The problem: every prompt is a potential leak
As soon as a user copies a document into ChatGPT, Claude or Gemini, the data leaves your perimeter: surname, first name, email address, IP address, case file number, health data and more. That information travels and can be retained by the vendor. For many organisations this is unacceptable, but banning these tools only pushes people towards shadow IT.
The approach we are testing: mask, send, restore
Detection of sensitive data
An engine automatically spots personal data in the text: names, emails, IP addresses, numbers, addresses and more.
Replacement with tokens
Each item of data is replaced by a neutral token. The meaning of the sentence is preserved, but no personal information leaves any more.
Sending to the third-party API
Only the anonymised text is transmitted to ChatGPT, Claude or Gemini. The mapping table stays with you.
Restoration on receipt
When the answer arrives, the tokens are replaced by the real values. The user receives a complete result; the vendor has never seen the real data.
The round trip, in one diagram
Only the masked text crosses the boundary. The mapping table, the only key for reading the real values, never leaves your infrastructure.
A concrete example
Original text (stays with you)
“Please get back to Jean Dupont at j.dupont@ville.fr about case file no. 74123.”
Sent to the API (anonymised)
“Please get back to [PERSON_1] at [EMAIL_1] about case file [REF_1].”
Restored answer (user side)
“A follow-up email has been drafted for Jean Dupont (j.dupont@ville.fr), case file no. 74123.”
Defence in depth, not a miracle cure
Anonymisation sharply reduces the exposure of personal data, but it does not replace a sovereign AI: detection is never 100 % perfect, and the channel is still that of a third-party vendor. It is an excellent complement when the use of external APIs is unavoidable.
For full control, the best answer remains sovereign AI hosted in house, where the question of leakage no longer arises.
Take part in the testing phase
This solution will be available soon (testing phase under way). Contact us to be told when it opens, or to take part in the trials with your own use cases.
Keep me informed