, 64\u003c/em\u003e(1), 51\u0026ndash;73. https://doi.org/10.1177/0047287523121299\u003c/li\u003e\n\u003cli\u003eKowald, D. (2025). Investigating Popularity Bias Amplification in Recommender Systems Employed in the Entertainment Domain. https://doi.org/10.48550/arXiv.2504.04752\u003c/li\u003e\n\u003cli\u003eLabrague, L. J. (2024). Utilizing Artificial Intelligence\u0026ndash;Based Tools for Addressing Clinical Queries: ChatGPT Versus Google Gemini. \u003cem\u003eJournal of Nursing Education\u003c/em\u003e, \u003cem\u003e63\u003c/em\u003e(8), 556\u0026ndash;559. https://doi.org/10.3928/01484834-20240426-01\u003c/li\u003e\n\u003cli\u003eLee, M., Kwon, W., \u0026amp; Back, K. J. (2021). Artificial intelligence for hospitality big data analytics: Developing a prediction model of restaurant review helpfulness for customer decision-making. \u003cem\u003eInternational Journal of Contemporary Hospitality Management, 33\u003c/em\u003e(6), 2117-2136. https://doi.org/10.1108/IJCHM-06-2020-0587\u003c/li\u003e\n\u003cli\u003eLichtenberg, J. M., Buchholz, A., \u0026amp; Schw\u0026ouml;bel, P. (2024). Large language models as recommender systems: A study of popularity bias. arXiv preprint arXiv:2406.01285.\u003c/li\u003e\n\u003cli\u003eLiu, H. (2015). \u003cem\u003eComparing Welch\u0026rsquo;s ANOVA, a Kruskal-Wallis test and traditional ANOVA in case of Heterogeneity of Variance\u003c/em\u003e.\u003c/li\u003e\n\u003cli\u003eLiebenow, L. W., Schmidt, F. T., Meyer, J., \u0026amp; Fleckenstein, J. (2025). Self-assessment accuracy in the age of artificial intelligence: Differential effects of LLM-generated feedback. \u003cem\u003eComputers \u0026amp; Education\u003c/em\u003e, 105385. https://doi.org/110.1016/j.compedu.2025.105385\u003c/li\u003e\n\u003cli\u003eLu, P., \u0026amp; Feng, X. (2024, May). Personalized recommendation algorithm in AI-assisted learning system. In \u003cem\u003e2024 6th International Conference on Communications, Information System and Computer Engineering (CISCE)\u003c/em\u003e (pp. 1289\u0026ndash;1294). IEEE. https://doi.org/10.1109/CISCE62493.2024.10653402\u003c/li\u003e\n\u003cli\u003eLu, H., Liu, W., Zhang, B., Wang, B., Dong, K., Liu, B., Sun, J., Ren, T., Li, Z., Yang, H., Sun, Y., Deng, C., Xu, H., Xie, Z., \u0026amp; Ruan, C. (2024). \u003cem\u003eDeepSeek-VL: Towards Real-World Vision-Language Understanding\u003c/em\u003e (arXiv:2403.05525). arXiv. https://doi.org/10.48550/arXiv.2403.05525\u003c/li\u003e\n\u003cli\u003eMalak, A., \u0026amp; Şahin, M. F. (2024). How Useful are Current Chatbots Regarding Urology Patient Information? Comparison of the Ten Most Popular Chatbots\u0026rsquo; Responses About Female Urinary Incontinence. \u003cem\u003eJournal of Medical Systems\u003c/em\u003e, \u003cem\u003e48\u003c/em\u003e(1), 102. https://doi.org/10.1007/s10916-024-02125-4\u003c/li\u003e\n\u003cli\u003eMasciari, E., Umair, A., \u0026amp; Ullah, M. H. (2024). A systematic literature review on AI-based recommendation systems and their ethical considerations. \u003cem\u003eIEEE Access, 12\u003c/em\u003e, 121223\u0026ndash;121241. https://doi.org/10.1109/ACCESS.2024.3451054\u003c/li\u003e\n\u003cli\u003eMladenović, D., Beheshti, M., Kolar, T., Ismagilova, E., \u0026amp; Dwivedi, Y. K. (2024). Synthetic WOM? The emergence of generative artificial intelligence-induced recommendations. \u003cem\u003eJournal of Computer Information Systems\u003c/em\u003e, 1\u0026ndash;18. https://doi.org/10.1080/08874417.2024.2409252\u003c/li\u003e\n\u003cli\u003eOmar, M., Sorin, V., Agbareia, R., Apakama, D. U., Soroush, A., \u0026amp; Sakhuja,A. (2025). Evaluating and addressing demographic disparities in medical large