Google plans to integrate its generative AI, Gemini, into all cars that support Android Auto in the coming months. This announcement was made during the company's Android event, prior to the 2025 I/O Developer Conference.
The company stated that adding Gemini functionality to Android Auto and, later this year, to vehicles running Google's built-in operating system, will make driving "more efficient and enjoyable." This update was shared in a post on the company’s official blog.
"We believe this will be one of the biggest transformations we've seen in the in-car experience," said Patrick Brady, Vice President of Android Automotive, during a virtual briefing with media members ahead of the conference.
Gemini will enhance the Android Auto experience in two significant ways.
First, Gemini will act as a more advanced voice assistant. Drivers (or passengers – Brady mentioned that they won’t perform voice matching with the owner of the phone running the Android Auto experience) will be able to ask Gemini to send messages, play music, and basically do everything Google Assistant already does. The difference is that, thanks to Gemini's natural language capabilities, users won't need to rely on rigid commands anymore.
In addition, Gemini can "remember" preferences, such as whether a contact prefers to receive messages in a specific language, and handle translations for the user. Google claims that Gemini will be able to carry out one of the most common in-car tech demonstrations: finding good restaurants along a planned route. Of course, Brady noted that Gemini will tap into Google’s listings and reviews to respond to more specific requests, such as "Mexican taco places with vegetarian options."
The second major way Gemini will appear is through a feature called "Gemini Live," where a digital AI listens continuously and is ready for full conversations. These discussions could range from brainstorming spring break travel ideas to creating recipes for a 10-year-old’s meal, or even diving into topics like "Roman history," according to Brady.
If all this sounds a bit distracting, Brady insists that Google believes otherwise. He argues that natural language processing will make it simpler to request Android Auto to perform tasks, reducing hassle, thus Gemini will "lower cognitive load."
This is a bold statement at a time when many are calling for car manufacturers to move away from touchscreens and bring back physical knobs and buttons – a demand some companies have already begun addressing.
There are still many unknowns. Currently, Gemini will run on Android Auto and cars with Google’s built-in systems using Google’s cloud processing. However, Brady mentioned that Google is collaborating with automakers to "build more computing power so [Gemini] can run on the edge," which would not only improve performance but also enhance reliability – a challenge in moving vehicles that connect to new cell towers every few minutes.
Modern cars generate vast amounts of data from onboard sensors, and in some models, even include interior and exterior cameras. Brady stated that Google has “nothing to announce” regarding whether Gemini can utilize this multi-modal data, though “we’ve had a lot of discussions about it.”
"We do think that as cars come equipped with more cameras, there are some really fascinating future use cases here," he added.
Gemini will roll out on Android Auto and Google Built-In in all countries where the company’s generative AI model is available, supporting over 40 languages.
