AI has demonstrated immense potential in revolutionizing various industries and their respective professions, such as programming, accounting, word processing, and art. We are only scratching the surface of what AI can accomplish, but Google is actively working on enhancing one particular capability: itinerary planning for travel. While still a concept, given the rapid progress of generative AI in generating possibilities, it is plausible that you may utilize this technology or something similar for your upcoming vacation or business trip.

The demo app created by Google serves as inspiration for app developers, aiming to showcase the potential of combining its generative AI-powered language model with existing Google APIs. Google believes that itinerary planning exemplifies a complex, research-intensive task that can be significantly simplified with the help of AI. As a result, the company has released an open-source demo to enable developers to experiment with and gain a better understanding of how AI can be deployed in this particular application.

In the demo app, Google utilizes the PaLM natural language model API to gather the traveler’s preferences in a conversational manner. Users can define their budget, areas of interest, and preferred activities. Subsequently, the AI can suggest destinations and utilize the Google Maps API to retrieve location details and images. Visual representations of a location and its attractions can provide a much better understanding and expectation of what to anticipate. Our team had a similar experience when using Bard’s recently acquired ability to display images in search results.

Google’s flowchart for polling the Maps API

However, Google’s travel planner demo offers more than just these functionalities. What if you fail to specify the duration of your trip? The default duration is set to five days. Additionally, Google has trained the PaLM API to respond in a particular manner and await your requests for itinerary refinements based on your desired activities during the trip. The app also possesses conversational memory, enabling it to comprehend the context when you revisit preferences mentioned earlier in the interaction.

Moreover, Google permits third-party integration, enabling individuals planning a trip to book flights and hotels directly through the AI-powered chat interface once they are satisfied with the tailored itinerary. Third-party developers can also leverage the AI to showcase curated tour packages and destinations based on commonly searched keywords such as “budget traveler” and “family trip,” eliminating the need for manual personalization during the search process.

Since Google’s entire AI-powered demo is open source, the company essentially allows other developers to examine its code and potentially build upon the experience provided by its example app. As it is not restricted by license agreements, it is only a matter of time before we witness AI integration in top-tier travel planning applications. Google has already begun discussions with partners in Asia, including Agoda, to integrate PaLM into the existing customer flow of the ticket booking platform.