Open a browser’s task manager and you may find more processes than visible tabs. That is not, by itself, evidence of a fault. A modern page can contain material from different sites, and the browser may separate that work for security as well as reliability.
Chromium’s site-isolation documentation explains an additional boundary intended to limit access across sites even when a renderer is compromised. This builds on other protections; it does not replace the browser sandbox or the rules that normally restrict cross-site access.
A tab is a visual container
Consider a page containing an embedded frame from another site. To the reader it appears inside one tab. Internally, the browser may handle it separately from the surrounding page. Consequently, counting tabs is not a reliable way to predict the exact process count.
The word “site” also has a technical meaning in the browser’s design. It should not be casually equated with every distinct URL or with a business name. Implementation details and platform constraints affect what is isolated and how.
If you are investigating memory use, first identify the process and the activity associated with it. Close an unused page and observe the result. Avoid disabling a security feature merely because a process list is longer than expected.
Keep its limits in view
Site isolation is not a promise that a website is trustworthy. It does not make a convincing phishing page safe, prevent you from deliberately entering a password into the wrong site or approve an extension’s permissions.
For everyday use, keep the browser updated and treat unexpected sign-in requests on their own merits. For technical troubleshooting, consult the documentation for the specific browser and platform before changing experimental flags. Understanding that visible tabs and security boundaries are different things can explain an unfamiliar task-manager entry without turning it into either a panic or a reason to weaken protection.