I spent my morning with Gemini in Chrome, the new integration that puts the auxiliary assistant on your browser. Instead of moving to the Chatbot web app, you can click on the new Gemini button in the right corner of Chrome to start a conversation-but the main difference is that Browsed browser assistant You can “see” what is on your screen while moving in the web.
For me, Gemini’s integration appears to be just the beginning of Google’s Important to make Amnesty International more “agent”, I also found myself wanting to do more than it was actually. Currently, you can only try an early access version from Gemini in Chrome if you are subscribed to AI Pro or Ai Ultra, use the trial version, DeV or Canary from Chrome.
I started using Gemini to summarize some articles on freedomIn addition to finding some news related to the home page, where the Game Boy Nink game indicated Two Film air conditioning, and the huge Dec Steam update from VALVE.
But Gemini only can “see” what is on your screen, so I found that if you want to summarize certain elements, such as freedomComments section, you will need to make it visible before Chatbot can provide a response. Gemini will follow you when switching the tabs as well, but it can only pull the information from one at a time.
If you do not want to write, the Gemini in Chrome also allows you to switch to the “Live” feature by selecting the button in the bottom corner of the dialog box. From there, you can simply ask a question loudly, and Gemini will respond to talking to you.
I found this especially useful to use it along with YouTube videos, where I presented a video to reshape the bathroom and asked him, “What is the tool that you use?” “It seems to use the nail gun to connect some wooden pieces together.” In another video, correctly select GEMINI on the motherboard, along with tweezers and hot air tool that YouTube used to remove. It can summarize videos and tell you about specific parts that you did not see either, but I found that this is not always accurate if it is not a video called chapters that can draw information from.
Perhaps my favorite use of integration is the presence of Gemini recipes from YouTube videos, so I did not have to write recipes myself or search for a link in the description. It was also useful when I asked him to refer to water -resistant bags on the Amazon search page.
Gemini was not always consistent. When I asked Gemini where Mrbeast was through a A video of him to explore the ancient Mayan citiesIncluding Chichén Itzá, answered, “I can’t access information in actual time, so I can’t locate the exact current Mrbeast Mrbeast.” When I asked it again, he responded with the site included in the video description: Mexico. Again, I asked GEMINI to a link to buy a specific pair of the scales offered in a video clip, but Gemini told me again that “he could not access information in actual time, including product lists or store stocks.” However, Gemini provided links to other products when claiming.
Sometimes, I felt that Gemini responses were very long for a small popup window in Chrome. You can extend it, but it does not leave much space on my 13 -inch MacBook Air screen. In addition, one of the main points of sale of artificial intelligence is that it is supposed to help you save time by providing quick and brief answers, which is not always done unless I ask for this specifically. Gemini follow -up questions, such as if you want to learn more about a specific topic, are somewhat frequent.
Even with these hiccups, I can easily see Google extending Gemini from Chrome to just just questions and answers. Google wants Amnesty International as an “agent”, which means that it can perform tasks on your behalf, and it appears that Gemini in the Chrome is ready to adopt these types of features one day. After I asked Gemini to summarize the restaurant list, for example, I thought about the request to submit a capacity – an agent that cannot be done yet. In the future, I could have seen it on hand by getting it by setting a reference signal on travel research for me, or perhaps even finding YouTube videos and saving them from different recipes to the play list at a later time.
Google seems close to making this a reality With “Agent’s Place” in the Mariner project to the application of GeminiWhich will allow it to manage up to 10 tasks at one time and search the web for you – and perhaps one day, these capabilities will bring to Gerome in Chrome as well.
