The use of artificial intelligence in the travel and hospitality industry. Civil liability profiles by Caterina del Federico IJTTHL PRE-PRINT The use of artificial intelligence in the travel and hospitality industry. Civil liability profiles Caterina del Federico University of Bologna International Journal of Tourism, Travel and Hospitality Law PRE-PRINT

The use of artifcial intelligence in the travel and hospitality industry. Civil liability profles Caterina del Federico, Italy, University of Bologna 1) A brief introduction This contribution will develop through some questions having a pivotal role in artifcial intelligence debate and on its use in the feld of tourism. The approach used is that of the jurist; thus, the acts issued in recent years by the European Union will be primarily considered to answer the various proposed questions. Certainly, this brief paper does not presume to provide defnite answers in the face of an issue that raises so many questions; rather, an attempt is made to raise some critical issues and provide some insights As normally happens, new phenomena are matched by new legal issues. Preliminarily, there is a question of whether there is a defnition of artifcial intelligence . In a second 1 step, the relevance of artifcial intelligence systems in the tourism industry will be demonstrated through some concrete examples. Finally, the most critical point will be raised: who is responsible for the damages produced by an artifcial intelligence system? What is the applicable liability regime? Therefore, the issue of civil liability will be analyzed, an issue on which uncertainty reigns and on which there is a proliferation of acts by the European Union. 2) What is artifcial intelligence First, there is the question of what is meant by the term “artifcial intelligence”. Frequently, in fact, the term is used inappropriately. For example, when reading the legal doctrine on the subject, one notices that some authors indiscriminately use the term artifcial intelligence and the term blockchain, or the term artifcial intelligence and the expression internet of things. It seems appropriate to provide clarity and identify a fxed point from which to start. On these profles and for a review of the diferent meanings the expression has taken on over the years 1 see, among many contributions, Floridi, L., Cabitza, F. (2021) L’intelligenza artifciale. L’uso delle nuove machine, Bompiani, Milano. On the practical and case history level see also Rothman, D. (2018) Artifcial Intelligence By Examples: Develop Machine Inteligence From Scratch Using Real Artifcial Intelligence Use Cases, Packt Publishing, Birmingham, UK. 1

Among the key acts issued by the European Union is undoubtedly the so-called "AI act. (AIA)"; this is the Proposal Regulation laying down harmonized rules on artifcial intelligence of April 21, 2021 . 2 The Proposal has the merit of providing several defnitions, contained in Article 3, which is headed "Defnitions". The frst defnition is precisely that of "artifcial intelligence". The provision reads: «'artifcial intelligence' (AI system) means software developed with one or more of the techniques and approaches listed in Annex I and can, for a given set of human-defned objectives, generate outputs such as content, predictions, recommendations, or decisions infuencing the environments they interact with». This is a rather vague defnition, referring to the specifc lists in Annex I. The Artifcial Intelligence Act (AIA) defnes various areas of focus of AI systems: applications prohibited because they cause unbearable risks to fundamental rights and freedoms ; high-risk applications 3 (not prohibited but subject to specifc conditions to manage risks) ; limited-risk 4 applications and other applications with negligible risk . 5 The list of banned AI systems includes online manipulative practices that produce physical or psychological harm to individuals or exploit their vulnerability based on age or disability; social scoring that produces disproportionate or decontextualized harmful efects; and biometric identifcation systems used by law enforcement in public spaces (when their use is not strictly necessary or when the risk of harmful efects is too high). For the frst time, European regulators are attempting to defne a boundary or limit that should not be crossed when employing AI-based services or products in society. Unlike prohibited AI systems, AI systems classifed as "high risk" are not prohibited by default, but subject to several compliance obligations. These obligations include, among others, a risk management plan, compliance certifcation, a data management plan, and human oversight. The list of high-risk AI systems in the AIA includes facial recognition; AI used in critical infrastructure; in educational, employment, or emergency contexts; in For an analysis of the proposed regulation see Finocchiaro, G. (2022) ‘La proposta di regolamento 2 sull’intelligenza artifciale: il modello europeo basato sulla gestione del rischio’, Diritto dell’informazione e dell’informatica, 2022/2, pp. 303-327; Towsend, B. (2021) ‘Decoding the Proposed European Union Artifcial Intelligence Act’, Insights, 25(20); Veale, M., Borgesius, F.Z. (2021) Demystifying the Draft EU Artifcial Intelligence Act, in Computer Law Review International, 4/2021, pp. 97-112. Such indications can be found in art. 5 of the AIA. 3 These kinds of systems are indicated analytically in the annex III of the Proposal. 4 Some information is provided in the explanatory memorandum of the AIA. 5 2

