Information has been leaked about a new Google project that may be officially revealed next December. It will be an artificial intelligence capable of browsing the web and performing everyday tasks on its own, instead of the user.
Forget chatbots GeneratorsGenerators From the pictures, it's already outdated. The next major advantage in the AI race is replacing the chair and keyboard interface, i.e. the user. While Anthropic announced last week that the new Claude 3.5 Sonnet can interact with it appsapps office, GoogleGoogle It will be working on a similar feature, according to Information.
According to their sources, the artificial intelligence will be codenamed “Project Jarvis,” and “ It responds to a person's commands by repeatedly taking screenshots of what is on their screen ComputerComputer and interpret snapshots before performing actions such as clicking a button or entering text into a field “.
Improved feature for Google Chrome
However, Project Jarvis will be a little more limited than Cloud, since this functionality will be limited to the web, and more specifically to the Google Chrome browser. The goal will be above all to automatically handle everyday tasks on the web, such as searching, shopping or making reservations.
Google reportedly plans to unveil Project Jarvis in December, alongside Gemini 2.0. Other companies are also working on similar features. OpenAI is said to be working on a model that can navigate the web on its own, and Apple Intelligence will be able to understand what's displayed on the screen, and even interact with apps. Finally, if MicrosoftMicrosoft Copilot Vision will not be able to click for you, and will still be able to understand and discuss the content on screen.
