Artificial intelligence and sustainable tourism: major challenges and opportunities
03/11/2024
iA+Tourism tour of the Eastern Townships: Our team’s highlights
29/11/2024
White paper “Generative AI and Tourism: Innovation without gimmicks”.
The white paper ” Generative AI and Tourism: Innovation without gimmicks ” published by Hauts-de-France Innovation Tourisme (HIT) highlights the growing impact of AI in the tourism sector. The technology is seeing rapid adoption: in November 2022, ChatGPT reached one million users in five days, and today the platform boasts 180 million. The AI market could grow by an average of 37% a year to exceed €1,600 billion by 2030. The white paper cites numerous applications by tourism companies, including the following:
Club Med: uses Google’s Vision AI to manage over 50,000 automatically qualified photos. The group has also set up an intelligent bot enabling its employees to obtain real-time information on offers and facilities.
Callbot in Hauts-de-France: A theme park works with The Sane Intention to automate call handling, freeing up time for complex requests requiring human intervention.
Jardin d’Acclimatation in Paris: uses AI fromApollo Plus to predict daily attendance and optimize resource management and marketing.
Center historique minier de Lewarde: Offers immersive interaction with avatars of former miners, enriching the visitor experience.
VisitDenmark: Conducted a communication campaign based on generative AI to promote the destination in 2023.
SNCF Connect: AI for creativity and customer relationship support
For the past 18 months, SNCF Connect & Tech, the entity responsible for the SNCF (Société nationale des chemins de fer français) ticket sales website and mobile application, has been taking an exploratory approach to artificial intelligence. Its strategy is based on four pillars: quality, safety, ethics and customer satisfaction. Here are a few concrete examples of how AI is being used in the organization:
Optimizing marketing and creative processes:
- AI-assisted advertising: During the opening of the 2023 winter sales, Adobe Firefly AI was used to transform a snow globe by replacing the original chalet with Notre-Dame-de-la-Garde, thus highlighting several travel destinations. A task that would normally have taken several days was accomplished in just a few minutes.
- Simplified production for TV campaigns : AI optimized the “Retrouvez-vous” campaign by replacing green screen shooting with automated generation techniques. It also made it possible to rejuvenate the actors, reducing the teams’ workload and production times.(See the campaign video)
Customer relations support with Claude from Anthropic:
SNCF’s 150 telephone advisors handle 1.3 million requests a year via various channels, including WhatsApp, Messenger and the messaging system integrated into the SNCF Connect application. For the past six months, Anthropic’s Claude language model, deployed via AWS, has been assisting advisors by analyzing questions and proposing answers that they can edit, validate or refuse. Each interaction enriches the Large Language Model (LLM), improving the quality of future responses.
- Measurable results: AI reduced average request processing time by 30 seconds and improved customer satisfaction by 150%.
- Training and continuous improvement: All advisors have been trained to use the tool, with a focus on continuous improvement. The team constantly monitors the quality of responses and strives to reduce the generation time, currently 14 seconds, to further optimize service efficiency.
Priceline gives its chatbot a voice to reinvent travel planning
Priceline equips its Penny travel assistant with voice capabilities thanks to OpenAI technologies, via GPT-4o and the new Realtime API. This evolution enables real-time voice interactions, making travel planning and booking more natural, intuitive and personalized. Priceline is among the first brands to adopt this innovative technology.
Features and use cases :
- Intuitive voice search: users can interact with Penny by voice to ask questions and get immediate answers about hotels, restaurants or activities, without having to handle their device.
- Anticipating needs: By analyzing past interactions, Penny anticipates user preferences and adapts its recommendations accordingly.
Future developments and improvements:
- Expanded search categories: Currently limited to hotels, Penny Voice will soon support reservations for flights, car rentals and vacation packages.
- Advanced multilingual voice recognition: In the coming months, Penny will be able to understand over 120 languages and recognize various accents, offering even smoother, more personalized voice interactions.
- Answers tailored to the customer’s tone: Penny will adjust its responses according to the emotion perceived in the customer’s voice, making conversations more natural and engaging.
Klook: An experience provider powered by generative AI
Klook, an Asian platform specializing in offering travel experiences, is strengthening its partnership with Google Cloud by deploying a generative AI infrastructure across its entire operations. The aim is to improve user experience, internal productivity and partner company operations through AI technologies. Here are the main use cases and results achieved:
- Content creation and localization : Generative AI has enabled Klook to reduce content production time by over 80%, while improving the quality of translations and the relevance of local information. This feature optimizes travel pages, delivering a more consistent and engaging experience for users.
- AI Buying Guide : A new feature helps travelers compare options with a summary of user reviews and platform data. Klook thus becomes one of the first online travel agents in Asia to integrate such an AI solution.
- Making customer reviews actionable: Klook uses user reviews to help its partners identify growth opportunities. For example, thanks to review analysis, one partner launched a new attraction that now accounts for 24% of total revenues.
- Improving internal processes: Klook has integrated Gemini Code Assist to automate and optimize the code review process, essential to software development. Code reviews examine source code to detect errors, guarantee quality and ensure compliance with standards. Thanks to Gemini Code Assist, the efficiency of code reviews has increased by 60%, accelerating development and facilitating the delivery of quality code. In addition, platform performance improved by 70%, delivering a smoother user experience. AI also optimized financial and marketing tasks, including invoice reconciliation and SEO optimization, boosting the efficiency of internal teams.
Auteur
Groupe de travail IA et Tourisme