asylum and border contexts; in social assistance; for credit scoring; for use by law enforcement; or for judicial purposes. The EU Commission may update this list based on the severity and likelihood of impact of present and future AI systems on fundamental rights. Finally, a third category of AI systems is considered of "limited risk". This category includes morally questionable AI applications such as AI algorithms that produce deepfakes (highly realistic fake videos or photos) as well as emotion recognition and biometric categorization systems. The use of these systems would imply no specifc duty of compliance, only very vague transparency requirements: a simple notifcation to consumers/citizens about whether an AI system is operating in that context. Therefore, to have a complete picture of artifcial intelligence systems, the defnition in Article 3 of the Proposal is not sufcient, but it is appropriate to view Annex I of the Proposal. However, a limitation should be pointed out: it is almost impossible to keep the lists up to date with all the intelligence systems that are being developed. 3) How is AI changing the travel industry Thus, according to what has been said, trivializing it can be said that artifcial intelligence is a branch of computer science where machines and computers simulate human intelligence. In AI machines are programmed to think like humans can do; it leverages the problem solving and decision-making capabilities of human beings. Artifcial intelligence is not just a single technology, but a rapidly evolving collection of technologies like deep learning, machine learning, and expert systems. AI has made advances in many industries in the last years. According to some reports , 6 the penetration of artifcial intelligence in any business feld in 2021 has increased by 56% compared to 2019. In other words, it is already a tool of the present with multiple applications. More specifcally, the private investment in 2021 totaled around $93.5 billion, more than double the total private investment in 2020, while the number of newly funded See the report by the consulting frm McKinsey & Company, available at the link https:// 6 See also the report of 2022 published by the Stanford Institute for Human-Centered Artifcial Intelligence (HAI), led by the AI Index Steering Committee, available at the link 2022/03/2022-AI-Index-Report_Master.pdf. The AI Index is an independent initiative at the Stanford Institute for Human-Centered Artifcial Intelligence (HAI), led by the AI Index Steering Committee, an i n t e rd i sc i p l i na r y group o f expe r t s f rom ac ross academi a and i ndus t r y. The annua l report tracks, collates, distills, and visualizes data relating to artifcial intelligence, enabling decision-makers to take meaningful action to advance AI responsibly and ethically with humans in mind. 3

AI companies continues to drop, from 1051 companies in 2019 and 762 companies in 2020 to 746 companies in 2021. In 2020 there were 4 funding rounds worth $500 million or more; in 2021 there were 15 . 7 The tourism industry is also taking advantage of AI to revolutionize the way it operates. Travel companies use AI in various processes from travel planning to landing in the destination. Before the AI revolution traveling was required time and research. This was before could recommend cheap fights to the tourist’s target destinations, before Airbnb could better assist the tourist’s accommodation based on previous trips, and before fellow tourists could tell some other tourist about destinations on Tripadvisor. Back then, the tourist had to spend time outlining his itinerary, searching for fights, and sorting through several flters to fnd that one hotel. The arrival of artifcial intelligence has changed all that, introducing new ways of experiencing tourism and the travel experience. Below just a few examples involving artifcial intelligence systems that can be identifed as trends in the feld of tourism: AI assistants for travel booking, robots for face-to-face customer services, AI driven applications for fight forecasting, data analytics to identify valuable insights, sentiment analysis through social media, room mapping + dynamic price tracking, facial recognition at the airports for border identifcation check. Chatbot, facial recognition and other AI systems used in tourism Thanks to artifcial intelligence, travelers no longer need to visit travel agencies to book fights or search for accommodation. AI assistants and intelligent chatbots have now taken the place of travel agents allowing travelers to book fights and accommodation and hire vehicles online, just to give a few examples. These chatbots are normally deployed in social media sites like Facebook messenger, Skype, WhatsApp to ofer users a more personalized booking experience. The travel reservation giants like, Skyskanner and Expedia are utilizing such chatbot by entering the details of the tourist’s intended journey. Then the bot will search through the booking sites and fnd the tourist the best deal. Chatbot technology cover an important part of AI systems used in the feld of tourism, and it is not surprising that many travel brands have already launched their own versions in the last year or so. That of Skyscanner is just one example, creating a smart bot to help See chap. 4 of the Report made by the Stanford Institute for Human-Centered Artifcial Intelligence (HAI). 7 4