language models: a systematic review. \u003cem\u003eInternational Journal for Equity in Health, 24\u003c/em\u003e(1), 57. https://doi.org/10.1186/s12939-025-02419-0\u003c/li\u003e\n\u003cli\u003eOpenAI. (2024). \u003cem\u003eIntroducing ChatGPT search\u003c/em\u003e. https://openai.com/index/introducing-chatgpt-search/\u003c/li\u003e\n\u003cli\u003ePande, A., Patil, R., Mukkemwar, R., Panchal, R., \u0026amp; Bhoite, S. (2024). \u003cem\u003eComprehensive Study of Google Gemini and Text Generating Models: Understanding Capabilities and Performance\u003c/em\u003e.\u003c/li\u003e\n\u003cli\u003ePetkovic, D. (2023). It is not \u0026ldquo;Accuracy vs. Explainability\u0026rdquo;\u0026mdash;we need both for trustworthy AI systems. \u003cem\u003eIEEE Transactions on Technology and Society, 4\u003c/em\u003e(1), 46\u0026ndash;53. https://doi.org/10.1109/TTS.2023.3239921\u003c/li\u003e\n\u003cli\u003ePillai, R., \u0026amp; Sivathanu, B. (2020). Adoption of AI-based chatbots for hospitality and tourism. \u003cem\u003eInternational Journal of Contemporary Hospitality Management, 32\u003c/em\u003e(10), 3199-3226. https://doi.org/10.1108/IJCHM-04-2020-0259\u003c/li\u003e\n\u003cli\u003ePriyanshu, A., Maurya, Y., \u0026amp; Hong, Z. (2024). \u003cem\u003eAI Governance and Accountability: An Analysis of Anthropic\u0026rsquo;s Claude\u003c/em\u003e (arXiv:2407.01557). arXiv. https://doi.org/10.48550/arXiv.2407.01557\u003c/li\u003e\n\u003cli\u003ePopesku, J. (2019). Current applications of artificial intelligence in tourism and hospitality. In \u003cem\u003eSinteza 2019-International Scientific Conference on Information Technology and Data Related Research\u003c/em\u003e (84-90). Singidunum University. https://doi.org/10.15308/Sinteza-2019-84-90\u003c/li\u003e\n\u003cli\u003eRay, P. P. (2024). A deep introspection into the role of ChatGPT for transforming hospitality, leisure, sport, and tourism education. \u003cem\u003eJournal of Hospitality, Leisure, Sport \u0026amp; Tourism Education\u003c/em\u003e, \u003cem\u003e35\u003c/em\u003e, 100504. https://doi.org/10.1016/j.jhlste.2024.100504\u003c/li\u003e\n\u003cli\u003eRemountakis, M., Kotis, K., Kourtzis, B., \u0026amp; Tsekouras, G. E. (2023). ChatGPT and persuasive technologies for the management and delivery of personalized recommendations in hotel hospitality. https://doi.org/10.48550/arXiv.2307.14298\u003c/li\u003e\n\u003cli\u003eReuters. (2025). \u003cem\u003eChinese universities launch DeepSeek courses to capitalise on AI boom\u003c/em\u003e. https://www.reuters.com/technology/artificial-intelligence/chinese-universities-launch-deepseek-courses-capitalise-ai-boom-2025-02-21/?utm_source=chatgpt.com\u003c/li\u003e\n\u003cli\u003eSaleh, M. I. (2025). Generative artificial intelligence in hospitality and tourism: Future capabilities, AI prompts and real-world applications. \u003cem\u003eJournal of Hospitality Marketing \u0026amp; Management\u003c/em\u003e, 1\u0026ndash;32. https://doi.org/10.1080/19368623.2025.2458603\u003c/li\u003e\n\u003cli\u003eSaid, A. (2025). On explaining recommendations with Large Language Models: A review. \u003cem\u003eFrontiers in Big Data, 7,\u003c/em\u003e 1505284. https://doi.org/10.3389/fdata.2024.1505284\u003c/li\u003e\n\u003cli\u003eSchmidgall, S., Harris, C., Essien, I., Olshvang, D., Rahman, T., Kim, J. W., \u0026amp; Chellappa, R. (2024). Evaluation and mitigation of cognitive biases in medical language models. npj \u003cem\u003eDigital Medicine, 7\u003c/em\u003e(1), 295. https://doi.org/10.1038/s41746-024-01283-6\u003c/li\u003e\n\u003cli\u003eSedgwick, P. (2012). Pearson\u0026rsquo;s correlation coefficient. \u003cem\u003eBMJ\u003c/em\u003e, \u003cem\u003e345\u003c/em\u003e(jul04 1), e4483\u0026ndash;e4483. https://doi.org/10.1136/bmj.e4483\u003c/li\u003e\n\u003cli\u003eSigala, M., Ooi, K.