The impact on travel from Google I/O 2025 launches

If you listened to the keynote speeches at Google’s I/O developer conference this week, you will have noticed that use cases in travel and tourism were peppered throughout.

What is abundantly clear from the presentations by Google CEO Sundar Pichai and colleagues such as DeepMind’s Demis Hassabis is that the world is rapidly moving away from traditional search engine results toward an artificial intelligence (AI)-enabled future.

The use of Gemini has exploded over the past year, Pichai revealed: The Gemini app now has over 400 million monthly active users, and 7 million developers are building with Gemini—five times more than this time last year. The Gemini ecosystem is now processing 9.7 trillion tokens a month—50 times more than in 2024.

The travel sector needs to move fast to recognize the transformational changes that are happening as a result of some of the I/O announcements.

In a LinkedIn post after the keynote, Dan Granath, CEO of GoTo Hub, said, “At Google I/O 2025, we saw the shift confirmed: AI now answers the questions people used to type into search engines. Instead of ten blue links, we get one synthesized response. And increasingly, that response decides what content gets seen, what products are promoted and what gets booked.”

In another LinkedIn post, Krzysztof Balon, CEO of Automate.travel, said, “The travel sector is about to undergo its biggest transformation since online booking emerged.”

The announcements at Google I/O that will unleash dramatic change in the sector came like a torrent, and these are summarized below.

AI Mode

Pichai announced a total reimagining of search—AI Mode—which rolled out in the United States this week after time in the research and testing environment with developers.

“AI Overviews have scaled to over 1.5 billion users and are now in 200 countries and territories. As people use AI Overviews, we see they’re happier with their results, and they search more often,” Pichai said.

“In our biggest markets like the U.S. and India, AI Overviews are driving over 10% growth in the types of queries that show them, and this growth increases over time.”

AI Mode will allow users to ask both longer and more complex queries. Users in beta “have been asking much longer queries—two to three times longer,” Pichai said.

In search, a tab for AI Mode will sit alongside the usual tabs such as All, Images and Video. This will bring up a chat window to launch those longer searches Pichai referenced.

This spells the end for the top ten links and their replacement with a single source of what Google believes is “the truth.”

Personalization and Deep Search

I/O revealed that Google is adding personal context to AI Mode searches this summer—a feature it is calling Deep Search.

Users will be able to connect to other apps in the Google ecosystem, including Gmail and Google Docs, to improve this personalization.

Deep Search will thus help Google level up with other AI providers such as OpenAI, as well as with online travel agents such as Kayak, which have already embraced personalization by leveraging cookies and search history from ad networks.

Agentic AI

Google’s Project Mariner, the codename for its agent mode, is designed to take actions on your behalf with capabilities to “interact with the web and get stuff done.” It is available to developers now through the Gemini application programming interface, and Google announced it will now be available this summer through Gemini, Chrome and Search.

Google also announced an agentic checkout feature focused on online fashion shopping. Say you have been looking for a dress but decide not to buy it; agentic checkout will watch the price, and if there is a price drop, it can prompt you to buy with just one tap. The potential for travel is clear.

From inspiration to booking

In AI Mode, Deep Search will remember the context of your search journey. You could start by asking for a perfect two-day trip to Austin, Texas, for example, and then narrow this down, asking for recommendations for a bus tour or restaurant in town.

Results from Google Travel and Google Things to Do are integrated into the results. Clicking on them doesn’t take you to those sites but keeps you within AI Mode. At various points, organic links appear in the right-hand column—currently these are organic links, but it is not hard to imagine these will become paid links in the future.

When agent mode is incorporated into this over the summer, it will allow you to book from that same window. This will get around the current limitations of large language models, which would not have access to availability and prices. The booking could happen completely within AI Mode. This will be a massive opportunity for suppliers to drive direct bookings.

One consequence of Deep Search is that users may decide that they no longer need to seek out expert articles or blogs to find credible answers. This will give suppliers and intermediaries an opportunity to establish authority for themselves rather than seeking third-party endorsements.

Visibility in this new world will require suppliers to focus on providing structured and authoritative content that AI systems can consume.

Other announcements

There was more for the travel sector in the I/O keynote speeches.

Real-time speech translation on Google Meet is now available for subscribers in English and Spanish. The example presented showed a conversation between a vacation rental host and guest. More languages will be rolled out in next few weeks and an enterprise version later this year.

Google Glass is dead; long live Glasses with Android XR. Google is partnering with optical companies Gentle Monster and Warby Parker to bring AI-enabled glasses to market later this year, incorporating camera, microphone and speakers. In a live demonstration, Shahram Izadi, Google’s vice president for Android XR, and Nishtha Bhatia, product manager of Glasses & AI at Google, conversed in Farsi and Hindi.

• Google Beam is a video communications platform, using an array of cameras to capture video from six angles which is then merged into 3D in real-time using AI. It will be launched through a partnership with HP later this year.
• Gemini Live will add camera and screen sharing to the Gemini app and can be connected with other Google apps such as Calendar and Maps. Use this to take a picture of a church in a destination, and you no longer need to book a guide. Could this be a killer app combined with Glasses with Android XR?
• Travel companies will also welcome better AI generation of content: Veo 3 for AI-generated videos including dialog and sound effects and Imagen 4, an improved AI image generation model.
• Hassabis said the ultimate vision for the Gemini app was “to transform into a universal AI assistant” and that this would be powered by a new mode for the Gemini 2.5 Pro model called Deep Think.

In summary

Pichai ended his keynote with statistics. He revealed he had mentioned AI 92 times and Gemini 95 times during the presentation.

“The opportunity with AI is truly as big as it gets,” he said in closing.

The travel sector needs to grasp that quickly.