consumers fnd fights in Facebook Messenger. Users can also use it to request travel recommendations and unplanned suggestions. Unlike e-commerce or retail brands that use chatbots, which can seem useless and deceitful, it is fair to say that examples like Skyscanner are much more relevant and useful to everyday consumers. After all, with the advent of so many other travels search sites, consumers are overwhelmed with choice. Therefore, a chatbot like Skyscanner can connect with consumers at their own pace and in the social media spaces they frequent most often. Let’s see quickly how Skyskanner uses its chatbot. When the Skyscanner chatbot is activated on Facebook or Whatsapp, users can enter their destination to start the conversation with the bot. If the user does not know where to go, he just types “anywhere”, and then the bot will provide him with some suggestions, including the price for each destination using real-time search trends. When the consumer tells the chatbot the travel date, it will show him the cheapest fights to his destination. Then the bot will redirect the user to the ofcial Skyskanner website to complete the booking upon selecting to book a fight. Moreover, the Skyskanner chatbot allows the user to receive price alerts for a fight. Skyscanner reported that it had surpassed one million traveler interactions with the application in February 2018 and 8 they are increasing year by year. One other example is that of Aero Mexico - the well-known airline for travel to Mexico and Latin America -, that also started using the Facebook Messenger chatbot to answer very general questions from customers. The main idea was to answer 80% of the questions, which are usually repetitive and on common topics. Airlines are benefting tremendously from the use of artifcial intelligence systems. Also, KLM Royal Dutch Airlines uses AI to answer customer questions on Twitter and Facebook. It uses an algorithm from a company called Digital Genius, that has learned from over 50,000 questions and answers. In addition, Deutsche Lufthansa's Mildred bot can help search for the lowest fares. A chatbot is considered an artifcial intelligence system of law risk, as specifcally stated in the explanatory memorandum of the AIA . 9 The report is available at the link: 8 unique-traveller-interactions/. The explanatory memorandum reads: «for some specifc AI systems, only minimum transparency 9 obligations are proposed, in particular when chatbots or ‘deep fakes’ are used». 5

Another good example of an artifcial intelligence system increasingly used in the feld of tourism is that of robots for face-to-face customer services. Robots, indeed, are gradually infltrating customer services in the travel industry, avoiding the need for human agents, this especially from the period of the covid-19. Some popular airports and hotels have now employed robots to assist the client in such situations. One good example is that of the robot in London Heathrow airport. This robot guides passengers in terminals. According to a report published by Vero Solutions , it is 10 expected to replace robots for the checking process by 2030. On the other hand, replacing the human workforce with machines is a trend already underway in various sectors, but in the past two decades, the number of robots in use worldwide has tripled to 2.25 million. However, in tandem with the extremely rapid technological evolution that now afects every sector, these numbers are increasing exponentially, generating disruptive consequences. This is confrmed by a study conducted by Oxford Economics , a British company specializing in global forecasting studies and quantitative 11 analysis. How Robots Change the World is the title of the report that is alarming workers and others from America to the United Kingdom. In fact, researchers have estimated that by 2030, 20 million manufacturing jobs globally could be in the hands of robots, including 14 million in China alone. In Japan, between 2015 and 2016, has opened the frst hotel where robots will be the ones to welcome clients, carry their bags and be their point of contact for any need. Henn-na Hotel in Nagasaki has been the worlds’frst hotel fully stafed by multilingual robots primarily used for check-in and checkout processing. Another innovative AI solution is Connie, the robot deployed in the hotel Hilton McLean in Virginia. Connie, the robot Concierge in Hilton uses domain knowledge of the IBM Watson AI program and Wayblazer to answer human queries related to hotel features and local attractions while providing recommendations. It uses Natural language processing to understand the sentences when someone talks to him. Furthermore, Connie can move its The report is available at the link: 10 coming-to-an-airport-near-you.html. The report ci ted is of June 2019 and i t is avai lable at the fol lowing l ink: https:// 11 _gl=1*1ddo0md*_ga*MTQ4NzYyNDc4LjE2NjU0ODUzMzQ.*_ga_GZFMW0F8TH*MTY2NTQ4NTUyOS4xLjAu MTY2NTQ4NTUyOS42MC4wLjA. 6