-B., Tan, G. W.-H., Aw, E. C.-X., Buhalis, D., Cham, T.-H., Chen, M.-M., Dwivedi, Y. K., Gretzel, U., Inversini, A., Jung, T., Law, R., \u0026amp; Ye, I. H. (2024). Understanding the impact of ChatGPT on tourism and hospitality: Trends, prospects and research agenda. \u003cem\u003eJournal of Hospitality and Tourism Management\u003c/em\u003e, \u003cem\u003e60\u003c/em\u003e, 384\u0026ndash;390. https://doi.org/10.1016/j.jhtm.2024.08.004\u003c/li\u003e\n\u003cli\u003e\u0026Scaron;tilić, A., Nicić, M., \u0026amp; Pu\u0026scaron;ka, A. (2023). Check-in to the future: Exploring the impact of contemporary information technologies and artificial intelligence on the hotel industry. \u003cem\u003eTurističko poslovanje, (31).\u003c/em\u003e https://doi.org/10.5937/turpos0-43739\u003c/li\u003e\n\u003cli\u003eTussyadiah, I. (2020). A review of research into automation in tourism: Launching the Annals of Tourism Research Curated Collection on Artificial Intelligence and Robotics in Tourism. \u003cem\u003eAnnals of Tourism Research\u003c/em\u003e, \u003cem\u003e81\u003c/em\u003e, 102883. https://doi.org/10.1016/j.annals.2020.102883\u003c/li\u003e\n\u003cli\u003eVashishth, T. K., Sharma, V., Sharma, K. K., Kumar, B., Kumar, A., \u0026amp; Panwar, R. (2024). Artificial intelligence (AI)\u0026ndash;powered chatbots: Providing instant support and personalized recommendations to guests 24/7. In \u003cem\u003eTechnology and Luxury Hospitality\u003c/em\u003e (211-236). Routledge. https://doi.org/10.4324/9781003488248-15\u003c/li\u003e\n\u003cli\u003eViglia, G., Adler, S. J., Miltgen, C. L., \u0026amp; Sarstedt, M. (2024). The use of synthetic data in tourism. \u003cem\u003eAnnals of Tourism Research\u003c/em\u003e, \u003cem\u003e108\u003c/em\u003e, 103819. https://doi.org/10.1016/j.annals.2024.103819\u003c/li\u003e\n\u003cli\u003eWong, I. A., Lian, Q. L., \u0026amp; Sun, D. (2023). Autonomous travel decision-making: An early glimpse into ChatGPT and generative AI. \u003cem\u003eJournal of Hospitality and Tourism Management, 56,\u003c/em\u003e 253-263. https://doi.org/10.1016/j.jhtm.2023.06.022\u003c/li\u003e\n\u003cli\u003eWu, S., Koo, M., Blum, L., Black, A., Kao, L., Scalzo, F., \u0026amp; Kurtz, I. (2023). \u003cem\u003eA Comparative Study of Open-Source Large Language Models, GPT-4 and Claude 2: Multiple-Choice Test Taking in Nephrology\u003c/em\u003e (arXiv:2308.04709). arXiv. https://doi.org/10.48550/arXiv.2308.04709\u003c/li\u003e\n\u003cli\u003eWu, Z., Chen, X., Pan, Z., Liu, X., Liu, W., Dai, D., Gao, H., Ma, Y., Wu, C., Wang, B., Xie, Z., Wu, Y., Hu, K., Wang, J., Sun, Y., Li, Y., Piao, Y., Guan, K., Liu, A., \u0026hellip; Ruan, C. (2024). \u003cem\u003eDeepSeek-VL2: Mixture-of-Experts Vision-Language Models for Advanced Multimodal Understanding\u003c/em\u003e (arXiv:2412.10302). arXiv. https://doi.org/10.48550/arXiv.2412.10302\u003c/li\u003e\n\u003cli\u003eZhang, X., Du, C., Pang, T., Liu, Q., Gao, W., \u0026amp; Lin, M. (2024). \u003cem\u003eChain of Preference Optimization: Improving Chain-of-Thought Reasoning in LLMs\u003c/em\u003e.