When you buy through affiliate links in our content, we may earn a commission at no extra cost to you. Learn 
Google’s Gemini-powered voice assistant on Android can now scan through webpages on Chrome and YouTube videos to produce short summaries.
Google bills AI as the central feature that will shape the future versions of Android as we witnessed on the latest Pixel 9 series. Accordingly, Gemini has been gradually replacing Google Assistant as it adds new features.
Now, the Gemini app on Android can summarize web pages and YouTube videos. The additions are part of the contextual abilities that Google showcased at the I/O event in May this year.
When you open Gemini over any webpage or video, you’ll get two buttons that will attach the sources as context for queries. After attaching the video or the webpage, you can ask questions related to the input and Gemini will provide a response.
In practice, the feature doesn’t always answer questions well when asked about the contents of YouTube videos. The answers are sometimess shallow and simply based on the title, video description, and chapters. Other times, it throws an error explaining its inability to understand a video.You also have to tap buttons to activate the feature, which undermines Gemini’s role as a voice-based assistant on Android.
Other times, YouTube summaries work wonderfully well with the assistant also offering responses based on specific parts of the video. It might rely on captions to getthe information.
Dealing with text on webpages is much easier for Gemini, and you can ask questions about information in the middle of the page and likely expect a reasonable response. However, the ability is currently limited to Chrome on Android. While using any other web browser, Gemini takes a screenshot instead of inspecting the entire webpage and responds based on the image.
These features are available on version 1.0.662093464 of the Gemini app on Android. I was able to activate it on multiple phones running Android 14 without having to update the app, which suggests it might be a server-side switch. In addition, Gemini’s floating card user interface is also rolling out, which ensures Gemini opens on top of the existing screen instead of moving to a new one.
This functionality on Gemini is powered by extensions. These extensions allow Gemini to interact with other apps and get contextual responses.
Gemini for web also supports other Google apps such as Gmail, Google Docs, Drive, and YouTube Music. But the functionality on Android appears limited at the moment. In addition to these apps, Gemini may soon open up extension support for popular third-party apps such as WhatsApp and Spotify. This would allow Gemini to perform actions such as placing calls, answering messaging, or skipping tracks to a specific timestamp.
It’s still unclear how Google addresses privacy concerns with these extensions, especially with encryption in WhatsApp. This could mean Google skips Gemini-backed features in the EU until the region defines specific regulations. For other regions, users won’t necessarily have a choice.