arms and legs to shoe directions to guests and express diferent human interactions and fne-tunes its capabilities. It was particularly useful in the covid-19 era to ensure customer safety. Another massive use of artifcial intelligence systems in tourism is represented by some widely used forms of AI in airports these days, such as facial recognition systems for airport security. Facial recognition is a biometric technology, that greatly increases the efciency of passenger screening by automating the process of verifying photo identifcation at the airport. According to the New York Times , airports are looking for additional ways to use 12 biometric verifcation methods like facial recognition. The National Institute of Standards and Technology has also discovered that facial recognition is 99.5% accurate . Soon, all 13 airlines will require passengers to go through biometric verifcation before boarding planes. Then again, the relevance of biometric systems in tourism, and particularly at airports, as well as the need to establish certain legal rules for such systems, is also underscored by the European Committee on Transport and Tourism, that in 2022 has issued a draft opinion on these topics. In particular, the relevance of biometric systems is highlighted 14 in Amendment 3 of the draft opinion cited, where is stated that the Proposal Regulation This information is available at the link: 12 security.html. It is possible to fnd this data at the link: 13 face-recognition-softwares-accuracy-fight-boarding. Draft opinion for the Committee on the Internal Market and Consumer Protection and the Committee on 14 Civil Liberties, Justice and Home Afairs on the proposal for a regulation of the European Parliament and of the Council laying down harmonized rules on artifcial intelligence and amending certain union legislative acts (COM(2021)0206 – C9-0146/2021 – 2021/0106(COD)). 7

on Artifcial Intelligence should accompany these developments by the highest and adequate level of protection, in particular when biometric data are used . 15 4) Civil liability profles It follows from the above that artifcial intelligence has been incorporated into many diferent areas of the travel and tourism industry, making lives easier for travelers around the globe. The European Union in recent years has published a huge number of acts focused on and dedicated to artifcial intelligence. Just to give some examples, the following can be mentioned in the past two years: a White paper on AI: a European approach to excellence and trust (2020) ; a Resolution 16 with recommendations to the Commission on a civil liability regime for AI (2020) ; a 17 Resolution on a framework of ethical aspects of artifcial intelligence technologies (2020) ; a Proposal for a Regulation laying down harmonized rules on AI (2021) ; a 18 19 public consultation on civil liability – adapting liability rules to the digital age and AI Amendment 3, Proposal for a regulation, Recital 8 a (new), text proposed by the Commission (8a): «the 15 use of biometrics in transport and tourism will vastly beneft user experience and overall safety and security. The application of fngerprint or retina scans to access cars could help prevent theft, while in-car biometrics could help detect drivers' stress levels and prevent intoxicated driving, directly contributing to the EU’s 2050 “VisionZero”. In the tourism sector, contactless check-ins, for example through facial recognition technology, will help attain a seamless travelling experience. This Regulation should accompany these developments by the highest/adequate level of protection, especially when use of biometrics data is involved, in line with the data protection framework of the Union, while fostering research and investment for the development and deployment of AI systems that can positively contribute to society». Similarly, Amendment 6, Proposal for a regulation, Recital 12 a (new), text proposed by the Commission (12a): «this regulation should support research and innovation for the application of AI systems in the transport and tourism sectors. For this reason, this Regulation should exclude from its scope applications of AI systems developed, applied, and assessed in a controlled testing environment, for the sole purpose of evaluating their use and functionality. As regards product-oriented research activity by providers, the provisions of this Regulation should apply insofar as such research leads to or entails placing an AI system on the market or putting it into service. All forms of research and development should be conducted in compliance with the highest ethical standards for scientifc research». The text is available at the link: 16 approach-excellence-and-trust_en. The text i s ava i l abl e at the l ink: ht tps: / /www.europar l / 17 TA-9-2020-0276_EN.html. The text i s ava i l abl e at the l ink: ht tps: / /www.europar l / 18 TA-9-2020-0275_EN.html. This proposal is known as “artifcial intelligence act” (o AIA), the text is available at the link: https://digital- 19 8

