Spell-check for privacy
Realtime Privacy

See sensitive data before you send it.

BeeSensible marks names, IBANs, BSNs and 65 data types while you type, in AI tools, email, and chat. You decide what to do with each one. It never changes your text on its own and never blocks sending.

Claude
Turn these notes into a client-safe summary without exposing direct identifiers.
Here is a cleaner summary draft. Review the marked details before sharing it.
Draft a short summary using the following context: Bjorn Hansen, NL12RABO0123456789, March 3, 2026, +31 6 12345678 and bjorn.hansen@northgate.eu.
Review sensitive details before sharing outside the browser.
Insights

See where sensitive data shows up.

Aggregated, never individual. Which platforms create the most exposure, how much your team handles before sending, and when it happens.

app.beesensible.eu/analytics

Insights

Monitor sensitive-data detections across your team

Refresh
7 days30 days90 days
28/05/2026 - 27/06/2026
Detections
3240
critical · 0.0%
Most active platform
Outlook
108 detections
Handled rate
76%
245 of 324 handled
Sent despite warning
9%
29 sent unremediated
Detections over time
030609012030 May6 Jun13 Jun20 Jun27 Jun
StandardCritical
Handled rate
76%handled
Replaced100(31%)
Masked87(27%)
Removed58(18%)
Sent despite warning29(9%)
Ignored50(15%)
Detections by platform
1OutlookOutlook72% handled108
2ChatGPTChatGPT81% handled94
3ClaudeClaude74% handled41
4PerplexityPerplexity88% handled31
5GrokGrok69% handled28
6GeminiGemini77% handled20
Top entity types
1FIRST_NAME56
2LAST_NAME48
3ORGANIZATION24
4URL23
5EMAIL21
6CITY20
7Drug name15
8First name13
Detection heatmap (day × hour)
Shown in your local time
0h6h12h18h
Mon
Tue
Wed
Thu
Fri
Sat
Sun
Less More
The decision is yours

Replace, mask, or remove. One click.

Each marking becomes a choice in the panel. Replace a name with a realistic stand-in so the prompt still reads naturally, mask it behind a placeholder, or remove it entirely. Apply it once, or to every match in the field.

Gemini
Drafting an AI workspace summary
Draft a short summary using the following context: Bjorn Hansen, NL12RABO0123456789
, March 3, 2026, +31 6 12345678 and bjorn.hansen@northgate.eu.
Detection engine

Bombus, our best engine yet.

Since July 2026, Bombus 1.2 is the standard engine behind BeeSensible. A smart combination of fast AI techniques and hard rules, working as one judgement. It builds on Osmia 3.3, our earlier detection engine, and reads sharper: higher recall and stronger on structured data like IBANs and medical codes. It runs on your own device or in the EU, live while you type and across full documents.

65 data types8 categoriesEN + NLLocal or EU
See everything it detects →

How it works

Context · AI techniques Patterns · rules
One result
Where it runs

On your own device. Or in the EU.

With the desktop app, the detection runs on the machine itself, so the text never leaves it. Prefer no install? Detection runs in EU working memory and is discarded right away.

Dear Jane Foster,

Someone types

in ChatGPT, email or a document

text stays local

Processed on device

the engine runs on the machine, nothing leaves

only anonymous numbers
Optional

Check your policy

aggregated numbers for analysis and compliance

Across the apps you already use

AI tools like ChatGPT and Claude, email in Gmail and Outlook, chat in Slack, and more. 20+ apps in total.

Three actions, always yours

Replace with a realistic stand-in, mask with a placeholder, or remove it. One click, or apply to every match in the field.

Nothing is stored

Detection runs on the user's own machine or in the EU, in working memory. The text itself is never stored either way.

Try it on your own text.

Install the extension, open ChatGPT or Gmail, and see what gets marked before anything is sent.