Realtime Privacy: detection while people type

Realtime Privacy is the module that flags sensitive data at the moment someone types or pastes it, before it is sent. It is the day-to-day layer your team sees: a coloured highlight under a name, an IBAN, or a BSN in an AI chat, an email, or a web form, with the choice of what to do about it.

#What it does

While a person writes, BeeSensible checks the text and underlines sensitive data: a calm warning colour for regular data, a stronger colour for critical data like IBANs and national IDs. The person can then replace the value with a realistic placeholder, mask it, or remove it. BeeSensible never blocks sending and never changes text on its own. The decision stays with the person, which is what builds the habit over time.

It runs in two places on the same detection:

• The browser extension covers AI tools, email, chat, and other web apps in Chrome and Edge.
• The desktop app adds native apps that the browser cannot reach, such as Outlook desktop and the ChatGPT desktop app.

#What you control

As an admin you tune the module from the dashboard:

• Detection profiles decide which of the 65 data types are active, and how strongly, per app. A finance team's IBAN profile does not have to apply to a support team's chat window. See Detection profiles, categories, and levels.
• The detection engine (Bombus or Osmia) and where detection runs (on-device or in the EU cloud) are organisation-level settings. See Detection engines.

#What you see

Activity appears in the dashboard as aggregated counts and patterns: which data types are flagged most, which apps carry the most exposure, and how that shifts over time. Never the text, never an individual person. See Reading the analytics dashboard.

#Where the text goes

Through the desktop app, detection runs entirely on the device and no text leaves the machine. For browser-only use, the text is checked in working memory on BeeSensible's EU servers and discarded straight after. Nothing is stored as content.