(2021) ; a Resolution on AI in a digital age (2022) ; last, but not least, a Proposal for an 20 21 AI liability Directive , which therefore apparently supersedes previous acts on the topic of 22 liability. Now, to connect the topic of civil liability to that of the use of AI systems in tourism, it is 23 worth considering a couple of the case cited above. Incidentally, that of the liability of artifcial intelligence systems is among the most debated in the legal world today. 24 As a frst example can be considered the cited case of the chatbot, an artifcial intelligence system that is considered of low risk, according to the AI act. There could be a malfunction of the artifcial intelligence system that causes a data breach. It happens very often since artifcial intelligence systems feed on data. Precisely for this reason, the The text is available at the link: 20 12979-Civil-liability-adapting-liability-rules-to-the-digital-age-and-artifcial-intelligence_en. The text i s ava i l abl e at the l ink: ht tps: / /www.europar l / 21 TA-9-2022-0140_EN.html. The text of the proposal is available at the link: 22 business-eu/contract-rules/digital-contracts/liability-rules-artifcial-intelligence_en. It has to be considered, among other things, in close connection with the issue of liability, that some 23 scholars attribute legal personality to artifcial intelligence systems. On the issue of recognition of legal personality, Pagallo, U. (2013) The Laws of Robots, Springer; Sartor, G. (2009) ‘Cognitive automata and the law: electronic contracting and the intentionally of software agents’, Artifcial Intelligence Law, 17(4), pp. 253-290; Teubner, G. (2016) ‘Rights of Non-Humans? Electronic Agents and Animals as New Actors in Politics and Law’, Journal of Law and Society, 33/2016, pp. 497-521. In the opposite direction, however, Bertolini, A., Aiello, G. (2018) ‘Robot companions: A legal and ethical analysis, Information Society, 33(4), pp. 130-140; Bertolini, A. (2013) ‘Robots as Products: The Case for a Realistic Analysis of Robotic Applications and Liability Rules’, Law, Innovation and Technology, 1/2013, pp. 214-247. The relationship between civil liability and artifcial intelligence is examined in depth within the 24 monographic section of the journal Giurisprudenza italiana devoted to the topic ‘Intelligenza artifciale e responsabilità’, edited by Rufolo, U., Gabrielli, E. For an analysis of the issues raised, see also volume curato da De Franceschi, A., Schulze, R. (2019) Digital Revolution. New challenges for Law, C.H.BeckNomos; Weber, R.H., Staiger, D.N. (2017) New Liability Patterns in the Digital Era, in Synodinou, T.E., Jougleux, P., Markou, C., Prastitou, T., (ed(s)) EU Internet Law, Springer, pp. 197-214. 9

