ChatGPT: its prototype was launched less than six months ago and it’s already taken the world by storm. Reuters reported that it had 100 million active monthly users in January, making it the fastest-growing consumer application in history. When you compare to another successful platform, Instagram, ChatGPT accumulated 10 million users in 40 days, while Instagram took 355 days to get there.
It’s no wonder that I can’t seem to be able to walk into a room these days without somebody asking me, “What impact will Generative AI have on the travel industry? And how are Amadeus and Microsoft planning to use it?” — the same questions my team and I have been investigating for some time.
I’m Fredrik Odeen, Innovation Program Director for Corporate Strategy at Amadeus. I recently had a conversation with Julie Shainock, Global Leader for the Travel and Transportation Industry at Microsoft to get to the bottom of the biggest questions on everybody’s mind.
FREDRIK: Is Generative AI and more specifically apps like ChatGPT actually that new? Chatbots have been around for a while. How is Chat GPT different, and where did it all begin?
JULIE: The GPT in ChatGPT stands for “generative pretrained transformer”. This means that when you interact with ChatGPT, it’s actually generating original answers based on your prompts.
But you’re right, chatbots and generative AI have been around for a while. The first early natural language program, dubbed ELIZA, was created in 1966 by MIT computer scientist Joseph Weizenbaum. It was the first program that allowed some kind of plausible conversation between humans and machines.
Other chatbot experiments followed, but all machine learning is based on data sets—the AI must have access to large swaths of text or images to study so that it can identify, predict and anticipate patterns in data. But it’s only recently that we have access to the transformers we need to give machine learning models access to the terabytes of data they need to be truly convincing and original.
While early machine learning models were trained by humans—a tedious and time-consuming process—contemporary text-based machine learning models can rely on self-supervised learning, which significantly increases the amount of data an AI model can study, thus greatly improving its capacity to anticipate how a sentence should start and end.
ChatGPT is unique because of the sheer volume of data it was trained with – it exceeds previous chatbots by orders of magnitude: ChatGPT 3 was trained with 17 billion trainable parameters, and 45 terabytes of data. For reference, the Hubble Space Telescope generates 10 terabytes of data per year. In other words, ChatGPT has read more text than anyone will read in a lifetime (ChatGPT tells me this is 90 million books!) This volume of knowledge is the reason it has such an exceptional command of language.
JULIE: Can you look at this through a travel lens, Fred? What does this mean for the industry?
FREDRIK: Generative AI is still so new and evolving every day, so it’s hard to anticipate exactly how it may be used, but experts in the travel industry are already talking about how it could impact travelers at each stage of the traveler journey.
One thing is for sure; it will impact all stages of travel:
Pre trip: It could be used in trip planning. For example, imagine if a traveler could say, “Hey, I want to do nature-based activities in southeast Asia. I have two weeks, and this is my price range. What do you recommend?” and they would receive a complete trip itinerary in just a few seconds, including flights, hotels and transfers to book. This could revolutionize how people search and book travel. However, an important aspect here would be to see whether ChatGPT would allow for sponsored content. This could have a negative impact on user trust and put a premium on travel advice that comes with the human touch.
On trip: AI-powered instant translation services could make it easier for travelers to negotiate last-minute changes directly with accommodation hosts and rental car providers who don’t speak the same language. Machine learning models could also automatically inform a host about a late arrival to their boutique hotel due to a delayed flight. Customer service in general throughout the traveler journey could be automated much more easily.
Post trip: It could help travel companies analyze customer feedback and identify trends or patterns, providing valuable insight into customer sentiment and opportunities for personalization. At the same time, it could build a profile for the traveler, asking for ratings on specific activities, looking to build trip recommendations based on preferences, seasons and prices.
These are just some of the ways we anticipate Generative AI will infiltrate the travel space.
JULIE: What are some specific ways that Amadeus is planning to incorporate Generative AI into its technology?
FREDRIK: As you know Julie, Amadeus and Microsoft have some exciting projects in the works that I can’t reveal just yet, but here are some things we can talk about...
Better travel puts travelers first, which is why we are focused on delivering technology that creates hyper-personalized travel experiences. By leveraging AI-generated content and data analysis, we aim to help our travel seller partners to create tailored travel experiences that cater to the unique needs and preferences of individual travelers. This includes providing personalized recommendations, real-time itinerary adjustments, and seamless integration of preferred services.
Of course, any journey, no matter how it’s planned, deserves the best customer support. Generative AI will revolutionize customer support by automating responses and providing instant, accurate, and relevant information to travelers. This will not only improve customer satisfaction but also reduce the workload on customer support teams, so they can focus on more complex tasks where their human touch will make all the difference.
Finally, travel industry collaboration is really important to us. Our partnership with Microsoft will foster collaboration across the entire travel industry. By using generative AI to share information and insights, we can create a more cohesive ecosystem, enhancing the services and offerings available to customers and unlocking new opportunities for growth.
FREDRIK: What about Microsoft? How are you integrating Generative AI and Azure Open AI into your solutions?
JULIE: Well in March we announced Microsoft 365 Copilot —which combines the power of large language models (LLMs) with user data across the Microsoft 365 Apps—including calendar, emails, chats, documents, meetings, and more—to unleash creativity, unlock productivity, and uplevel skills. For example, you could give it natural language prompts like "tell my team how we updated the product startegy" and it will generate a status update based on the morning’s meetings, emails, and chat threads.
JULIE: We’ve focused on the positives and huge changes Generative AI could bring to our industries, but what would you say to people who are concerned about AI in the travel space?
FREDRIK: This is an important question. AI can automate many decisions and it’s critical to make sure that the decisions it makes are fair and not biased against certain groups of people. There are also concerns about privacy, security, malicious usage, accountability, and sustainability of AI systems. That’s why we have clear guiding principles that help us navigate how and when to use AI technology. We have an AI Ethics policy containing [a number of core] principles that guide AI development and its use at Amadeus:
Fairness: “AI systems should treat all people fairly”
Reliability and safety: “AI systems should perform reliably and safely”
Privacy and security: “AI systems should be secure and respect privacy”
Transparency: “AI systems should be explainable”
Accountability: “People should be accountable for AI systems”
Sustainability: “AI systems should be sustainable”
Of course, this is the beginning of a new era. We’re constantly following new regulations and keeping our eyes open to new ways to use this technology to maximize its benefits while mitigating any potential challenges or risks.
FREDRIK: What’s next for Microsoft and Amadeus? Is there anything you want to tell our customers and other industry partners?
JULIE: We’re working on some very exciting projects at the moment, and I’m very excited to see how this journey evolves to transform the future of travel. For now, all I can say is that Microsoft and Amadeus' AI-driven tools can easily be integrated with existing systems and platforms used by travel professionals, such as booking engines, CRM systems, and marketing automation tools. I do think that generative AI is going to have a very big impact on how we work to help travelers make the most of their trip. Watch this space!
Fredrik Odeen is the Program Director, Corporate Strategy, at Amadeus. Connect with Frederik on LinkedIn.