OpenAI released an update notice, comprehensively updating the ChatGPT extension for Google Chrome browser. Users can now directly use ChatGPT in the Chrome sidebar without switching to a standalone application or web version, making the AI assistant truly a "browser native."
According to the official introduction, the updated extension allows ChatGPT to research the market, compare sources, extract information from websites, or open and refine files in Google Workspace and Microsoft 365 within the app. It can use the browser to introduce the latest context, perform step-by-step operations between web pages, and continue working while the user reviews and guides the results. This means ChatGPT is no longer just a passive chat box waiting for questions but an intelligent agent that can actively browse web pages, read content, and coordinate tasks across tabs.
It can highlight content, control tabs, and access local files
James Sun from OpenAI revealed more technical details. The extension can view the complete context in the browser, including content that users have highlighted, control the user's tabs, and access the local file system and installed plugins. Every conversation thread started in the browser will be displayed in the desktop application, and it can also reference any task initiated in the desktop application, achieving seamless integration between the browser and desktop version.
These capabilities are built upon OpenAI's previous experience with the Atlas browser. Atlas was an AI browser that OpenAI had tried to develop, but now the company has chosen to directly embed all the functions of ChatGPT and Codex into Chrome, the browser with the largest global market share, rather than continue maintaining an independent browser product. The official stated that they will gradually discontinue the standalone Atlas browser and provide users with information on how to migrate to the ChatGPT extension.
From a "standalone browser" to a "parasitic strategy," OpenAI chose a pragmatic approach