European legislator and the European Data Protection Supervisor emphasize the 25 26 connection and the need for harmonization between acts on artifcial intelligence and the General Data Protection Regulation (hereafter: GDPR). In case of a data leak, there is (obviously) a breach directly involving the processing of personal data. Therefore, as also the same acts on artifcial intelligence would seem to suggest, the GDPR will be applied, with all that goes with it. It would seem natural, therefore, to apply the liability regime provided by the above-mentioned legislation, that establishes for a very strict liability system; it is referred to as strict liability in charge of the personal data controller or data processor. Hence, the crucial importance of coordination with other regulations. Let’s take now the case mentioned before of facial recognition at the airport, at security checkpoints. In such a case, as provided by the AIA, it is a biometric system, which is categorized as high risk, with the resulting liability regime. It may happen that the person undergoing the security check is mistaken for a terrorist and stuck at the airport. These are the well-known cases of mismatch. In such cases, the European Parliament Resolution of 2020 provided for a strict liability regime for high-risk AI systems. 27 Today there is a change of course. On 28 September 2022, the European Commission published two proposals aimed at modernizing product liability rules in the digital age. The As an example, one may recall the explenatory memorandum of the AIA, point 1.2., entitled «Consistency 25 with existing policy provisions in the policy area». The text reads: « (…) Consistency is also ensured with the EU Charter of Fundamental Rights and the existing secondary Union legislation on data protection, consumer protection, non-discrimination, and gender equality. The proposal is without prejudice and complements the General Data Protection Regulation (Regulation (EU) 2016/679) (…) ». Moreover, the picture would not be complete if other legislative proposals were neglected. These are initiatives concerning the subject of personal data, aimed at realizing the project of a single European data market. Some legislative projects have already been approved (the reference is to the Digital Service Act and the Digital Market Act) while others are still in a gestation phase. The reference is to the Data Governance Act and the Data Act. As regards the Digital Service Act the text is available at the link: doceo/document/TA-9-2022-0269_EN.pdf. As regards the Digital Market Act the text is available at the link: The profound interrelationship between AI and data protection was underscored by the joint opinion n. 5, 26 2021 of the European data protection supervisor (hereafter: Edps) and the European data protection board (hereafter: Edpb). Moreover, on the Italian side, the following should be noted: the European Data Protection Supervisor in the Memorandum presented last March 9 to the IX and X Joint Committees of the Chamber of Deputies on the Proposal: both transversal and united by representing the challenge, current and future, thrown by technology to the law and its possibility of regulating even what appears, in its incessant evolution, to be more refractory to the norm. European Parliament resolution of 20 October 2020 with recommendations to the Commission on a civil 27 l iabi l i ty regime for art ifcial intel l igence (2020/2014( INL) ) , avai lable at the l ink: https:// 10

proposal is the first in the world specifically designed to address compensation for damage caused by artificial intelligence (AI) systems. Two Directives are proposed: one on adapting non-contractual civil liability rules to AI (the AI liability Directive); and the other for defective products (the Product Liability Directive). The proposed Directives are said to complement each other and the EU AI Act. The Product Liability Directive addresses a producer’s “no fault” or “strict” liability for defective products (including AI) and associated compensation for damages. Whereas the AI Liability Directive is concerned with liability arising from “wrongful behavior” by AI systems . 28 The line between the two responsibilities turns out to be very thin and unclear. However, if anything certain can be said, it is that there is no longer a clear diferentiation of the liability regime applicable to high-risk and low-risk AI systems. This diference, on the other hand, was apparent in the previous recommendation. However, the Proposal for an AI Civil Liability Directive of September 28, 2022 provides 29 for fault-based liability for both high-risk and low-risk systems, albeit with some diferentiations (which apparently sounds strange enough). In conclusion, in the light of the examples given before, there is something wrong. There is no harmony between the Proposals on artifcial intelligence and the GDPR. In fact, in the case of chatbot malfunction (low-risk system) the GDPR will be applied and therefore strict liability. Whereas, in the case of facial recognition (high-risk system), applying the rules introduced on September 28, we will have fault-based liability. The other question is: who can be sued? There has been a lot of debate as to who should be accountable in the event of a failure by an AI system. The new Proposals conclude it should be the providers of AI systems and, in some cases, the user of AI systems (each as defined in the EU AI Act). Therefore, even on that profile there is still no certainty. In favor of a "strict liability" regime for AI systems see Vladeck, D.C. (2014) ‘Machines without Principals: 28 Liability Rules and Artifcial Intelligence’, Wash. L. Review, 117/2014, pp. 117-150. Proposal for a Directive of the European Parliament and of the Council on adapting non-contractual civil 29 liability rules to artifcial intelligence (AI Liability Directive). The text is available at the link: https:// 11

