How Realtime Privacy works
What the coloured highlights mean and what you can do about them.
Hi team, Ingrid Lindqvist will be joining the project. You can reach [email] directly; her account is NL91 ABNA 0417 1643 00.
Realtime Privacy is the part of BeeSensible you see every day. While you type or paste in an AI tool, email, or chat, sensitive data gets an underline before you send. What happens next is your call.
What you see
An underline appears under sensitive data as you write:
- Yellow, a single line: sensitive. Everyday data like a name or an email address.
- Red, a double line: critical. Higher-risk data like an IBAN or a national ID.
Once you handle a detection, the line turns teal and dotted, so you can see what is already done. BeeSensible never blocks sending and never changes your text on its own. See Highlights while you type for the colours in detail.
What you can do
Move your mouse over a highlight and pick one of three actions. In the product they are called Replace, Mask, and Delete:
- Replace swaps in a fictional but realistic-looking value.
- Mask swaps in a numbered label like
[NAME1]. - Delete takes the value out of your text.
The Apply to all occurrences switch handles every occurrence of the same value in one go. See Review detections and replace, mask, or delete for the full walkthrough.
Where it works
The extension runs in Chrome and Edge, across AI tools, email, chat, and other supported sites. If your organisation uses the desktop app, the same underlines also appear in four apps outside the browser: Outlook desktop, the ChatGPT app, the Claude app, and Microsoft Copilot.
Your text stays private
Through the desktop app, detection runs on your own machine; your text never leaves it. Without the desktop app, the text is checked in working memory on BeeSensible's own EU servers and discarded straight after. Nothing you type is stored.