All detection in BeeSensible runs on Bombus. One engine, for every organisation, on the device and in the cloud.

#Bombus

Bombus 2.1 combines fixed rules for structured data (bank numbers, national IDs, card numbers) with small, fast AI models for everything that depends on context (names, addresses, medical details). Those layers are merged into one result; overlapping hits become a single marking.

It recognises 65 data types in eleven categories, in Dutch and English. Which of them get marked in your organisation is up to you, through detection profiles; see Detection profiles, categories, and levels.

#On the device or in the cloud

Every organisation has a detection mode: on the device (local) or in the EU cloud. BeeSensible manages that setting per organisation; there is no switch for it in your own dashboard.

In local mode the engine runs inside the desktop app, on the user's machine. The text never leaves the device, and the desktop app does the detection work for the browser extension as well. This mode never quietly falls back: if the local engine is unavailable, the extension shows "Detection offline".

In cloud mode the client sends the text to BeeSensible's own EU servers, where it is analysed in working memory and discarded straight after; nothing is stored as content.

For users both modes feel the same. Which one is active is shown in the extension settings: Detection: on-device or Detection: cloud. On the dashboard's rollout page you see the same mode as Local or Cloud under the On-device operation section.