5) Conclusions It has become clear in the preceding pages that the world of tourism is making and will make extensive use of artifcial intelligence systems in the future, changing consumer habits and making the services ofered more agile. The European legislator had the merit of being the frst to issue a legislative proposal on the topic of artifcial intelligence. At the state of the art, we can consider three pillars: the AI act, the Proposal on adapting non-contractual civil liability rules to AI (the AI liability Directive) and the Proposal for defective products (the Product Liability Directive). However, the process still appears to be at an embryonic stage, and most of the legal questions remain unanswered, the few answers still unclear. There also seems to be a lack of coordination, at least on some profles, between the artifcial intelligence proposals and the GDPR. All that remains is to await the next movements of the legislature, hopefully also in feld of tourism. 12

Contact Estoril Higher Institute for Tourism and Hotel Studies IJTTHL Celebrating five years of the first publication Vincenzo Franceschelli, On Tourism Law | Quest for general principles, Alejandro Corral Sastre, A new Administrative Law for a new Tourism: now or never, Jonas Thyberg, The Package Travel Act and the Covid19 pandemic, Caterina del Federico, The use of artificial intelligence in the travel and hospitality industry. Civil liability profiles, "We are celebrating the 500 anniversary of the discovery of the Magellan strait with two online publications. On October 21, the launch of Derecho del Turismo en Las Américas brought together colleagues from all Latin American countries. On the same day, the Collective Commentary about the New Package Travel Directive, with colleagues from the then 28 Member States. More recently, Competition Law in Tourism and Tourism Law in Europe are publications demonstrating this group's dynamism. The pandemic has not broken the bonds that have united and unite us. Here we are again here in Lisbon, more numerous and more motivated than ever, to consecrate what will be the voice of our group: The new international journal of Tourism Law, The International Journal of Tourism, Travel and Hospitality Law.” • Vincenzo Franceschelli, The Lisbon Group and International Tourism Law Five years after the first international publication (The New Package Travel Directive, October 2017) at the ESHTE | INATEL International Conference in October 2022, bringing together 50 speakers from 27 countries, a new project was announced: The International Journal of Tourism, Travel and Hospitality Law (IJTTHL). IJTTHL brings together several universities and is published both in print and online. The central part will be in English but will have an IberoAmerican chapter, and the texts are immediately available through Pre-Print. Monika Jurkova, Liability of online platforms in the tourism sector, https:// Francesco Torchia, Tourism enterprise and cultural heritage protection, as a legal instrument for valorization of the territory and of the person, https://,-as-a-legal-for-valorization-of-the-Territory-and-ofthe-Person-by-Francesco-Torchia/ Valérie Augros, Holiday lettings in France: tips and tricks, Holidays-lettings-in-France---Tips-and-tricks-by-V.-Augros/ Sarah Prager, Competition law: online travel agents and airlines, https://,--Competition-Law---OTAs-and-airlines/ Andrej Micovic, Legal Tech and Online Dispute Resolution, Microsoft-Word---A.-Micovic---Legal-Tech-and-ODR-in-Tourism./ Pilar Juana Garcia Saura, The inspection of tourist accommodation by Public Administrations: problems with the use of the robot inspector (web spider) and the responsibility of collaborative platforms, Michael Wukoschitz, A Wicked Deed’s Curse – Will X v Kuoni change the Organiser’s Liability?,,---M.-Wukoschitz/ Bertold Bär-Bouyssiere, Sustainability and Article 101(1) TFEU, Exploring (almost) virgin territory, Tatjana Josipović, Modernisation of information requirements for consumers on online tourism services market, Matija Damjan, The new online platform rules and the accommodation booking services, Ilie Dumitru, EU legislation and contractual relationship between the travel package organizer and the air carrier in case of charter flights. Liability for cancelled and delayed charter flights, João Almeida Vidal, Arbitration and tourism: a field to explore, https://