\u003c/li\u003e\n\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"information-technology-and-tourism","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"jitt","sideBox":"Learn more about [Information Technology \u0026 Tourism](https://link.springer.com/journal/40558)","snPcode":"40558","submissionUrl":"https://submission.springernature.com/new-submission/40558/3","title":"Information Technology \u0026 Tourism","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"Springer Hybrid","inReviewEnabled":true,"inReviewRevisionsEnabled":false},"keywords":"Artificial Intelligence (AI), Hospitality, Prompt, large language models (LLM)","lastPublishedDoi":"10.21203/rs.3.rs-7687581/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-7687581/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eThe rise of generative AI tools like ChatGPT, Claude, Gemini, DeepSeek and Grok is transforming the way users interact with digital information, particularly in the global hospitality industry. This study evaluates hotel recommendations generated by these AI chatbots across the top 10 most visited cities worldwide. A comprehensive comparative analysis is conducted to evaluate whether these tools provide reliable and unbiased suggestions by comparing their outputs with verified real hotel data, including price, hotel category, and scores from Booking.com and TripAdvisor. The findings reveal a significant difference between AI-generated data and actual real-world values, especially in pricing. ChatGPT consistently recommends higher-category hotels but often underestimates scores and prices. Gemini achieves the closest alignment with star ratings. DeepSeek and Grok present increasingly promising multimodal capabilities. The study highlights the potential and current limitations of AI-driven hotel recommendations, offering strategic insights for hospitality businesses that are adapting to rapidly changing AI-driven search behaviour.\u003c/p\u003e","manuscriptTitle":"Are all Generative AI Chatbots the Same? Analysing the Reliability of Hotel Recommendations","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-11-14 09:39:35","doi":"10.21203/rs.3.rs-7687581/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"decision","content":"Revision requested","date":"2026-02-17T10:27:39+00:00","index":"","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2026-02-16T09:46:25+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"83711110612638254103379939373001496798","date":"2026-02-04T20:42:19+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-11-24T10:39:43+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"110453755054223645575424243081980278148","date":"2025-11-05T09:02:09+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2025-11-04T23:24:28+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2025-10-10T16:45:53+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2025-09-25T10:08:22+00:00","index":"","fulltext":""},{"type":"submitted","content":"Information Technology \u0026 Tourism","date":"2025-09-22T19:03:53+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"information-technology-and-tourism","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"jitt","sideBox":"Learn more about [Information Technology \u0026 Tourism](https://link.springer.com/journal/40558)","snPcode":"40558","submissionUrl":"https://submission.springernature.com/new-submission/40558/3","title":"Information Technology \u0026 Tourism","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"Springer Hybrid","inReviewEnabled":true,"inReviewRevisionsEnabled":false}}],"origin":"","ownerIdentity":"4ec16ac2-f128-4dcf-bb48-76fe9cf31911","owner":[],"postedDate":"November 14th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"under-review","subjectAreas":[],"tags":[],"updatedAt":"2026-04-17T16:38:42+00:00","versionOfRecord":[],"versionCreatedAt":"2025-11-14 09:39:35","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-7687581","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-7687581","identity":"rs-7687581","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}