I spent my morning with Gemini in Chrome, a new integration that puts AI -driven assistant in your browser. Instead of going to the chat boot web app, you can click on the new Gemini button in the Chrome’s top right corner to start the conversation-but the main difference is that the browser’s built-in assistant can “see what is on your screen while navigation on the web.”
To me, the integration of Gemini in Chrome looks like Google’s mission began to make its AI more “agents”, because I want myself to do more than that. Now, you can only try the initial access version of Gemini in Chrome if you are AI Pro or AI Ultra subscriber, and can either use Chrome’s beta, giant, or canary version.
I started using Gemini to summarize some subjects StuffyAlso look for some gaming news on the homepage, where it pointed to the new game Boy Games Nintendo to include its switch online service, next Alden color Film adaptation, and large scale steam deck update of valve.
But Gemini can just “see” what’s on your screen, so I noticed if you want to summarize some of the elements like that, such as StuffyThe comments section, before providing a chat boot response, you will need to make it visible. Gemini will follow you when you switch to tabs, but it can only pull information from only one at a time.
If you don’t feel like typing, Gemini in Chrome also allows you to switch to the “straight” feature by selecting the button in the bottom right corner of the dialog box. From there, you can just ask a question out loud, and Gemini will talk to you and answer.
Gemini’s abstracts can be a bit tall in such a small window. Screenshot: Verge
I found it especially useful in using YouTube videos, where I prepared the bathroom -making video and asked, “Which tool is he using?” Gemini replied, “It looks like he is using a nail gun to tie some pieces of wood together.” In another video, Gemini correctly identified a capacitor on the Mother Board, as well as a brush and hot air tool, as well as Utber used to remove it. It can summarize the videos and tell you about specific parts you have not seen, but I have come to know that if a video is not a label label, it is not always correct that it can draw the information.
Perhaps there are recipes for Gemini bridges from YouTube videos in terms of my favorite use for integration, so I don’t need to write recipes or find the link in detail. It also worked when I asked to identify a water proof bag on the Amazon search page.
However, Gemini was not always permanent. When I asked Gemini that Mr. Best was a video of him, he had a video that he was looking for in ancient Mayan cities, including Chechen Atza, he replied, “I do not have access to real -time information, so I can’t identify the exact current location of Mr. Best.” When I asked him again, he replied with the location listed in the video: Mexico. Another time, I asked for a link to buy a specific twenty couple shown in a video shown from Gemini, but Gemini once again told me that it did not have access to real -time information, including product listing or store inventory. ” However, Gemini provided me with links to other products on indicating.
Sometimes, I found that Gemini’s reactions were just too long for a small popup window in Chrome. You can increase it, but it doesn’t leave much space on my 13 -inch display of my MacBook Air. In addition, one of the most important soldiers of AI is that it has helped save time by providing quick and comprehensive answers, which it doesn’t always do unless I especially ask for it. Gemini’s follow -up questions, such as I want to know more about a particular topic, happened a little repeatedly.
Even despite these hiccups, I can easily see that Google Chrome’s Gemini integration is just beyond easy questions and answers. Google wants its AI “agent”, meaning it can work on your behalf, and Gemini in Chrome is one day ready to adopt such features. After asking Gemini to summarize a restaurant menu, for example, I even thought of asking him to order a pickup – an agent that can’t do it yet. In the future, I could even see that it could be for me by keeping travel research -related bookmark pages, or later searching and saving YouTube videos of various recipes on my watch playlist.
Google looks like the project is approaching the fact that the Gemini app with the “agent format” of the Marine, which will allow it to manage 10 tasks and find the web at the same time – and one day, it will also bring these abilities to Gemini in Chrome.
