ChatGPT’s Chrome extension can now work across your open tabs

OpenAI has expanded ChatGPT’s Chrome extension so it can use information across open tabs and, in Work, carry out browser tasks such as filling in forms and downloading files. The extension listing reports seven million users, putting computer use within reach of a sizeable existing audience rather than a handful of curious testers.

OpenAI Watch analysis

What happened

The update lets ChatGPT use open tabs as context, organise them and take actions in the browser. OpenAI’s examples include comparing products across tabs, preparing a spreadsheet from vendor pages and filling in a form while stopping before submission. The extension listing also describes creating and editing documents and working with sites where a user is already signed in.

The update was announced on 15 October, according to Unite.AI. Its account of OpenAI’s documentation says the extension supports Chrome, Edge, Brave, Opera and Vivaldi through the ChatGPT desktop app, with some differences in side-chat availability.

Why it matters

This moves ChatGPT from answering questions about a page towards doing work across several pages and services. That could save tedious copying and switching, especially for research or routine form-filling. It also means the assistant may encounter private or authenticated information, not just whatever someone types into a prompt.

OpenAI’s documented controls include asking before interacting with each new website, with options to allow an action once or more broadly. The documentation also labels browser-history access and allowing all sites as elevated risk. Permission prompts are useful; they are not a substitute for understanding what the extension can access.

Our read

The useful change is not that ChatGPT has acquired another place to sit on screen. It is that the browser can become part of the task. Before handing it anything sensitive, check the site permissions and review actions before they are submitted. The assistant can do the clicking; responsibility for the click remains stubbornly human.

What to watch

• Whether OpenAI’s examples translate into reliable results on real, signed-in websites.
• How clearly the extension signals what information it reads and what actions it takes.
• Whether users find its permission controls easy to understand and manage.

Discussion spark: Would you let ChatGPT act across sites where you are signed in if it asked before each new website, or should browser agents be limited to read-only tasks?

OpenAI Watch is independently operated by WittyWires. It is not affiliated with, endorsed by, or operated by OpenAI.